摘要:

C. PROPERTIES OF BIOLOGICAL MEMBRANES THE STRUCTURE OF FILMS OF PROTEINS ADSORBED ON LIPIDS BY D. D. ELEY AND D. G. HEDGE University of Nottingham Received 26th January, 1956 The interaction between lipid monolayers and proteins in the substrate has been inves- tigated quantitatively, using a Langmuir surface trough, by studying the changes in surface pressure produced on injection of various quantities of protein solution under the film. This isotherm shows two marked discontinuities which are considered to indicate the com- pletion of the first and second adsorbed sub-layers of protein. Lipo-protein layers have been formed with stearic acid and cholesterol as the lipids, and plasma albumin, fibrinogen, lysozyme and insulin as the proteins. In every system studied, the first sub-layer of protein has been found to consist of denatured protein, the molecules unfolding to bring every amino-acid residue under one polar group of the lipid monolayer.Indications have bcen obtained that the second sub-layer consists of native protein. These results have a bearing upon cell membrane structure. There is little need to stress the biological importance of lipo-proteins since these materials form the basis of the plasma membranes in the living cell. A variety of techniques has been employed to study their structure and mechanisms of formation, but little correlation has been made between the results obtained and membrane structure. Putnam and Neurath 1 have used diffusion, viscosity and electrophoresis methods to investigate the interaction between serum albumin and dodecyl sodium sulphate.These, and other workers, have attributed complex formation to electrostatic forces between oppositely charged groups. An emulsion technique has been used by Elkes, Frazer, Schulman and Stewart,z who made an estimate of the amount of protein adsorbed at the surface of the drops. The extent of the surface denaturation of an enzyme at an oil/water interface has been measured by Fraser, Kaplan and Schulman 3 using oil/water emulsions stabilized by various lipids. Applying an analytical method Pankhurst 4 has studied the composition of complexes formed between gelatin and sodium alkyl sulphates. The cell membrane has been supposed by Danielli and Davson,s on the basis of permeability, surface tension and other evidence to consist of a bimolecular layer of lipid, with an adsorbed monolayer of surface-denatured protein on each side of the lipid, and on top of these a further layer of native protein.It is evident that the techniques of surface chemistry will be valuable in the study of protein- lipid interaction of the kind envisaged in this model. One method, used by Sher and Sobotka,6 has been to examine the stripping of built-up fatty acid monolayers by serum albumin. Conditions more nearly approaching the physiological are achieved by an injection technique, as used by Schulman and Rideal.7 Qualitative studies on injecting protein solution beneath a unimolecular film of a lipid were carried out by Schulman,s and later by Doty and Schulman,g and Matalon and Schulman,lo who found an appreciable increase in surface pressure at constant film area, indicating a marked adsorption, when the monolayer and the protein are oppositely charged.In the present work an attempt has been made to study quantitatively lipo-protein films formed in this way, and to draw an analogy between each film and the cell membrane. 22 1222 STRUCTURE OF FILMS OF PROTEINS EXPERIMENTAL MATERIALS.-The lipids used were cholesterol and stearic acid, both of which gave areas agreeing with published values when spread to form a monolayer on water. Crystallized bovine plasma albumin (B.P.A.) and egg white lysozyme were obtained from Armour Laboratories. Human fibrinogen was a sample supplied in sealed ampoules by the Lister Institute in 1950, and sheep insulin was supplied by Burroughs Wellcome and Co.Electrophoretic examination of the proteins showed that all were homogeneous except fibrinogen, which contained about 5 % of another component. Buffer solutions were prepared from A.R. reagents, using water which had been distilled from alkaline permanganate. The polyacrylic acid was a high molecular weight sample (M - 2.2 x 106) prepared in this laboratory by Dr. M. R. Porter. Insulin was dissolved in 0.01 N HCl since it is most stable at pH2. Other proteins and polyacrylic acid were dissolved in conductivity water. All solutions were stored at 1" C and were discarded after one month. APPARATUS AND PROCEDURE. The Langmuir surface trough used was of chromium- plated brass. Surface pressures were measured by a single horizontal torsion wire and were accurate to 0.02 dynes/cm.The authors are indebted to Professor D. 0. Jordan for the construction of the surface trough. The lipids were spread from benzene solution, and the film brought to the required initial surface pressure, which was either 2 dynes/cm, or 10 dynes/cm. A known volume of a 0.02 % or a 0.04 % solution of the protein was injected, using an Agla micrometer syringe, and the resulting rise in surface pressure at constant area followed over about one hour. The final value of the surface pressure was taken to be that after 50 min, after which time there was little further change following a rapid initial rise. Sweeping the surface on the " clean "side of the mica float showed the absence of leakage of injected protein past the barrier.The protein solution was injected through the lipid monolayer and the same results were obtained whatever the site of injection. Measurements were carried out varying the amount of protein injected, from 0.005 mg to 0.14 mg, and the increase in surface pressure at constant area, A n , plotted against amount injected. The substrate was a 0.04 M phosphate buffer of pH 7-4, with the ionic strength made up to 0.15 with sodium chloride; this ionic strength was used throughout all the experiments. Similar experiments were carried out, using B.P.A. and stearic acid, at pH 1.9. The effect of pH on the rise in surface pressure was studied using insulin and polyacrylic acid with both lipids. The pH was varied using HCl, acetate and phosphate buffers, and was measured with a glass electrode.The effect of urea on the interaction was determined by injecting B.P.A. under a film of stearic acid on a 4 M urea substrate. All experiments were carried out at a room temperature of about 17" C. RESULTS The initial pressure in the adsorption by cholesterol films was constant at 10 dynes/cm, and a considerable rise in surface pressure was observed with B.P.A., fibrinogen and insulin. Lysozyme, however, produced no such effect. Using stearic acid monolayers the adsorp- tion at 10 dynes/cm was negligible. This was probably due to the closer packing of the hydrophilic groups in stearic acid, since the carboxyl groups occupy only about one half the area of the hydroxyl groups in cholesterol at the same surface pressure. Thus, at this high pressure the carboxyl groups will interact with each other rather than with the protein.Support for this explanation is given by studying the effect of expanding the stearic acid film. On decreasing the initial surface pressure the adsorption became greater, owing to less close packing of the carboxyl groups. The most convenient initial surface pressure for stearic acid monolayers was 2 dynes/cm. Variation of the amount of protein injected beneath the lipid film had an interesting effect upon the total increase in surface pressure, A n , of the film (see fig. 1, 2 and 3). Since the amount of lipid varied slightly each time, A17 is shown as a function of the ratio of the number of molecules of protein np to the number of lipid molecules nL. In all cases the curves are of the same type as shown for B.P.A.adsorption. To interpret these curves it is assumed that all the injected protein is adsorbed on the lipid monolayer. This must be very largely true, since protein remaining in solution (perhaps in reversible equilibrium with the adsorbed multilayer) would sooner or later have given rise to a protein film on the " clean " water surface beyond the barrier, and this was not found. The graph can then be divided into two sections, each section being approximately a straight line, which is taken to represent the forming of a separate layer of protein. Thus the first layer has a large effect upon the surface pressure, and it is completed by a small amount of protein.D . D . ELEY AND D. G . HEDGE 223 The second layer requires much more protein but it produces very little change, relatively, in the surface pressure.The further addition of a third layer of protein has no effect upon the surface pressure, and no indication is given of the amount of protein needed to complete I I I I I I I I I I 0 0 8- 1 I I, 2 4 6 8 to 12 14 16 FIG. 1.-The variation of surface pressure increase, An, with molecular ratio of B.P.A. to cholesterol, np/nL, at pH 7.4. FIG. 2.-The variation of surface pressure increase, A n , with molecular ratio of B.P.A. to stearic acid, iip/iiL, at pH 7.4. it, or indeed if one is formed at all. The molecular ratios of proteinllipid for the first and second layers at pH 7.4 are shown in tables 1 and 2. TABLE 1 .-ADSORPTION BY CHOLESTEROL MONOLAYERS no. of layers molecules cholesterol wt.of protein A n in B.P.A. 1 620 0.01 I 6.0 B.P.A. 2 90 0-084 8.9 fibrinogen 1 4200 0.01 5.4 fibrinogen 2 440 0.089 9-1 insulin 1 120 0.009 9.5 insulin 2 14 0.09 13.2 molecules protein in mg dynes Icm protein224 STRUCTURE OF FILMS OF PROTEINS TABLE 2.-ADSORPTION BY STEARIC ACID MONOLAYERS protein B.P.A. B.P.A. fibrinogen fibrinogen lysozyme lysozyme insulin insulin no. of layers molecules stearic acid wt. pf protein molecules protein in mg 660 0 . 0 2 1 1 4 0 0.10 4200 0.018 830 0-098 1 4 5 0.020 35 0.080 1 1 5 0.01 8 30 0,084 A D in dynes/cm 5.2 7.4 8.5 1 3 . 4 7.0 8.8 1 6 . 2 1 7 . 7 FIG. 3.-The variation of surface pressure increase, An, with molecular ratio of B.P.A. to stearic acid, np/nLy at pH 1.9. It is evident that considerable surface denaturation of the protein will occur in forming the first sub-layer. The extent of this is seen by considering the area taken up by the protein. In table 3 the difference in the specific area of the protein in the first and second sub-layers is shown.TABLE 3.-sPECIFIC AREAS OF ADSORBED PROTEIN lipid cholesterol Y Y Y7 stearic acid Y ? # Y 7 7 protein B.P.A. fibrinogen insulin B.P.A. fibrinogen ly sozy me insulin specific area of protein in m2/mg 1st layer 2.4 2.6 2 . 8 1.2 1.4 1 . 4 1 . 4 2nd layer 0 . 3 6 0.33 0 . 3 2 0.33 0 . 3 3 0.43 0.35 Adsorption of B.P.A. by stearic acid films at pH 1 - 9 was studied to determine the effect of pH on the amount of protein adsorbed. Although A n was far greater than at pH 7.4 the adsorption followed the same pattern as before (see fig.3). The protein/ lipid relationships and the specific areas for the first and second layers of protein at this lower pH are shown in Table 4. TABLE 4.-ADSORPTION OF B.P.A. BY STEARIC ACID AT pH 1 . 9 A17 in wt. of B.P.A. ~ ~ ~ ~ , f i p T a " f ~ ~ dynes/ molecules stearic acid molecules B.P.A. in mg layer, in mz/mg cm no. of layers 1 640 0.022 0 . 1 4 1 3 . 4 2 1 7 0 0.086 0.46 16-6D. D . ELEY A N D D . G . HEDGE 225 Measurements of insulin adsorption at various pH values were carried out in an attempt to explain the mechanism of the interaction and the significance of the change in surface pressure. The initial surface pressure for both stearic acid and cholesterol monolayers was 2 dynes/cm throughout. As found by previous workers 99 10 a fairly sharp change in A17 occurs at the isoelectric point, which is at pH 5.6 for insulin (see fig.4). In order to compare insulin adsorption with that of a simpler compound, injection of polyacrylic acid under films of both lipids was carried out at various pH values. Here A n , which was never very large, fell to almost zero at pH 5 (see fig. 5). t I I I I I I I I i I I I I i I 1 I 2 7 8 - 6 2 3 4 5 I 2 I P " FIG. 4.-The effect of pH on the surface pressure increase, A n . produced on injection of insulin under films of cholesterol (open circles), and stearic acid (full circles). I I . I 1 I 1 I 1 2 I I I 1 I I 2 - I 3 4 5 6 7 D H FIG. 5.-The effect of pH on the surface pressure increase, A n , produced on injection of Polyacrylic acid under films of cholesterol (open circles), and stearic acid (full circles).The effect of using a urea substrate was to reduce Al7 considerably. Thus the value for B.P.A. adsorption by stearic acid fell from 4.6 dynes/cm to 2-2 dynes/cm. DISCUSSION The most striking fact emerging from the results is that, for the same protein, the ratios of the number of molecules of lipid to the number of protein molecules are approximately the same for both lipids. This is the more remarkable because a cholesterol molecule a t 10 dynes/cm has nearly twice the area of a stearic acid molecule at 2 dynes/cm surface pressure. A good indication is given here as to the structure of the lipo-protein layer, since it appears that the protein molecule always interacts with about the same number of lipid molecules.These numbers H226 STRUCTURE OF FILMS OF PROTEINS are of the same order as the number of amino-acid residues in the protein molecule. The values quoted by Waugh 11 for the number of residues for bovine serum albumin, fibrinogen and insulin are 588, 3380 and 103 respectively. Fevold 12 gives a value of 124 for the corresponding figure for lysozyme. These values compare fairly well with those given in tables 1, 2 and 4, the most marked discrepancy being with fibrinogen. It appears reasonable to assume from this that the first sub-layer of protein is arranged with one amino-acid residue under each lipid molecule. This result would definitely rule out the theory that the observed pressures are due to the globular protein molecules in their entirety " penetrating " the adsorbed monolayer, since there is insufficient free space for this to occur.A penetration of the lipid monolayer by the lyophilic side chains of the protein is, however, very likely. It is probable that the hydrophilic groups in the lipid monolayer interact with the peptide linkages of the protein. That this is dimensionally possible has been shown by molecular models. In a condensed unimolecular film cholesterol molecules are arranged vertically, with the hydroxyl group 1 ointing into the substrate.13 In this position the hydroxyl groups will be at the corners of rectangles approxi- mately 10 8, by 4 A. The length of one peptide group 11 is about 3.6 A, so the protein molecule can be arranged fairly conveniently beneath the monolayer in the manner suggested.Similarly, with stearic acid films, the area per molecule at 2 dynes/cm is 24 A2, giving a maximum separation of carboxyl groups of 4-9 A. Allowing for the rectangular shape of the carboxyl group, viewed vertically, inter- action with the peptide bonds is possible. It is noticed that the adsorption figures are all somewhat larger than the number of amino-acid residues. This may be caused by the injected protein failing to spread over the whole surface area, thus giving rise to an edge effect. An analogy may be drawn here with gas adsorption by solids, the B.E.T. theory of which allows for sorption of a second layer to commence before the first layer is coniplete.14 With fibrinogen, the rather larger discrepancy may possibly be correlated with the small secondary " peak ", which was demonstrated electrophoretically.Adsorption effected by peptide bonds is not unknown for examples are found in the work on water sorption by polypeptides. Water sorption on polyglycine, studied by Mellon, Korn and Hoover 15 is evidently by means of the peptide link, this being the only polar grouping present. Sponsler, Bath and Ellis 16 concluded that a peptide bond would adsorb a maximum of four molecules of water, but that steric factors can reduce this figure, which explains the observation that sorption by peptide bonds in solid proteins is comparatively unimportant compared to that by other polar groups. In solution, however, the peptide chains become consider- ably separated, as pointed out by McLaren and Rowen,l7 leaving the peptide bonds more free to interact with other groups, such as the hydrophilic groups of the lipid monolayers.It appears from the results obtained using a urea substrate that hydrogen bonds are important in lipid-protein interactions, just as they are in protein-water inter- actions. The action of urea is assumed here to be entirely one of a hydrogen-bond breaking agent, although Pauling 18 has pointed out that the ability of urea to act in this manner has not been tested adequately. In the work of Pankhurst4 on complex formation between sodium dodecyl sulphate and gelatin sols it was found that, at a pH above the isoelectric point of the protein, the amount of detergent needed to complete one layer corresponded to the number of peptide links in the protein.However, below the isoelectric point far less detergent was adsorbed. This was explained by postulating an ion-dipole association, and assuming resonance of the peptide link, by analogy with amides. The addition of hydrogen ions would inhibit this resonance, thus preventing the formation of a dipole. This is evidently not the case for protein adsorption by monolayers since both B.P.A. and lysozyme give a hydrophilic group peptide link relationship at pH values below the isoelectric points of the proteins.D . D. ELEY AND D . G. HEDGE 227 Considerable surface denaturation accompanies the formation of the first sub- layer of protein. The results of Fraser, Kaplan and Schulman showed adsorption of catalase on various lipid surfaces caused unfolding of the molecule, the degree of which was measured by the loss in activity of the enzyme.In the present work, the extent of surface denaturation is shown by a consideration of the specific area of the protein. The first sub-layer of protein occupies a greater area than a protein at an air-water interface. This is more marked with cholesterol, owing to the wider spacing of these molecules, though it is probable that the actual protein does not occupy a larger area under this lipid than it does under stearic acid. It appears that this first layer resembles more nearly the film obtained at an oil/water interface, for which Alexander and Teorell 19 obtained a value of 1.8 to 2.0 m2/mg for the limiting area, at zero compression, of serum albumin. The second sub-layer of protein, however, has a far smaller area, which appears to be approximately the same for all proteins under both lipids, with the possible exception of lysozyme, and B.P.A.at pH 1.9. In both cases the pH is well below the isoelectric point of the protein, which is pH 10.5-1 1 for lysozyme and pH 4.9 for B.P.A. It would appear that the area of the first sub-layer of protein depends on the lipid area, whereas the area of the second layer is constant, or depends on the pH of the substrate to some extent. Thus indications are that the second layer is of native protein. This is supported by measurements of the surface area of proteins by water adsorption. Bull 20 obtained a value of 0.238 m2/mg for the area of solid serum albumin. Allowing for some unfolding of the molecule in solution this compares with the values for the area of the second sub-layer obtained in the present work.A study of the weight of protein required to complete two layers under the lipid shows that these results can be compared with those of Elkes, Frazer, Schulman and Stewart.2 These workers obtained a vahe of 2.5 mg of protein per m2 of surface area for the amount needed to form a stable layer on the lipid surface. The weight of protein needed to complete one layer on a lipid monolayer is only 0-4 mg/m2 for cholesterol, and 0.77 mg/m2 for stearic acid, whereas that required for two layers is about 3.2 mg,'niz. Thus the value of 2.5 mg probably corresponds to two layers of protein molecules. The effect of pH on the adsorption gives some indication of the mechanism of the interaction.Previous workers,9~10 found the changes in surface pressure were related to the charge on the protein, the maximum pressure rise occurring when the charges on the protein and the lipid were of opposite sign. However, this is not the complete story, since the maximum pressure rise using insulin and stearic acid occurs when both are negatively charged (see fig. 4). That the inter- action is not purely ionic is also shown by the fact that the curve for insulin and cholesterol, an uncharged molecule. is of the same shape. Thus the change in AI7 with pH depends on the chemical behaviour of the injected protein, and is not simply a charge effect. A possible explanation of this phenomenon may be found in the properties of the insulin molecule.At a pH below 2-0 the molecular weight of insulin is 12,000. Studies in the bulk phase by Sjijgren and Svedberg21 have found a molecular weight of 36,000 at a higher pH, and surface studies by Fredericq 22 have shown that when the isoelectric point is reached insulin has a molecular weight of around 120,000. This change in molecular weight with pH could contribute to the change in 111 with pH. This is not the only factor, because a greater value of AI7 is obtained with insulin, even at pH 2, than is experienced with a similar amount of fibrinogen, which has a molecular weight of about 400,000. The most satisfactory explanation of the surface pressure change is that it is caused by penetration of the non-polar side chains of the protein molecule. Support for this idea is given by comparing the energies of adsorption for different proteins under stearic acid films.23 Further, it is possible, using an analysis similar to that employed by Pethica,24 to calculate the number of side chains contributing to A n .228 STRUCTURE OF FILMS OF PROTEINS Indications are that about half the available non-polar side chains penetrate the stearic acid film in forming the first sub-layer of protein.The dependence of A17 upon the number of lyophilic groups in the molecule is shown again by the effect of pH on the adsorption of polyacrylic acid. For the uncharged molecule, at pH 1.9, a low value of A n is observed, owing to the low hydrocarbon content of the material; at pH 5, A17 decreases almost to zero. Since this is the case with both cholesterol and stearic acid it is evident that the interaction is not purely ionic.The decrease in A17 with increasing ionization of the polyacrylic acid rules out an ion-dipole mechanism, since stearic acid has a considerable dipole at pH 5, as shown by surface-potential measurements.25 The reason for the pH effect found here is shown in the work of Crisp,26 who found films of polyacrylic acid to be stable up to pH 3. On increasing the pH, no film can be detected. This was explained by assuming that the electrical potential set up by ionization lowers the energy of adsorption of the ionized residues, rendering the net energy too small to stabilize the film. These investigations support Danielli’s model of the plasma membrane, and give it a quantitative signiticance.The analogy is necessarily very approximate, as pointed out by Cheesman and Davies,27 since the surface tensions employed are far greater than the interfacial tensions found in living cells. Nevertheless, the structure of a lipo-protein layer has been determined, and may be compared to the cell membrane. The authors wish to thank the Department of Scientific and Industrial Research for the award of a maintenance grant to D. G. H., Dr. R. A. Kekwick for the fibrinogen sample used, and Dr. T. Malkin for a specimen of stearic acid. 1 Putnam and Neurath, J. Amer. Chern. Soc., 1944, 66, 692. 2 Elkes, Frazer, Schulman and Stewart, Proc. Roy. Soc. A., 1945, 184, 102. 3 Fraser, Kaplan and Schulman, Faruday Society Discussions, August, 1955. 4 Pankhurst, Surface Chemistry (Buttenvorths, 1949), p. 109. 5 Danielli and Davson, Permeability of Natural Membranes (Cambridge University 6 Sher and Sobotka, J. Colloid. Sci., 1955, 10, 125. 7 Schulman and Rideal, Proc. Roy. Soc. B, 1937,122, 29. 8 Schulman, Biochem. J., 1945, 39, lvi. 9 Doty and Schulman, Faruday Soc. Discussions, 1949, 6, 21. Press, 1943), p. 60. 10 Matalon and Schulman, Faraday Soc. Discussions, 1949, 6, 27. 11 Waugh, Adbances in Protein Chemistry, 1954, 9, 325. 12 Fevold, Advances in Protein Chemistry, 1951, 6, 230. 13 Adam, The Physics and Chemistry of Surfaces (Oxford, 1941), p. 81. 14 Brunauer, Emmett and Teller, J. Amer. Chem. Soc., 1938, 60, 309. 15 Mellon, Korn and Hoover, J. Amer. Chem. Soc., 1951,73, 1870. 16 Sponsler, Bath and Ellis, J. Physic. Chem., 1940, 44, 996. 17 McLaren and Rowen, J. Polymer. Sci., 1951, 7 , 289. 18 Pauling and Corey, Les Proteins-Rapports et Discussions, ed. R. Stoops (Institut 19 Alexander and Teorell, Trans. Faruday Soc., 1939, 35, 733. 20 Bull, J. Amer. Chern. Soc., 1944, 66, 1499. 21 Sjogren and Svedberg, J. Amer. Chem. Soc., 1931: 63, 2657. 22 Fredericq, Biochim. Biophys. Actu, 1952, 9, 601. 23 Eley and Hedge, unpublished work. 24 Pethica, Trans. Faraday SOC., 1955, 51, 1492. 25 Glazer and Dogan, Trans. Faraduy Soc., 1953, 49,448. 26 Crisp, J. Colloid Sci., 1946, 1, 161. 27 Cheesman and Davies, Advances in Protein Chemistry, 1954, 9, 439. intern de Chimie Solvay, Brussels, 1950), p. 63.

ISSN:0366-9033    

DOI:10.1039/DF9562100221

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

COMPARTMENTALIZATION OF THE CELL SURFACE OF YEAST IN RELATION TO METABOLIC ACTIVITIES" BY ASER ROTHSTEIN The University of Rochester, School of Medicine and Dentistry, Received 7th February, 1956 Rochester 20, New York, U.S.A. The surface of the yeast cell is a complex structure, containing enzymes, permeability barriers, cation-binding sites, and active-transport mechanisms. Among the enzymes are phosphatases and saccharases. Cell surface reactions in sugar uptake also possess many of the properties of enzyme reactions. There are at least three distinct permeability barriers, one to proteins, one to divalent cations, and one to univalent cations. The latter two barriers are also impermeable to anions, and therefore the net movements of K+ across them are largely balanced by movements of H+ in the opposite direction.Peripheral to the divalent cation barrier there are cation-binding sites of two chemical species, both involved in glucose uptake. Internal to the same barrier there are reactions in sugar uptake which are influenced by the extracellular H+ and K+, but not by divalent cations. Active transport systems are capable of moving K+ and H2PO4- into the cell against large activity gradients, when substrates are present, but K+ is retained in high concentrations in the cytoplasm by a barrier whose permeability is also dependent on metabolic factors. The complex structures and functions of the cell surface impose limitations upon analysis of its behaviour as a membrane, except for a few substances. The properties of the cell surface as a semipermeable membrane have been characterized in some detail by many investigators.It has been established that the movements of a variety of substances across the membrane can be predicted by modifications of the laws of diffusion, Furthermore, the rate at which a given substance penetrates the membrane is often related to its lipid solubility, its mole- cular size, and sometimes to its acidic or basic properties. On the basis of per- meability behaviour, several types of membrane structure have been suggested involving lipid and aqueous phases in mosaic or concentric arrangements. In recent years, the concept of the cell surface as a membrane of relatively constant mechanical and physico-chemical properties, controlling the movements of substances by its fixed permeability properties, has given way to the concept of a structure which also participates actively in certain physiological or biochemical functions.1 A number of specific enzymes has been localized at the surface of the cell which are concerned with the digestion of certain extracellular substrates.Other cellular activities in which the cell surface has been implicated are the synthesis of extracellular macromolecules, both protein and carbohydrate, and in active transport phenomena, Active transport is of special interest insofar as membrane phenomena are concerned, because it involves the movement of a substance against its activity gradient with energy supplied by metabolic reactions. A number of mechanisms of active transport have been postulated, many of which require the presence in the membrane of a " carrier system " and of energy-supplying metabolic reactions In some theories it is suggested that the carrier is itself' an enzyme.*This paper is based on work performed under contract with the United States Atomic Energy Commission at the University of Rochester Atomic Energy Project, Rochester, N.Y. 229230 COMPARTMENTALIZATION OF CELL SURFACE Are the physiological and biochemical functions of the cell surface associated with modifications of the single membrane visualized from permeability studies ? Or is the cell surface a more complicated structure? In the present paper, evidence will be presented supporting the latter view. In fact, it is suggested that the cell surface is a multimembranous, compartmentalized system.The data are taken from studies with yeast cells but the conclusions drawn may be applicable to other types of cells as well. The cell-surface phenomena have been explored by various techniques most of which depend on the principle that the peripheral portions of the cell are in- fluenced by the extracellular rather than the intracellular environment. The outside of the cell is directly subjected to the extracellular environment. Those portions of the cell periphery which are protected by a membrane are influenced only by those extracellular substances which penetrate beyond the membrane. In the present discussion, a number of phenomena associated with the cell surface will be reviewed briefly. These include (i) enzymes of the cell surface, (ii) active transport of ions, (iii) permeability barriers, and (iv) cation binding.An attempt will then be made to construct a crude model of the cell surface from the data on hand, More detailed discussions of certain aspects of the problem are to be found in recent reviews (cell surface enzymology,l sugar uptake,2 and electrolyte metabolism 39 4). ENZYMIC ACTIVITIES OF THE CELL SURFACE A number of specific enzymes have been localized on the cell surface of yeast by direct procedures. For example, P32-labelled phosphate compounds are hydrolyzed by the cells, with no mixing of cellular phosphate with P32-labelled phosphate derived from the substrate.5 Thus the phosphatases responsible for the hydrolysis must be located on the surface of the cell. Saccharases such as invertase, lactase, and maltase have been studied by several techniques : (a) the pH activity curve for invertase 6 or lactase 7 in the living cell is the same as that for the purified enzymes.Thus the enzyme must be directly exposed to the extracellular pH and not protected by the cytoplasmic buffers ; (b) mild acid treatment of the cells destroys the invertase and maltase activity, but not the fermenting ability of the cell; 8 (c) the invertase activity of the cell is inhibited by UO;f, which does not penetrate the cell.9 During the uptake of sugar, interactions occur at the cell surface which have properties of enzyme reactions.2 For example, uranyl ion which acts on the cell surface, specifically but reversibly blocks sugar uptake. The action is not an interference with diffusion of glucose through the membrane, for the kinetics are those of a saturation phenomenon (Michaelis-Menten) indicating the forma- tion of a complex between glucose and a cellular constituent present in limited amount.The rate of sugar uptake is also influenced by extracellular H+ and K+ con- centrations despite a constant level of these ions in the cytoplasm. In the presence of K+, the relationship between the rate and pH is a wide plateau from pH 2.0 to 8.5, but in the absence of K+, it is a biphasic curve with optima at pH 4.5 and 8.5. If the action of H+ is on a surface enzyme reaction essential to sugar uptake, then the biphasic curve suggests that there are two enzymes. A well-known enzyme which catalyzes the phosphorylation of glucose is hexokinase, crystal- lized some years ago.It has a pH optimum of 8.5. Another hexokinase with a pH optimum of 5.0 has recently been separated by starch electrophoresis.10 Preliminary data suggest that the effects of H+ and Kf on cell surface reactions of the intact cell can be explained in terms of the actions of these ions on the two hexokinases.A. ROTHSTEIN 23 1 Other evidence suggesting that glucose is metabolized at the cell surface has been summarized in detail elsewhere.2 Only a few of the arguments can be presented here. (i) Extracellular K+ at low pH stimulates fermentation but not respiration of glucose.11 If the surface reactions are simply a delivery system to transfer sugar into the cell, then both modes of metabolism should be stimulated by the increased delivery of sugar.(ii) When the extracellular pH is above 7-0, there is a large increase in the glycerol production during fermentation, despite the fact that there is no change in intracellular pH. Again, if glucose were simply transported into the interior of the cell, alteration of the rate of transport should not alter the nature of the fermentation. (iii) The sugar specificity of the cell is the same as that of the enzyme hexo- kinase. Other sugars do not penetrate into the cell. For galactose, the cell can be adapted to use this sugar. In the process of adaptation the cell becomes able to take up the sugar, and at the same time the enzyme, galactokinase appears.12 ACTIVE TRANSPORT OF IONS A variety of substances can be taken up by the yeast cell against their activity gradients.Potassium uptake has been studied most intensively. In the presence of substrate, K+ is taken up in exchange for H+ derived from metabolic reactions, with resultant acidification of the medium.131 14 In the process, both ions move against large activity gradients. The cytoplasmic concentrations of K+ and Hf are approximately 2 x 10-1 and 1 x 10-6 M. Yet there is a K+ uptake and Hf excretion when the environmental concentrations are both 1 x 10-4 M. The details of the process and postulated mechanisms are discussed elsewhere.3~ 4 The transport mechanism for K+ is highly specif1c,3 discriminating between K+ and Na+ by a factor of 20 to 1. Ordinarily, yeast is rich in K+ and low in Na+. However, under certain conditions the cells become rich in Na+.Such cells if placed in a medium containing K+ will actively 15 extrude Na+ in exchange for K+. The extrusion system is highly specific for Na+ as compared to K+. Phosphate uptake also proceeds against an activity gradient in the presence of glucose. For example, the orthophosphate concentration of the cytoplasm is about 0.02 M, but absorption of phosphate occurs 16 even though the con- centration in the extracellular medium is as low as 1 x lO-4M. The process has not been studied as intensively as has the transport of K+. However, it seems likely that the formation of phosphorylated intermediates of fermentation may play a direct role. If divalent cations are present during the absorption of phosphate, they, too, can be absorbed as a complex ion with phosphate.16 PERMEABILITY BARRIERS The yeast cell possesses an effective permeability barrier to many substances.Organic acids such as pyruvic,l7 lactic, and others penetrate readily and can be respired, but if the pH is raised so that the acids are in the ionic form, the anions do not penetrate and are not respired. It can be generalized that the yeast cell is impermeable to anions. For example, phosphate esters which are normal intermediates of fermentation, when added to the medium cannot be directly utilized because they cannot penetrate into the cell.5 Even orthophosphate, during its active absorption, cannot readily diffuse through the membrane, as shown by the very low rate of exchange of P32-labelled phosphate in the medium with that in the cells.16 The impermeability of the membrane to anions is also reflected in the fact that in Kf-uptake or loss, the electrical balance is largely maintained by an equivalent loss or gain of H+ rather than by an inflow of anions. The permeability of yeast to various substances has been studied by volume of distribution techniques.18 Those substances that penetrate the cell membrane,232 COMPARTMENTALIZATION OF CELL SURFACE distribute in the total volume of the cell.These include substances such as alcohols and undissociated acids. Other substances, such as sugars and electro- lytes, distribute in a volume equal to about 10 % of the cell, probably equivalent to the " cell wall space " observed in plant cells. Very large molecules such as inulin and protein do not distribute in any fraction of the cell volume.Divalent cations behave in a manner similar to phosphate. That the mem- brane is relatively impermeable to the divalent cations is shown by the absence of exchange between cellular and extracellular ions, measured with Mn54 or Ca45, even during their rapid uptake.19 T i m e i n m i n u t e r FIG. 1.-The uptake and loss of K+ by yeast cells at different initial concentrations of K+ in the presence of glucose. Exchanges of univalent cations can occur readily. It has already been pointed out that K+ is rapidly accumulated against large concentration gradients in exchange for H+ derived from metabolic reactions. But the exchanges in the reverse direction, with the concentration gradient, proceed at a much slower rate.For example, cells suspended in distilled water lose K+ only slowly 13,20 despite the fact that the cellular concentration is 0-2 M. It can be concluded that the cell is able to accumulate and maintain high concentrations of K+ because of a combination of two factors, the active inward transport and the low per- meability which slows the outward diffusion. At very low concentrations, the two forces balance each other. The active transport of K+ is exactly balanced by its outflow, and no net changes of K+ occur. Thus yeast given glucose and potassium (3 x 10-4 M) takes up the cation until the extracellular concentration is reduced to about 5 x 10-5 M (fig. 1). With no potassium added, the cell leaks the cation until the extracellular K+ is also 5 x 10-5 M.If the initial concentration of K+ is 5 x 10-5 M, 1ittIe change in concentration occurs, the rate of uptake is exactly balanced by the rate of leakage. Because the movements of K f are balanced by those of H+, the steady state between inflow and outflow is shifted markedly in the upward direction as the pH is reduced. For example, it is increased some 20-fold to 1 x 10-3 M when the pH is reduced from 4.5 (the experiment of fig. 1) to pH 3.0A . ROTHSTEIN 233 The permeability of the cell membrane to K+ has been difficult to study, especially during metabolism, even with isotope techniques, because of the great disparity that may obtain between the rates of inflow and outflow. In addition the short half-life of K42 is discouraging. The problem has been investigated by the application of the technique of fractional elution to a column of yeast cells.K+-free medium is continually passed through the column of cells at an appropriate rate. Potassium never accumulates (except to the minimal extent allowed in order to make accurate analyses), and absorption of the ion is minimized. By use of a standard fraction collector and a sensitive flame photometer, the out- flow of K+ can be accurately followed for long periods of time under a variety of conditions, without any concomitant inflow to obscure the results. T i m e i n minutes FIG. 2.-Outflow of K+ from a column of yeast cells into a K-1-free eluting medium. A, 1 x 10-4 M HgC12, pH 3.5, aerobic; B, pH 2.0, aerobic; C, pH 4.0, aerobic; D, pH 4.0, anaerobic. The resistance of the potassium barrier is dependent on many variables in- cluding 0 2 , substrate, pH, temperature, other cations, and inhibitors.In the resting cell (no substrate) at 25" C in the presence of 0 2 , at pH 45, the rate of K+ leakage is 10 mM/kg h, but if 0 2 is excluded, the rate is reduced to a much lower level (fig. 2). If the temperature is reduced to 5" C , the rate is also reduced to very low levels, less than 2 mM/kg h. It should be kept in mind that the cells contain some 200 mM/kg, so that the rates of leakage represent a loss of only a few percent of the cellular Kf per hour. Furthermore the rates are very small compared to those at which the metabolizing cell can take up K+ against the Concentration gradient (as high at 400 mM/kg h for short periods of time).In the absence of salts in the extracellular medium, outflow of K+ from cells without substrates is largely balanced by an inflow of Hf. For this reason the medium tends to become more alkaline, and the outflow of K+ is accelerated as the H+ concentration is increased. The effect of substrates is of considerable interest. For example, glucose, which induces a rapid Kf-uptake, at the same time induces a marked increase in the rate of K+-outflow. Values of 20 to 25 mM/kg h are observed under aerobic or anaerobic conditions (fig. 2). The substrate-effect is somewhat confusing234 COMPARTMENTALIZATION OF CELL SURFACE because glucose and pyruvate markedly increase the K+ outflow, fructose has a smaller effect, whereas ethyl alcohol and lactate have no effect.Yet all of these substrates are reported to induce K+ uptake.22 Certain metabolic inhibitors also have a dramatic effect. Mercuric ion has the greatest effect. At 1 x 10-4 M, the rate of K+ outflow is increased 11-fold to 110 mM/kg h or 50 % of the cellular K+ per hour (fig. 2). At 1 x 10-5 M, there is also an effect, but only after a delay of about 40 min. Dinitrophenol also increases the outflow of K+, but not to the same extent as Hg2f. Azide and cyanide have little effect unless glucose is also added, in which case there is a large increase in K+-outflow which persists for about 1 h followed by a return to low levels. The properties of the cellular barrier to K+-outflow can be summarized briefly. In the first place, the barrier is essentially impermeable to anions, so that the K+ outflow is an exchange of K+ for H+ (or for other extracellular univalent cation if present).Whether the limiting resistance is toward K+ or toward H+ is difficult to determine. Despite the large outward gradient of K+ and inward gradient of H+, the exchange rate is very small relative to that which occurs in the reverse direction during active metabolism. Thus the barrier is a real one and an effective one. The resistance of the barrier is not fixed, but can be increased markedly by temperature, substrates and metabolic inhibitors. This suggests that K+ passes out of the cell in large measure through the same channels in the barrier which are used to pump it into the cell by energy expenditure. BINDING OF CATIONS BY THE CELL SURFACE It has been pointed out that the yeast cell possesses a peripheral barrier to the inward and outward diffusion of divalent cations which prevents exchanges between the medium and the bulk of the cytoplasm.There are, however, anionic sites on the outer surface of the cell which bind the extracellular cations in a stable but reversible complex.19 The surface anions are of two chemical species, poly- phosphate or nucleic acid phosphate, and protein carboxyl. Both species bind cations, but the former give complexes of considerably greater stability. The interaction of cations and cell-surface anions can be characterized by a simple mass-law equation based on a 1 -to-1 combination. Relatively stable complexes are formed with divalent cations. especially UO$+. Univalent cations such as Na+ and K+ also can be bound in competition with divalent cations.However, the complex with the univalent cations is considerably less stable. The anionic groups of the cell surface are involved in certain metabolic activities. For example, if the “ phosphate groups ” are saturated with UO3+, the cells are unable to ferment glucose or other sugars and are able to respire glucose at only 40 % of the normal rate. If the “carboxyl groups” are also saturated by the use of higher concentrations of UOg+ then the cells are unable to respire glucose at all.23 In addition, blockage of the “carboxyl groups” results in the inhibition of the invertase activity of the cell.9 A GENERALIZED CONCEPT OF THE CELL SURFACE STRUCTURE The available information does not give a clear picture of the cell surface structure.The experimental tools are too clumsy. Nevertheless, it can be in- ferred that no single membrane is responsible for all of the cell surface properties, but that the cell surface is a compartmentalized structure. By the volume of distribution technique mentioned earlier, it was found that the cell possessed an outer zone, probably the cell-wall space, into which ions and small molecules could penetrate but into which larger molecules such as inulin and protein couId not. Conway 18 presents evidence that certain redox reactions of metabolism which supply energy for the active transport of K+ and H+ are located in the outer zone. Evidence of a compartmentalization is also found when the surface action of divalent and univalent cations is compared.Both are bound by cell surfaceA. ROTHSTEIN 235 anions in a competitive manner.19 However, there seems to be a separate per- meability barrier for divalent cations and for univalent cations. Kf and H+ can penetrate beyond the divalent cation barrier before reaching a univalent cation barrier. This can be demonstrated in several ways. The first is concerned with the distribution of K+ shown in fig. 3 (measured 21 by isotope technique with K42). Low concentrations of Kf and high concentrations of yeast were used so that the amount of K+ distributing the surface compartments was a measurable fraction of the total extracellular Kf. Under aerobic conditions, with no sub- strate (which permits slow outflow of intracellular K+) there is an initial, rapid I I I I 1 L 10 2 0 3 0 4 0 5 0 6 0 7 0 8 0 lo( Time i n minutes FIG.3.-The effect of Mn2+ on K42 uptake by yeast cells under aerobic and anaerobic conditions in the absence of substrate. The yeast concentration was 400 mg/ml, at pH 4.0. The extracellular K+ concentration remained at a level of 2-0 to 2.5 x 10-3 M. A, aerobic ; A', aerobic with 0.1 M KCl added at 60 min ; B, aerobic with 1 x 10-3 M MnC12 ; C, anaerobic ; C', anaerobic with 0.1 M KCI added at 60 min ; D, anaerobic uptake of K42 followed by a slower prolonged rate. At the end of 1 h, about 15 to 20 % of the K+ of the cell has been exchanged. In the presence of high concentrations of Mn2+, the curve is lower by a constant increment. Addition of high concentrations of K+ at the end of 1 h results in the back-exchange of a small amount of K42.In contrast, under anaerobic conditions (which permit little outflow of intra- cellular K+), only an initial, rapid uptake is observed, amounting to about 2 to 3 % of the cellular Kf. Mn2+ can displace only a portion of the K42 from the cell, but high concentrations of Kf can rapidly displace nearly all of the K42. Apparently, then, K42 can distribute in a fraction of the cell outside of the permeability barrier to K+. A fraction of the K42 is associated with the cation binding sites of the outer surface and this portion can be competitively displaced by Mn2+. Another fraction, however, penetrates into a compartment inaccessible to Mn2f and cannot be displaced except by unlabelled Kf.A similar conclusion is based on the effects of the ions on surface reactions in sugar uptake. At pH 2.5 the rate of sugar uptake is depressed. However, the rate is raised to maximal levels in the presence of Kf, despite the fact that under with 1 x 10-3 M MnC12.236 COMPARTMENTALIZATION OF CELL SURFACE the conditions of the experiment, no K+ is absorbed.11 Divalent cations stimulate only slightly (table 1). Nevertheless, divalent cations do not influence the stimulation due to K+ even though the concentrations are sufficient to displace all of the K+ from the surface binding sites. The interaction of K+ on sugar uptake must occur in a compartment of the cell surface inaccessible to Mg2+ and Ca2+. TABLE TH THE EFFECTS OF VARIOUS IONS ON FERMENTATION OF GLUCOSE IN THE PRESENCE AND ABSENCE OF POTASSIUM pH 2.7 ; ion concentration, 0.02 M control Li Na K Rb cs M g Ca individual ions individual ions COz in pl./mg h 24.1 44.0 27.2 43.6 28.2 44.3 44.0 - 33.8 41 5 26.5 44.5 28.6 46-4 30-0 43.4 plus K C02 in pl./mg h All values are averages of 2 experiments.Finally, flooding the cell with divalent cations does not interfere with K42 exchanges into the cytoplasm in the absence of sugar (fig. 3, aerobic experiment) and only reduce the active K+ uptake during metabolism to a small extent (fig. 4).*1 Time in minutes FIG. 4.The effect of Mg*+ on K+ uptake by yeast given glucose. A, with 1 x 10-2 M MgCL ; B, no MgCl2. The site of K+ transport and exchanges into the interior of the cell is also inaccessible to Mg2+ and Mn2+.The barriers to divalent and univalent cations are both impermeable to anions, for the movements of K+ into the outer compartment or into the cell is an exchange of cations rather than a movement of K+ together with an anion.A. ROTHSTEIN 237 Experiments on acid inactivation of cell surface enzymes are of some interest. Mild acid destroys the invertase 8 and phosphatase activity 21 of the cell, but has little effect on the fermentative activity.8 The former enzymes are directly accessible to the Hf of the medium. The fermentative enzymes themselves are sensitive to acid, so in the cell they must be protected by a barrier. Stronger acid will, however, reduce the fermentative activity of the cell.21 In this case, protection is afforded by extracellular K+, in concentrations the same as already present in the cytoplasm. Thus compartmentalization is again indicated, with invertase and phosphatase exposed directly to the medium and fermentative re- actions in a compartment within the cell surface, protected by a membrane across which H+ and K+ can exchange.There is some evidence that the phosphatases and saccharases are on the outside surface of the cell wall. Thus immune serum prepared against yeast phosphatases will inhibit the phosphatases of the living cell 24 even though proteins cannot penetrate the cell-wall space.18 Also colloidal starch which cannot pene- trate the cell-wall space is hydrolyzed by a surface-bound amylase.25 The cell wall may not be a biologically inert structure at all. In fact chemical evidence indicates that the cell wall, although largely composed of the polysaccharides glucan and mannan, also contains proteins and lipids.26 DISCUSSION On the basis of the information presented in this paper, a crude model of the cell surface is shown in fig.5. No further argumsnts will be advanced to support the details of the model. Perhaps other versions would fit the known information Cell surface Cell interior A- Cell wall- Glucose - Ma ..._ Phosphate pump Me taphospha te Rerpira tory on K+ - H+ Phorphatax ''.I \ FIG. 5.-Schematic representation of the cell surface of yeast. as well or better. In other cells the arrangement may be quite different. It is, however, important to realize that the cell surface behaves as a simple permeability barrier to only a limited number of biologically important substances.It behaves in a far more complex and often poorly understood manner toward many bio- logically important substances, in some cases transporting them against an activity gradient. The complex structures and functions of the cell surface impose serious limitation upon any simple analysis of its behaviour as a membrane,238 RED CELL PERMEABILITY 1 Rothstein, The Enzymology of the Cell Surface, Protoplasmatologia, I1 E 4 (Springer 2 Rothstein, Symp. SOC. Expt. Biol., 1954, 8, 165. 3 Conway, The Biochemistry of Gastric Acid Secretion (Charles C. Thomas Publishing 4 Rothstein in Electrolytes in Biological Systems, ed, Shanes (Amer. Physiol. SOC., 5 Rothstein and Meier, J . Cell. Comp. Physiol., 1949, 34, 97. 6 Wilkes and Palmer, J. Gen. Physiol., 1932, 16, 233. 7 Myrback and Vasseur, 2. Phsiol. Chem., 1943, 277, 171. 8 Myrback and Willstaedt, Arkiv Kemi, 1955, 8, 367. 9 Demis, Rothstein and Meier, Arch. Biochem. Biophys., 1954, 48, 55. 10 Scharff and Rothstein (unpublished results). 11 Rothstein and Demis, Arch. Biochern. Biophys., 1953, 44, 18. 12 Spiegelman, Reiner and Morgan, Arch. Biochem., 1947, 13, 113. 13 Rothstein and Enns, J. Cell. Comp. Physiol., 1946, 28, 231. 14 Conway and Brady, Biochem. J., 1950, 47, 360. 15 Conway, Symp. Soc. Expt. Biol., 1954, 8, 297. 16 Goodman and Rothstein (unpublished observations). 17 Barron, Ardao and Hearon, J. Gen. Physiol., 1950, 34, 211. 18 Conway and Downey, Biochem. J., 1950, 47, 347. 19 Rothstein and Hayes, Arch. Biochem. Biophys., 1956, 63, 87. 20 Scott, Jacobson and Rice, Arch. Biochem., 1951, 30, 282. 21 Rothsteh and Bruce (unpublished observations). 22 Orskov, Acta Physiol. Scand., 1950, 20, 62. 23 Rothstein, Meier and Hurwitz, J . Cell. Comp. Physiol., 1951, 37, 57. 24 Derrick, Miller and Sevag, Fed. Proc., 1953, 12, 196. 25 Rahn and Leet, J. Bact., 1949, 58, 714. 26 Northcote and Horne, Biochern. J., 1952, 51, 232. Verlag, Vienna, 1954). Company, Springfield, Illinois, 1952). 1955).

ISSN:0366-9033    

DOI:10.1039/DF9562100229

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

238 RED CELL PERMEABILITY STRUCTURE AND FUNCTION TN RED CELL PERMEABILITY BY W. D. STEIN AND J. F. DANIELLI Department of Zoology, King’s College, London, W.C.2 Received 24th January, 1956 It is postulated that facilitated diffusion through the cell plasma membrane takes place through a hydrogen-bonding stereochemically specific membrane component which extends through the thickness of the membrane. It was predicted that facilitated diffusion will be slowed by the presence of other hydrogen-bonding molecules, as a result of com- petition for hydrogen-bonding sites in the membrane component. This competitive inhibition is shown to be produced by ethylene glycol, propylene glycol, ethanol, butanol, urethane and phenol. Data are also presented for inhibition by copper, which acts as a non-competitive inhibitor.The passage of molecules through the membranes of red cells occurs by at least three * distinct types of processes : (i) difusion (involving no specific structural relationship between the membrane and the permeating species) ; (ii) facilitated difusion (involving a specific structural relationship between the membrane and the diffusing species) ; (iii) active transport (involving both a structural relationship between the membrane and the permeating species, and the supply of energy from metabolism).l~ 2 * in some cells at least five, including mass flow through pores, and pinocytosis.W. D . STEIN AND J . F . DANIELLI 239 These three processes will be considered in relation to (a) competitive inhibitors, (b) non-competitive inhibitors, (c) hydrogen bonding systems.The interpretation of permeability results cannot proceed without knowledge of the physical nature of the membrane of the cell. In most of the work which will be discussed in this paper, the red cell has been used as the experimental material. The main body of information about the nature of the cell plasma membranes can be given under the following six headings. The red cell contains in its membrane roughly equal weights of lipoidal molecules and of complex protein molecules. It also contains complex polysaccharides which are responsible for the blood grouping phenomena, and a considerable variety of enzymes. (2) Gorter examined the physical properties of the lipoid component of red cells and found that the lipoid in the membrane was just sufficient to constitute a layer two molecules thick.He therefore suggested that the most important component of this membrane is a bimolecular leaflet of lipoidal molecules. (3) As a result of surface tension and related studies Danielli and Harvey concluded that in addition to a lipoid layer the plasma membrane must have adsorbed at each of its interfaces with water at least a unimolecular layer of protein. (4) A variety of optical studies including electron optical studies indicate that the membrane of the red cell contains a thin layer of lipoid molecules with protein molecules on either side. Some of these protein molecules appear to be oriented in the plane of the membrane and other protein molecules are oriented perpendicular to the plane of the lipoid layer.The molecules which are oriented perpendicular to the lipoid layer form a dilute protein gel extending some distance into the interior of a cell. This layer of protein gel is probably not of much importance in determining the permeability properties of the plasma membrane. (5) Impedance studies indicate that the membrane has a relatively high electrical resistance, showing that it is not readily permeable to ions, and that it has a static capacity of the order of 1 pF cm-2. These properties are such as would be expected from a bimolecular lipoid leaflet. (6) Cytochemical studies have shown that certain enzymes are present in or close to the surface of the plasma membrane. The enzyme alkaline phosphatase is quite commonly found in this situation.There is also some indication that phosphokinases may be closely associated with the permeability function of the plasma membrane. The analysis of diffusion kinetics made by Danielli in the 1930’s showed that the rate of permeation of plasma membranes by most organic molecules is that which would be expected for penetration of an approximately bimolecular leaflet of lipoidal molecules by a process of activated diffusion. The correspondence between theoretical predictions and experimental results was so good that we must regard the plasma membrane as being a thin lipoid layer, at least to a first approximation. However, the analysis of diffusion kinetics also showed very clearly that there are certain molecules which penetrate certain cells by processes other than simple activated diffusion. These molecules are often those of most physiological in- terest.Indeed, it may be said that one of the great advantages of deriving a quan- titative theory to describe the rate of permeation of molecules for which there was no specific permeability process, was that it became quite clear which molecules were permeating by specific processes. This has cleared the way for the analysis of the true specific processes, which are of the greatest physiological interest. At the present time it is convenient to divide the specific processes into two groups. The first of these groups involves molecules such as glycerol, urea, the sugars and (1) Gross chemical analysis.240 RED CELL PERMEABILITY the amino acids. These molecules often penetrate cell membranes by a process of diffusion which, however, is in some way conditioned so that only a very limited range of chemical structures can pass the membrane.Other apparently closely related molecules may be completely unable to penetrate by the specific process. For example, glucose often passes through cell membranes at a rate about 1000 or more times greater than it should do by simple activated diffusion. The specific process by which this occurs is not available to methyl glucoside. This first specific process we now call facilitated diffusion. The second type of specific process is usually called active transport and involves, in addition to a specific limitation to a small range of chemical types, the provision of energy by the cell.We do not know the mechanisms of either of these two specific processes. But certain sug- gestions have been made in recent years concerning possible mechanisms, and it is the purpose of this paper to approach the analysis of the mechanism of facilitated diffusion. If we list the characteristics of those molecules which penetrate cell membranes by diffusion, but at a rate greater than would be predicted by simple activated diffusion, we find four main characteristics. (i) Small hydrogen-bonding molecules such as water, methyl alcohol, urea and formamide penetrate more rapidly than they would be expected to permeate a homogeneous lipoid layer. (ii) Relatively large hydrogen-bonding molecules, such as the sugars, penetrate cell membranes (in particular instances) very much more rapidly than would be expected ifthey penetrated by non-specific diffusion through a lipoid layer.(iii) Molecules such as sugars, when penetrating by activated diffusion, have high temperature coefficients for the rate of penetration. When penetrating by facilitated diffusion the tempera- ture coefficient for penetration is usually much lower. (iv) Penetration by activated diffusion is not normally inhibited by enzyme poisons; on the other hand pene- tration by facilitated diffusion or by active transport is frequently brought to a standstill by specific poisons. For example, the penetration of cells by glucose by facilitated diffusion is frequently inhibited by phloretin. The main factor responsible for the slowness with which molecules such as sugars will penetrate a lipoid layer by activated diffusion is the formation of hydro- gen bonds with water.In order to enter a lipoid layer a sugar will normally need to break simultaneously five or six hydrogen bonds, so that the activation energy is very high and the process is correspondingly very slow. Rapid penetration can only be assured by a mechanism which evades the necessity for breaking a large number of hydrogen bonds simultaneously. So far as can be seen this can only be achieved by providing a hydrogen-bonding structure which extends right through the membrane, i.e. by the provision of hydrogen-bonding channels which com- pletely replace lipoid through the thickness of the membrane. It has been suggested that such channels may be provided by the presence of protein lamellae which extend through the thickness of the membrane.2 When two protein lamellae are face to face the space between them is a hydrogen-bonding space which will have a highly specific configuration, and which thus could be held accountable for the highly selective nature of the pores which appear to exist in cell membranes.Having postulated the presence in the membrane of specific hydrogen-bonding (possibly protein) pores, we are confronted with the problem of finding out whether in fact such pores exist. So far three suggestions have been made which enable us to approach this problem. (i) If the pores are composed of proteins, reagents for the typical groupings of proteins should inhibit facilitated diffusion. For example, it has been suggested that diazonium hydroxides and dinitrofluorobenzene will inhibit facilitated diffusion.3 This has now been shown to be the case for dinitro- fluorobenzene.4 (ii) If hydrogen bond formation through the thickness of the membrane is an integral part of facilitated diffusion, then the presence of alternative molecules which will block such hydrogen-bonding groups, or alternatively which will compete for the hydrogen-bonding groups, should diminish the rate of facilitated diffusion.(iii) It may be possible by tagging the pores with specific components,W. D . STEIN A N D J . F . DANIELLI 241 for example dinitrofluorobenzene, to fractionate the membrane and ascertain the precise chemical nature of the pore components. In this paper the main new experimental contribution arises from the second of these suggestions, i.e.we have investigated the action upon facilitated diffusion of certain inhibitors, both competitive and non-competitive. SATURATION AND INHIBITION OF THE GLYCEROL TRANSPORT SYSTEM The photometric method of measuring erythrocyte volume changes 5 was used in a study of the diffusion of glycerol. To investigate the rate of permeation of a substance, the substance is dissolved in isotonic saline (the final solution being hypertonic). Known volumes of red cell suspension and hypertonic solution are mixed. As a result the environment of the red cells is hypertonic, there is an osmotic contraction of the cell volume-and the light transmitted by the solution decreases. if the added substance is a permeant, the cells subsequently increase in volume and the optical density of the solution decreases. It is found that this decrease occurs in an exponential manner with time. It was considered that the half-time of attain- ment of the final volume could perhaps be used as a measure of the rate of volume change and hence as a measure of the permeability of the cells to the added solute.In fact, if the permeant is assumed to diffuse passively through the membrane, in accordance with Fick’s law, then the permeability constant P being given by dG/’dt 1 PA(Ce - Cj), it can be shown that 11-12- - where t4 is the half-time for attainment of the equilibrium volume, Vis the equili- brium volume of the red cell, A the surface area of the cell, Se the tonicity of the external medium and Ce the concentration of permeant added, and Ci in eqn.(1) is the internal concentration of permeant at time t. The following approxima- tions were made in the derivative of eqn. (2) : that the rate of movement of water is far greater than that of the permeants studied, i.e. the cells are always in osmotic equilibrium with the environment ; that the cells behave as perfect osmometers ; that the cell surface area remains constant throughout ; the purely numerical approximation that (In 2 - +)/Se N 0.693 N 0.7 for eqn. (2n). In fig. 1 the effect on the observed t g values of altering the concentration of permeant added is shown for a number of different substances penetrating into horse, rabbit and human red blood cells. All the fg values are expressed as ratios of t) at 0.1 M permeant. Curve A is a plot of t4 against Ce, as given by eqn.(2a). It will be seen that for both triethylene glycol and 1 : 2 : 4-trihydroxybutanol, the plot of t4 against Ce is in fair accordance with the approximate law of eqn. (2a). However, the plots for glycerol deviate markedly and characteristically from those predicted. In fact, t: is greater than would be expected and the discrepancy appears to increase with Ce. Eqn. (2) suggests that, in these cases, the permeability “ constant ” apparently decreases with concentration of permeant. The permeability constant as defined in eqn. (I) is the ability of the membrane to allow inward or outward movement of permeant per unit concentration insidc or outside the membrane respectively. Thus PCe gives the total number .of moles entering the membrane per unit area and we can term this product of PC, the inward242 RED CELL PERMEABILITY flux of the permeant through the membrane. If the fiux is directly proportional to the concentration, we have and we recover P = const.For permeants not obeying the equation derived from Fick's law, P is no longer constant and hence the flux through the membrane is not directly proportional to the concentration. Let us assume that the flux obeys a limiting law of the Michaelis-Menten type, i.e., PCe= const. Ce, 6- 5 - 4 - Here apparent P is the permeability " constant " derived from the observed tt value by the application of eqn. (2). Fmax is the limiting (saturation) flux. ' E I L Y 9 c ,.:ex/ 01 I I f I . 0 . 2 0.6 1.0 M pcrmcant FIG.1.-Variation of Half- equilibrium Time with Per- meant Concentration. A theoretical curve - simple diffusion, B penetration of 1 : 2 : 4-tri- hydroxybutanol, C penetration of triethylene glycol, D penetration of gIycerol into human erythrocytes, E penetration of glycerol into rabbit erythrocytes, F theoretical curve-satura- tion behaviour (K, = 1 M). We can write Fick's law in the more general form, dG/dt = A(flux inward - flux outward), and substitution of the saturation condition (3) gives ci ). dG d t = F , a x A (KmYCe K m f Ci Analysis of the cell volume changes with time under this condition gives making the same approximations and assumptions as for (2a) above. Curve Fin fig. 1 shows a plot of t+ against Ce derived from this equation with Km = 1 M.It would appear that eqn. (3) to (6) adequately describe the observed variation of t ) with Ce for glycerol, suggesting that the movements of this substance obey a law derived from consideration of saturat on phenomena, rather than from simple diffusion theory.W. D . STEIN AND J . F . DANIELLI 243 A similar conclusion for the diffusion of glucose into certain mammalian erythrocytes has been reached by Wilbrandt,6 by Lefevre 7 and by Widdas * using slightly different methods. These authors consider that the law (3) and the signi- ficance of K, can be explained on the basis of saturation of an enzymic or carrier system, R,, being an inverse measure of the affinity of the enzyme with the substrate, by analogy with the results of classical enzyme kinetics.It is, however, possible to derive the saturation condition of eqn. (3) and ( 5 ) in a way which does not involve consideration of enzymic activity. Let it be assumed that there exist specially differentiated portions of the membrane which act as pathways for the diffusion of permeant-these pathways will be termed pores. There are n such pores per unit area of cell surface. Let it be further assumed that if a pore is occupied by a permeant molecule, no further permeant molecules can enter the pore until this molecule leaves. At a given time let there be x such pores per unit area “ occupied” by permeant molecules. Then n - x will be un- occupied. The rate of movement of molecules per unit area out of the pores will be propor- tional to the number of pores occupied, i.e.transfer out of the membrane = Kout. x molesiunit area sec. The rate of entry of molecules into the pore system is similarly given by and at the steady state, therefore, transfer in = Kin. (n - X) Ce Kout. x = Kin. (n - x) Ce or Kout c e = - . - K P , Kin. n - x X (7) where Kp will be termed the ‘‘ pore constant ” for the permeant. and Vmax = nzn where rn is a constant. with the same proportionality factor m, hence Solving eqn. (7) and (8) we obtain an equation of the same form as (3). The significance of Kp will be discussed later. It will first be necessary to discuss some observations on the inhibition of glycerol transport. Since the work of Jacobs and Corson,9 copper in small traces has been known to inhibit the transport of glycerol in certain species of erythrocyte.“ Narcotic ” substances have a similar action. Lefevre extended this work and showed that mercury and some organic protein reagents in small concentrations can also inhibit glycerol transport. In the present study the effect of copper and of “ narcotics ” on inhibiting transport has been studied quantitatively. It has also been found that high concentrations of certain glycols have an inhibiting effect on this trans- port : this effect does not appear to be related to the oil-water partition coefficients of these substances, i.e. it is not a non-specific narcosis. It will be shown in what follows that the narcotics and the glycols act as competitive inhibitors while copper inhibits in a non-competitive manner. It will be convenient to return to the model of the transport system considered in eqn.(7) to (9). Let it be further considered that in the presence of an inhibitor certain of the pores are “ occupied ” and let 1 be the number of pores so occupied per unit area of membrane surface. Then n - x - I will now be the number of unoccupied pores and eqn. (7) becomes When the system is operating at maximum velocity all the pores are occupied At any lower velocity V, V = mx V/Vmax= xln. (8) (9) V = Vmax CeI(Kp + Ce>, C,(n - x - I)]x = Kp, Kp being unchanged.244 RED CELL PERMEABILITY NON-COMPETITIVE INHIBITION For a non-competitive inhibitor, the number I will depend only on the concentration of inhibitor molecules in the solution and will not be affected by C,, i.e. permeant molecules cannot displace inhibitor molecules from the inhibitor- occupied pores.This defines non-competitive inhibition. Hence when all free pores are occupied by permeant molecules, there will be only n - I such pores per unit area. The maximum velocity of eqn. (8) is thus reduced and (8) becomes (12) where the suffix i refers to inhibited velocities. Solving eqn. (1 1) and (12) we obtain c e fi = Vmax i - Kp + Ce' but also where Vmax is the maximum velocity in the absence of inhibitor, whence Substituting this expression in the generalized Fick's law of eqn. (4), it can be shown that It will be convenient, in what follows, to ignore the variations with C, of the term within the square brackets. Over the range of concentrations studied (0.1 to 0.4 M) these variations do not affect t3 significantly in comparison with the effects of the inhibitor.Thus we obtain the approximate equations from (6) and also from (13) and Here t g max is the value of t3 at saturation, i.e. at Ce = Cmax. Eqn. (13a) shows that at constant inhibitor concentration, the graph of tti against Ce (the external concentration of permeant) should be a straight line, with intercept at t+; = 0 of - Kp. Eqn. (10) is a special case of (1 3a) for which I = 0. Thus plots of tBi against C, at constant I , for various values of the constant I , should give a series of straight lines, all intersecting on the tAi = 0 axis at C, = - Kp ; the straight lines being of varying slopes, increasing with increasing I. Hence, also the intercepts at C, = 0 should increase with increase of I.Fig. 2 shows a series of such plots for the inhibition of glycerol transport in the human erythrocyte by copper. The plots are in good accordance with theory suggesting the non- competitive nature of the copper inhibition. Eqn. (14a) shows that at constant Ce, the plot of l/tgi against I should be a straight line of negative slope. If it be assumed that the number of pores occupied by inhibitor per unit area be directly proportional to the copper concentration added to the suspension, then a plot of l / t i i against concentration of copper addedW. D. STEIN AND J. F. DANIELLI 245 may be expected to show a similar behaviour. Fig. 3 shows such a plot for the copper inhibition of human red cells. A good straight line of the required negative slope was obtained, again suggesting that the assumption of non-competitive binding is valid.I 5 0 I00 v 0 .I f L 50 I I 0 0 . 2 0 . 4 0.6 0.8 M qlycerol FIG. 2.-Copper inhibition of glycerol transport with varying concentrations of glycerol. I00 - + '/2 x 10 h Y n 0.2 x ~ ~ 4 ~ ~ u 1.~0 ' OI.4 ' FIG. 3.-Copper inhibition of glycerol transport at constant glycerol concentration (0.1 M). (In passing, it may be pointed out that eqn. (13) and (14) contain the expression n, the number of pores per unit area. From eqn. (14) it will be observed that the intercept of l/tai against I, at l/tti = 0 gives I = n (or, at maximum inhibition all the pores are occupied by inhibitor). Thus if it were possible to measure Z directly, the very useful quantity n could be found). COMPETITIVE INHIBITION For competitive inhibition, eqn.(1 1) still holds but the number of pores occupied by inhibitor is a function of both the inhibitor and the permeant concentrations. This is because permeant molecules are able to displace inhibitor from the occupied pores, in a competitive fashion.246 RED CELL PERMEABILITY By exact analogy with (1 1) we have (n - x - I)CI 1 = K I , where CI is the concentration of inhibitor in the solution and KI is the pore con- stant of the inhibitor. For competitive inhibition, all the pores can be available for permeant " OCCU- pation ", at sufficiently high permeant concentration, so that Vmax inhibited = Vm, uninhibited, and we have Wvmax == xln (1 6) On solving eqn. (1 l), (15) and (16) we obtain and as before, substituting in eqn.(4) we have or where the variation of the term in square brackets is again neglected. Eqn. (17a) shows that at constant inhibitor concentration Ci, the plot of t+i against Ce should be a straight line. Tn fact, such plots for different Ci should give a series of straight lines of the same slope t+max/Cmax, and of the same slope as the uninhibited plot (since (17a) reduces to (10) for Ci = 0). The intercept on the ti, axis should increase with Ci as, of course, should the intercept on the C, = 0 axis. Fig. 4 shows such a series of curves for the inhibition of glycerol transport in the human erythrocyte by the glycol 1 : 2-dihydroxypropane. The agreement with theory is good, suggesting that this glycol acts by inhibiting glycerol transport in a competitive manner.Furthermore, from eqn. (17a) it will be seen that a plot of t+i against Cj at constant Ce should give a straight line the slope of which depends on the ratio K p / K ~ , the intercept on the C, = 0 axis being constant. Fig. 5 shows such a plot for the action of the glycols 1 : 2-dihydroxypropane and ethylene glycol on glycerol transport in the rabbit. The agreement with theory is again good. Eqn. (17a) shows that KI for ethylene glycol is lower than for 1 : 2-dihydroxypropane. These experiments then also support the view that the inhibition by glycols is competitive. 1 compound glycerol glycerol ethylene glycol propylene glycol ethanol butanol urethane phenol 2 cell species rabbit human human rabbit human human human human TABLE 1 3 Kp and KI M 0.5 0-7 0.1 0.1 0.1 0.01 5 0.01 5 0.003 4 olive oil/water partition coefficient 1 x 10-4 1 x 10-4 0.5 x 10-3 - 6 x 10-3 2 x 10-2 13 x 10-2 13 x 10-2 10 5 product of 3and4 5 x 10-5 7 x 10-5 5 x 10-5 20 x 10-4 20 x 10-4 20 x 10-4 300 x 10-4 6 x 10-4W.D . STEIN A N D J . F . DANIELLI 4 0 - 20 - 247 .I I I - 0-5 M c a 0 . 2 5 M O M I I I I 1 1- 1 / I M qlycol I I I ,2 0 . 4 M qlycerol FIG. 4.-Propylene glycol inhibition of glycerol transport ; glycerol concentration varied. M qlycol FIG. 5.-Inhibition by glycols ; concentration of glycerol fixed at 0.1 M248 RED CELL PERMEABILITY A comparison of fig. 2 and 4 shows most clearly the difference between the behaviour of copper and of 1 : 2-dihydroxypropane, a difference presumably due to their different modes of inhibition of the glycerol transport system.Fig. 6 and 7 show the apparently competitive nature of the inhibition by the narcotics ethanol and urethane. From fig. 1 to 7 it is possible to obtain values of Kp and KI for the various permeants and inhibitors. These values are recorded in table 1 . urethane 40 20\ I L I 0 . 1 I 0 . 2 0 - 3 0 . 4 FIG. 7.-Inhibition by urethane (" narcosis "). Although the basic eqn. (6), (13) and (17), which appear to describe adequately the phenomena of saturation and inhibition, were derived from the " occupied pore " model of the pathway system, they do not necessarily suggest that such a model is in fact operative. It is very likely that from any model comprising a system capable of saturation, a similar set of equations could be derived.In particular, it is not necessary to assume that each pore can contain only a single permeant molecule. DISCUSSION In the introduction it was postulated that, although the basic structure of the plasma membrane is lipoid, there are special regions in which a hydrogen-bonding structure extends through the thickness of the membrane, constituting polar pores. It was suggested that polar molecules would penetrate rapidly through such a structure, provided the stereochemistry of the penetrant corresponded closely to that of the hydrogen-bonding structure. It was also indicated that molecules capable of forming hydrogen bonds would compete for the hydrogen-bonding sites in the membrane, and so reduce the rate of penetration of a penetrant. There are, however, serious limitations to the amount of information which can be obtained from kinetic studies.It is highly desirable that we should have precise information about the molecular structure of the parts of plasma membranes which are responsible for active transport and facilitated diffusion. Kinetic studies can to some extent guide us as to what to expect. Facilitated diffusion appears to require a specific hydrogen-bonding space. This could probably be provided by protein lamellae arranged as in fig. 8. But although protein molecules could certainly provide the degree of specificity which is required, we cannot exclude the possibility that polysaccharides and nucleic acids may also be involved.W. D. STEIN A N D J . F . DANIELLI 249 It is of interest to consider the nature of the rate-limiting processes in permeation through a pore such as that shown in fig.8. Entry into, and exit from, pores may be rate limiting. So also may diffusion through a pore. In all these three cases, in addition to stereochemical factors, it is probable that the hydrogen- bonding sites which can be occupied by a penetrant will normally be occupied by P o l o r i i d e N on-p o lo r FIG. 8.-(u) diagrammatic cross-section of a protein molecule consisting of four polypeptide Iamellae. (b) diagrammatic cross-section of a pore composed by two polypeptide lamellae in a membrane. The basic structure of the membrane is a bimolecular lipoid leaflet stabilized by adsorbed protein monolayers. water or a displaceable element of the membrane. If alternative hydrogen- bonding substances, such as glycol, or urea, are present, they will also compete for these sites, If they compete effectively with water, they will also make passage through a pore more difficult for a penetrant, and in fact a penetrant may readily inhibit its own passage through such a pore.THE ACTION OF NARCOTICS Previous workers have shown that alcohols and urethanes inhibit the facilitated diffusion of glycerol. The extent of this inhibition has been shown to be related to the partition coefficient of the inhibitor in a manner similar to the action of the fat-soluble narcotics, to which class of compounds these inhibitors belong. The present study has confirmed the earlier work, but the further finding that these narcotics act as competitive inhibitors enables some suggestions to be made as to their possible mode of action.On the model of the pore-system depicted in fig. 8, glycerol is able to form hydro- gen bonds with the hydrophilic groups forming the " lumen " of the pore. The strength of these bonds will depend on the character of these binding groups and this in turn will depend on the molecular environment of the binding groups. Fat-soluble molecules will dissolve in hydrophobic portions of the membrane, and by so doing may alter the molecular environment and hence the hydrophilic character of the adjacent hydrogen-bonding groups. In this way, inhibition of glycerol transport can occur. Since, on this hypothesis, the narcotic does not interfere irreversibly with the facilitation system, a sufficient excess of glycerol molecules may be expected to reverse the inhibition and hence such inhibition is competitive-at a sufficiently high level of glycerol concentration, all pores will be able to act as hydrogen-bonding systems.Alternatively, it may be expected that the presence of a lipophobic glycerol molecule in the interior of the pore would depress the solubility of lipophilic groups in the regions adjacent to the pore-the permeant and the narcotic mutually250 RED CELL PERMEABILITY depress their respective solubilities in the pore region. The character of the pore is thus determined by whether permeant or narcotic is the first to enter the pore region. Since both permeant and narcotic have high turnover numbers through the membrane, " occupation " of the pore region is a reversible phenomenon and the competitive nature of narcotic action follows. As will be seen from table 1, ethylene glycol has as great an inhibitory effect on glycerol transport as has ethanol, although the partition coefficient of the glycol is some thirty-fold lower.It may thus be inferred that ethylene glycol inhibits by a direct action on the hydrogen-bonding system, presumably being able to bond with the same groups as the glycerol molecule. Such an action would also result in competitive inhibition. Thus the hypothesis of a specific hydrogen-bonding pore extending through the otherwise lipoidal interior of the membrane explains the observed competitive inhibitions. The protein nature of the pore is consistent with the non-competitive inhibition by copper and the action of protein reagents.PORE THEORY AND ENZYMIC REACTION THEORY Studies on the facilitated diffusion of glucose into human red cells have led to hypotheses based on " diffusible substrate-carrier complexes " to explain this facilitation An enhanced rate of diffusion results from a presumably enzymic reaction by which the glucose molecule becomes converted into a product capable of penetrating through the lipoidal cell membrane. The kinetics of saturation and inhibition of the transport system are explained as deriving from saturation and inhibition of the postulated enzymic reaction. The enzymic reaction is reversible, or must be reversed, in that the lipophobic character of the glucose molecule is regained when it reaches the interior of the cell.In the present study, it has not yet been necessary to postulate such an enzymic system for the penetration of glycerol. A purely physical hypothesis involving a limited number of hydrogen-bonding pores has been adequate to account for the experimental observations of saturation and inhibition, and it has been suggested that the hypothesis is adequate to explain the non-competitive inhibition by copper, and the partition coefficient-dependent competitive inhibition by the narcotics, as well as the inhibition by the glycols. A possible advantage of this model over an enzymic system is that it does not require a subsidiary hypothesis explaining the action of the enzyme. An enhanced rate of diffusion follows directly from the local lowering of membrane resistance afforded by a hydrogen-bonding pore. A further difficulty encountered by a " diffusible carrier " hypothesis is that if the carrier is able to reduce the lipophobic character of the glucose substrate, the carrier must itself be lipophobic for reaction to occur. It is thus necessary to explain how the diffusion of free carrier occurs. The low activation energy for glycerol facilitations accords with diffusion in a lipophobic environment and the recorded values are lower than those of esterifica- tion reactions in general. The competitive action of the narcotics is readily explained on the alternative hydrogen-bonding pore hypothesis. In terms of a diffusible carrier mechanism, narcotics could be expected to affect the enzymic reaction involved-but by a denaturation. It is not. clear how such a denaturation could be reversed by excess glycerol. To sum up, the results we have obtained from studies of competition are fully compatible with the view that facilitated diffusion through plasma membranes occurs through a hydrogen-bonding component which extends through the thick- ness of the membrane, But kinetic studies can never prove this is the case. The urgent requirement in this field is the isolation of the effective membrane components.251 One of us (W. D. S.) is indebted to the University of Witwatersrand for a research W. D . STEIN AND J . F. DANIELLI scholarship. 1 Davson and Danielli, The Permeability of Natural Membranes (Cambridge University 2 DanieUi, Colston Symp., 1954, 1, 1 ; Symp. SOC. Expt. Biol., 1954, 7, 502. 3 Danielli, Symp. SOC. Expt. Biol., 1952, 6, 1. 4 Bowyer, Nature, 1954, 174, 355. 5 Orskov, Biochem. Z., 1935,279, 241. 6 Wilbrandt, Symp. SOC. Expt. Biol., 1954, 8, 136. 7 Lefevre, Symp. SOC. Expt. Biol., 1954, 8, 118. 8 Widdas, J. Physiol., 1953, 120, 23 p. 9 Jacobs and Corson, Biol. Bull., 1934, 67, 325. Press, 1943).

ISSN:0366-9033    

DOI:10.1039/DF9562100238

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

W. D. STEIN AND J . F. DANIELLI 251 THE FACILITATED TRANSFER OF GLUCOSE AND RELATED COMPOUNDS ACROSS THE ERYTHRO- CYTE MEMBRANE BY FREDA BOWYER * and W. F. WIDDAS Physiology Department, King’s College, Strand, London, W.C. 2 Received 16th January, 1956 The penetration of sugars into human red cells is not a simple diffusion and probably requires as a first step the formation of a complex between the sugar and some membrane component. The kinetics of the process account for the different rates of penetration for ketose and aldose sugars and explain competition between the sugars. Competitive and irreversible inhibitors of hexose transfer have been described. The kinetics of the inhibitory reaction by dinitrofluorobenzene (DNFB) are complex. Although this special mechanism of hexose transfer is conveniently studied in human red cells, comparative studies in other species and tissues suggest that it is one of wide biological significance.Little is known about the mechanism of transfer on a molecular level but several possibilities in keeping with the observed kinetics are described. The penetration of sugars into human red cells has long been regarded as one of special interest. Kozawa 1 showed that related sugars penetrated at different rates and Ege 2 observed that the penetration rate of glucose decreased at higher concentrations. Quantitative measurements during the last ten years 3-8 have confirmed this fall in transfer rate at higher concentrations and have shown mutual interference between sugars. This evidence, coupled with the effect of inhibitors, has given rise to the view that the sugar first combines with some component of the cell membrane and that transfer is effected by a carrier system or a process of facilitated transfer.4.9, 8 KINETICS Attempts to derive kinetics appropriate to the process visualized have been made by LeFevre,4 LeFevre and LeFevre,6 Wilbrandt and Rosenberg,lo and Widdas 11912.In their simplest form the kinetics of the latter three authors reduce to an expression in which transfer across the red cell membrane is proportional to the difference in the fraction of membrane components saturated with hexose a t the inner and outer interfaces. The fraction of components saturated at the * In receipt of a personal grant from the Medical Research Council.252 FACILITATED TRANSFER interface is considered to be related to concentration according to the Michaelis- Menten equation or Langmuir adsorption isotherm.Thus C C' rate of transfer = K - - -} S + S ' {c++ C ' + + ' where C and C' are the concentrations outside and inside the cell respectively and 4 is the equilibrium constant of the sugar complexing reaction. This equation has two useful approximations. When transfer = (K/+) (C - C') S - t S ' k ( C - C'), an equation of the diffusion type. When c>+aY, transfer = K+{:' - $} (3) s -+ S' = k' {k, - k}. (34 This last approximation although deduced on theoretical grounds 11.10 has con- siderable application to practical studies 1 2 ~ 8 ~ 13 in the analysis of results of glucose penetration experiments. The ketose sugars sorbose and fructose show penetrations which are better fitted by eqn.(2), that is an equation of the diffusion type. It is clearly important to be able to distinguish between a process in which a membrane component is involved but having diffusion-type kinetic equations and a process of simple diffusion. For penetration of red cells by sorbose and fructose, simple diffusion would appear to be excluded by the following evidence : (1) Red cells are impermeable to rn-inositol 1495S7s8. This substance is an isomer of glucose with a ring structure and steric arrangements of hyroxyl (-OH) groups similar to glucose but with the exception that it is homo-cyclic and has no reducing group. However, if pores were available in the red cell membrane large enough to allow sorbose, fructose or glucose to enter by a process of simple diffusion it seems unlikely that in-inositol could be excluded so effectively.(2) Inhibitors which retard glucose penetration such as p-chloromercuri- benzoate (PCMB)4 and dinitrofluorobenzene (DNFB) 15 also inhibit the penetration of ketose sugars. (3) The presence of glucose in the medium causes sorbose penetration to be retarded. This effect is increased with higher glucose concentrations in a manner which can be accounted for on a basis of competitive inhibition 7.8. (4) In a comparative study of human and certain foetal bloods 13 there is always a close correlation between the rate of penetration of glucose and that of sorbose. There is thus strong evidence for considering that the ketose sugars use the same transfer mechanism as glucose 5 3 8 and that the fit of experimental results to equa- tions of the diffusion type is due to a high value of the equilibrium constant of the complex formed between the ketose sugars and the membrane component.EQUILIBRIUM CONSTANTS Accepting the assumption that the hexoses all have the same transfer mechanism their different rates of penetration can be explained on the different values of the equilibrium constants of the complex formed with the membrane component. T.he qualitative differences heve been confirmed by experimental determinations of the equilibrium constants 7 1617.F . BOWYER AND W . F . WIDDAS 253 THE MEMBRANE CONSTANT The equation for hexose transfer contains a second constant, K in eqn. (l), which depends on the cell volume, the number of glucose adsorbing sites in the membrane and the speed of translocation of adsorbed glucose through the mem- brane.Quantitatively, the constant represents the maximum rate of transfer which would be possible if the membrane component concerned was fully saturated with glucose at one interface and unsaturated at the other. This ideal cannot be achieved in experiments involving inward transfer of glucose as accumulationwithin the cell rapidly occurs. In experiments on the exit of glucose, however, it is possible to approach this ideal since cells equilibrated in a high concentration of glucose can be packed by centrifugation and a small volume of packed cells added to a large volume of glucose-free medium. Under these circumstances eqn.(1) takes the form rate of transfer = -K (4) s-S’ which approximates to --K when C’ is large. COMPARATIVE STUDIES The earliest comparative study of hexose permeability of erythrocytes 1 suggested that the process is peculiar to man and other primates. It has more recently been shown that cells from foetal blood of a number of mammals (sheep, guinea-pig, etc.) have a glucose permeability of the same order of magnitude and type as human erythrocytes 18% 13. In addition there have been experiments suggesting a much wider biological significance, e.g., in intestinal absorption 19.20 and the penetration of hexoses into heart muscle.21 In regard to the penetration of sugars into muscle cells the problem has a special interest in that several authors have adduced evidence that insulin has the effect of increasing the cell permeability to glucose and related sugars.This has been reviewed by Stadie.22 Further evidence is given by Park et al.23124 but this action of insulin has been doubted.25 Insulin has no effect on the hexose perme- ability of erythrocytes.3 RELATION OF GLUCOSE PERMEABILITY TO THAT OF POLYHYDRIC ALCOHOLS There are great species differences in red cell permeability to polyhydric alcohols.26s 27 Table 1 shows the permeability of adult and foetal red cells of human, guinea-pig and sheep to glycerol, erythritol and glucose. There is no correspon- dence between the permeability to glycerol and glucose nor is there any detectable inhibition of glucose penetration by glycerol or vice versa so that different mechan- isms of transfer may be postulated.TABLE 1 permeability to glycerol hexose erythritol type of erythrocyte human, adult + + + foetal + + + guinea-pig, adult + 0 0 foetal + + + sheep, adult 0 0 0 foetal 0 + + + indicates permeability typical of cells showing rapid penetration of the substance 0 indicates slow or zero permeability Studies with meso-erythritol28 and pentaerythritol show transfer rates which parallel those of hexose but not glycerol. That these alcohols use the hexose system254 FACILITATED TRANSFER (in those cells showing fast hexose transfer) is also confirmed by glucose com- petition and inhibitor studies. On the other hand, 1 : 2 : 4-trihydroxybutane and dihydroxyacetone do not appear to use the hexose transfer system. These observations modify the view 4 that only molecules with aldehydic or ketonic groups can use the hexose transfer system.They also suggest that there is a transition between molecules with three and four -OH groups. The different values of the equilibrium constant for compounds using the hexose system may be a function of the number and steric relationships of the -OH groups involved in complex formation. INHIBITOR STUDIES The hexose transfer system is sensitive to the inhibitors p-chloromercuribenzoate (PCMB), mercuric chloride, 2 : 4-dinitrofluorobenzene (DNFB), 2 : 4-dinitro- chlorobenzene (DNCB), 2 : 4-dinitrobromobenzene (DNBB) and polyphloretin phosphate. PCMB and mercuric chloride react readily with sulphydryl (-SH) groups and less readily with amino (-NH2) groups.29 The reaction is not readily reversible but can be reversed by the addition of, e.g., glutathione.Red cells partially inhibited by the mercurial inhibitors show a restored glucose permeability after addition of glutathione. The inhibition produced by PCMB and mercuric chloride has been found to decrease with time on standing. It is suggested that the inhibitor is at first attached to the membrane but either slowly penetrates the cells and possibly combines with intracellular substances, e.g., glutathione or combines with similar substances released into the medium by a small amount of haemolysis. DNFB and DNCB react irreversibly with -SH, -NH2 and phenolic hydroxyl. The inhibition of glucose transfer in red cells by these inhibitors was not reduced by repeated washings with alcoholic saline buffer.Experiments with a diazonium hydroxide, which reacts with tyrosine, showed no inhibition of glucose transfer.15 In order to find out whether amino groups are essential, preliminary experiments have been carried out using nitrous acid which de-aminates and acetic anhydride which acetylates -NH2 groups. So far no inhibition of glucose transfer has been observed though under the mild conditions employed it is possible that de-amination and acetylation were not complete. If it is -SH groups which are blocked they cannot be of the readily available type characterized by Barron and Singer,30 as LeFevre4 showed that glucose transfer was not inhibited by copper, alloxan, mapharsen, iodoacetate or arsenite. They could, however, correspond to the intermediate or not readily available type of -SH groups which are blocked by p-chloromercuribenzoate and mercuric chloride.MODE OF ACTION OF INHIBITORS An inhibitor could produce its effect either by combining with (i) the membrane site required by glucose, (ii) part of the membrane distinct from the glucose adsorbing site but which (iii) the movement of hexose through the membrane, (iv) the equilibrium constant of the complex formed by glucose and the site. Suggestions (i), (ii) and (iii) would cause a change in K, the membrane constant (see eqn. (I)). Postulates (i) and (ii) would, in effect, cause a reduction in the number of glucose adsorbing sites available in the membrane while postulate (iii) would affect the speed of translocation of adsorbed glucose through the membrane.Suggestion (iv) would cause a change in the equilibrium constant 4 of glucose and other hexoses with the site. prevented the combination of glucose with the site ; or by affectingF. BOWYER AND W. F. WIDDAS 255 For glucose : k' = K4, (see eqn. (3) and (3a)), while for sorbose : k = K/$ (see eqn. (2) and (2a)). Thus a decrease in 4, the equilibrium constant, would cause a decrease in the transfer of glucose but an increase in the transfer of sorbose. Experiments on the inhibition of glucose and sorbose by PCMB show a change of penetration in the same direction in both cases (i.e., a decrease) in the presence of inhibitor. Determination of 4 for glucose, by the competition with sorbose method, shows no change in 4 in the presence or absence of p-chloromercuri- benzoate.Experiments on the exit of glucose from red cells in the presence and absence of the inhibitors indicate that it is a change in K which causes the inhibition. So far, however, it is not possible to say whether this is due to postulate (i), (ii) or (iii). t m i n FIG. 1-Time course of inhibition produced in red cells incubated with glucoseIandlDNFB at 20" C. Glucose exit at 37" C. K in isotonic units. Concentrations of DNFB: 0 0.93 mM, A 1.4 mM, X 1.86 mM and 0 3-7 mM. The reaction of the red cell with phloretin is reversible and the experiments of LeFevre 17 show that it is a competitive inhibition, i.e. a plot of l/rate against inhibitor concentration is linear and the plot of rate against glucose concentra- tion at fixed inhibitor concentration is likewise linear.As the effect is on K, the membrane constant, postulate (iv) is ruled out. Also, as the inhibition is competitive postulate (iii) would appear to be eliminated. LeFevre considers that the phloretin acts by direct competition with the sugars for the membrane site, i.e. hypothesis (i). But the evidence does not exclude postulate (ii) if there can be two mutually exclusive sites for glucose and inhibitor. As the reaction of PCMB and mercuric chloride is only slowly reversible, these inhibitions do not behave in a competitive manner with glucose and a plot of l/rate against inhibitor concentration is not linear.17 A study of the reaction with the non-competitive inhibitor DNFB indicates that a complex process is involved.KINETICS OF INHIBITION BY DNFB The membrane constant from glucose exit experiments has been determined in cells equilibrated in glucose solution and incubated with various concentrations of inhibitor. The decrease of the membrane constant with time of incubation is shown in fig. 1.256 FACILITATED TRANSFER Taking K to be proportional to the number of glucose adsorbing sites and noting that the concentration of DNFB is in excess, a plot of log K against time was examined to see if the reaction was pseudo-unimolecular. This was not SO but a plot of 1/K against time is reasonably linear as is shown in fig. 2. This result would indicate that the inhibitory reaction is second order in terms of K. The slopes of the lines in fig. 2 are proportional to the velocity constant of the inhibitory reaction and when compared with the concentration of inhibitor in a log, log plot would appear to depend on the square of inhibitor concentration.I , ' , , 50 100 150 200 2 5 0 300 3 5 0 4 0 0 450 FIG. 2.-The same data as fig. 1 plotted to show the approximately linear relation of 1/K and time. The slopes of the lines kl may be regarded as proportional to the velocity constants of the in- hibitory reaction at the different DNFB concentrations. Pts 0, A, X, and 0 have the same significance as in fig. 1. The results thus suggest that the rate of the inhibitory reaction depends on both the square of the concentration of inhibitor and the square of the concentration of remaining glucose adsorbing sites. In terms of a simple mass action a fourth order reaction of this type seems most improbable.However, we are concerned with a reaction occurring in or near the cell membrane and if one or other component was associated or concentrated in the membrane a complex overall reaction order might result. That the inhibitory reaction is complex is also suggested by the high temperature coefficient of the reaction (Qlo = 7). Partially inhibited human red cells behave very much like adult guinea-pig red cells which have only a slow glucose penetration. The experiments of Morgan et aZ.31 with adult rabbit red cells (which also show slow glucose penetration) indicate that $ in this case is similar to $ for human red cells. This means that the large difference in permeability rate between rabbit and human red cells is due to a lower value for K, just as the difference between inhibited and uninhibited cells is due to a reduction in this constant.Taking the comparative studies with the in- hibitor evidence it would be sufficient to postulate that the main differences in permeability are due to variations in the number of available sites. MOLECULAR MECHANISMS The special features of erythrocyte permeability to hexoses as described in previous sections do not differentiate between several possible mechanisms ofF. BOWYER AND W. F. W I D D A S 257 transfer. The most we can say is that any hypothesis must give rise to rate-deter- mining kinetics closely approximating to those described. Reviewing all the experi- mental evidence it is difficult to escape the conclusion that the formation of a complex between the hexose and a component of the membrane is an essential first step.An important step in elucidating the molecular mechanism would be to deter- mine the nature of the membrane componentwithwhich the hexose forms a complex, but since only a short-lived complex need be presumed its identification presents a difficult problem. Subsequent steps in the transfer mechanism are more hypo- thetical since there is no finality of opinion as to the detailed membrane structure or the physico-chemical reactions most likely to occur. It might be appropriate, however, to conclude by describing some hypothetical mechanisms which would yield the same kinetics in order to consider their feasibility from a physico-chemical point of view.A. MOBILE MEMBRANE COMPONENT OR MEMBRANE CARRIER. This type of mechanism involves the transfer of the complex of hexose and the membrane com- ponent through the membrane to the opposite interface where the hexose dissociates. The same course of events can occur in the opposite direction andunsaturated carriers may also diffuse backwards and forwards. Ussing 32 has applied the term " ferryboats " to such carriers. The rate-determining steps of this system may be such as to give the required kinetics.33. 34 The chemical identity of such a carrier might be : (i) lipid, (ii) lipoprotein, (iii) protein or (iv) contractile protein. The possibility that a protein by coiling and uncoiling might pass backwards and forwards through a membrane to give special permeability was suggested by Goldacre and Lorch.35 The extended portion of the protein chain would have to be capable of forming complexes with hexoses.Further understanding of the possibility of this type of mechanism might be obtained from physico-chemical experiments on the mobility of proteins and lipids in biological membranes, A slight modification on the mobile membrane type of transfer has been suggested.36 This involves the rotation of the section of the membrane to which the hexose is attached, followed by its release in the interior of the cell. A further possibility is a sub-microscopic process akin to pinocytosis in which part of the surface forms an inclusion. In contrast to the above hypothesis, it has been suggested that after the initial adsorption of glucose on the membrane site the glucose only travels through the membrane.37 This could be accomplished by postulating a series of glucose adsorbing sites along a polar region from the out- side to the inside of the membrane.The transfer of glucose could then be con- sidered as a chain or " creep '' process along these adsorption sites. The polar region could rather loosely be termed a " polar pore ". A rather similar mechanism has been postulated by Hodgkin and Keynes 38 to explain a discrepancy which they find between measured and expected potassium flux in Sepia and Loligo axons. The transfer of hexose could take place symmetrically both to the inside and to the outside and the rate-determining steps of this system could also give the required kinetics.B. NON-MOBILE MEMBRANE COMPONENT. 1 Kozawa, Biochem. Z., 1914, 60, 231. 2 Ege, Thesis (Copenhagen, 1919), cited by Bang and Orskov. J. Clin. Invest, 1937, 16, 279. 3 Wilbrandt, Guensberg and Lauener, Hefv. physiol. Acta, 1947, 5, C20. 4 LeFevre, J. Gen. Physiol., 1948, 31, 505. 5 LeFevre and Davies, J . Gen. Physiol., 1951, 34, 515. 6 LeFevre and LeFevre, J. Gen. Physiol., 1952, 35, 891. 7 Widdas, J. Physiol., 1953, 120, 23 P. 9 Wilbrandt and Rosenberg, Int. Rev. Cytol., 1952, 1, 65. 10 Wilbrandt and Rosenberg, Helv. physiol. Acta., 1951, 9, C 86. 11 Widdas, J. Physiol., 1951, 115, 36 P. 12 Widdas, J. Physiol., 1952, 118, 23. 8 Widdas, J. Physiol., 1954, 125, 163. 13 Widdas, J. Physiol., 1955, 127, 318. I258 PERMEATION MECHANISMS 14 Hedin, Pfiig. Arch. ges. Physiol., 1897, 68, 229. 15 Bowyer, Native, 1954, 174, 355. 17 LeFevre, Symp. SOC. Expt. Biol., 1954, 8, 11 8. 18 Widdas, Abstr. XZX Int. Physiol. Congr., 1953, 885. 19 Fisher and Parsons, f. Physiol., 1953, 119, 210. 20 Fisher and Parsons, J. Physiol., 1953, 119, 224. 21 Fisher and Lindsay, J. Physiol., 1954,124,20 P. 22 Stadie, Physiol. Rev., 1954, 34, 52. 23 Park, Bornstein and Post, Amer. J. Physiol., 1955, 182, 12. 24 Park and Johnson, Amer. J. Physiol., 1955, 182, 17. 25 Beloff-Chain, Catanzaro, Chain, Masi, Pocchiari and Rossi, Proc. Roy. SOC. B., 26 Jacobs, Glassman and Parpart, J . Cell. Comp. Physiol., 1935, 7, 197. 27 Ulrich, Pfiig. Arch. ges. Physiol., 1934, 234, 42. 28 Bowyer and Widdas, f. Pliysiol., 1955, 128, 7 P. 29 Edsall. Ion Transport across Membranes, ed. Clarke. (Academic Press Inc., New 30 Barron and Singer, J. Biol. Chem., 1945, 157, 221. 31 Morgan, Kalman, Post and Park, Fed. Proc., 1955, 14, 103. 32 Ussing, Adv. Enzymol., 1952, 13, 21. 33 Wilbrandt, Symp. SOC. Expt. Biol., 1954, 8, 136. 34 Rosenberg and Wilbrandt, Expt. Cell Res., 1955, 9, 49. 35 Goldacre and Lorch, Nature, 1950, 166, 497. 36 Danielli, Symp. SOC., Expt. Biol., 1954, 8, 502. 37 Danielli, Recent Developments in Cell Pliysiology, Colston Pap. 7. ed. Kitching 38 Hodgkin and Keynes, J. Physiol., 1955, 128, 28. 16 LeFevre, Fed. Proc., 1953, 12, 84. 1955, 143,481. York, 1954), p. 229. (Butterworth, London, 1954), p. 1.

ISSN:0366-9033    

DOI:10.1039/DF9562100251

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

258 PERMEATION MECHANISMS PERMEATION MECHANISMS IN BACTERIAL MEMBRANES BY P. MITCHELL AND J. MOYLE Zoology Department, University of Edinburgh Received 30th January, 1956 Bacteria possess a plasma-membrane which acts as an osmotic barrier between the internal medium of the cell and the external medium for many solutes, but establishes osmotic linkage between the internal and external media for the transport of nutrient and waste between the cell and its environment. The specificity and kinetics of phosphate transport in Staph. aurezis resembles that of enzyme-linked reactions and the temperature characteristics suggest a complex movement of components of the plasma-membrane during the passage of phosphate from one side to the other. The material of the plasma- membrane is a complex lipo-protein, of which there is sufficient to form about one mono- layer of lipid and one of protein.The protein component includes the cytochrome system and a number of enzymes. It is suggested that some of these enzymes may themselves be the carriers of the substrates which are found to pass through the membrane during metabolism and in some cases, where exchange diffusion occurs, also during rest. The plasma-membrane of living organisms performs a dual function : it causes the separation of the internal medium of the cell from the external medium with respect to many solutes, but for other solutes (notably the nutrients and end- products of metabolism) it allows osmotic connection or acts as a specific osmotic link between the internal and external media. The two questions which we would like to answer are (i) what physical and chemical properties of the plasma-membrane determine its effectiveness as an osmotic barrier, and (ii) what is the molecular mechanism of osmotic linkage through the membrane?P .MITCHELL AND J. MOYLE 259 1. PLASMA MEMBRANE AS AN OSMOTIC BARRIER At the end of the last century, the botanist Alfred Fischer 1 carried out a care- fully controlled series of studies on the penetration of salts and polyhydric alcohols into bacteria. He observed that the protoplasm of many bacteria was caused to retract from the rigid outer cell-wall by salt and sugar solutions of high osmotic pressure but not by glycerol, urea or chloral hydrate solutions of the same osmotic pressure, and obtained some evidence that the protoplasm did not possess intrinsic rigidity, He therefore suggested that although bacteria are so small- having a volume of the order of lp3-the protoplasm is nevertheless covered by a delicate plasma-membrane which is an effective barrier to the free diffusion of salt and sugar molecules between the outer medium and the interior of the cell.Fischer's view met with considerable opposition 2 because the very nutrients which, according to his plasmolysis experiments, could not pass through the plasma-membrane into the cell, were rapidly metabolized under appropriate conditions. However, evidence in favour of the existence of a plasma-membrane of low permeability in bacteria gradually accumulated. Some of this evidence has been reviewed by Knaysi,3 Mitchell,4 Weibull5 and Mitchell and Moyle.6 Six main methods have been employed to measure the passive permeability of bacterial plasma-membranes : (i) Microscopic observation of the plasmolysis and rate of deplasmolysis of individual cells suspended in hypertonic solution.(ii) Macroscopic measurement, by light scattering, of the dependence of the shrinkage and swelling of the cells on solute concentration and time. (iii) Microscopic measurement of the change of volume of individual cells suspended at different solute concentrations. (iv) Macroscopic measurement, by light scattering, of the rate of lysis of sus- pensions of naked bacterial protoplasts in initially isotonic solution. (v) Measurement of the net rate of efflux of a solute by serial chemical analysis of the " internal medium " obtained by subsequent treatment of the cells with trichloroacetic acid, organic solvents, detergents or heat.(vi) Measurement of the net rate of influx of a solute by serial chemical analysis of the suspension medium and/or " internal medium " in suspensions adjusted so that the total volumes of water on either side of the plasma- membrane are approximately equal. NON-ELECTROLYTES. Methods (i) and (ii) supplemented by method (iii) have been applied to Bacterium coli (American strain B).7 In suspension media buffered at neutral pH with 0.02 M (Na2HP04 + NaH2P04) at 20" C, the apparent permeability of the plasma-membrane to the following non-electrolytes is too low to be measured : sucrose, lactose, D-glucose, D-fructose, D-rnannose, D-galactose, D-sorbose, D-sorbitol, L-rhamnose, L-arabinose, D-xylose.But for some other non-electrolytes the approximate times for half equilibration across the membrane are as follows : erythritol and pentaerythritol 15 min, D-ribose 5 min, glycerol less than 1 min. The rates of penetration of ribose and the polyhydric alcohols are not materially affected when carbohydrate metabolism is inhibited by mM mercuric chloride, mM sodium iodoacetate or mM potassium cyanide. In the presence of 7 mM K+, however, the metabolism of D-glucose, D-galactose, D-niannose or D- ribose causes a limited rise in the internal osmotic pressure of the cells, which can be abolished by mM mercuric chloride, mM sodium iodoacetate, mM potassium cyanide or 10-1 mM sodium dinitrophenate.Methods (iv), (v) and (vi) have been applied to Staphylococcus aureus (strain Duncan),*. 99 10 Micrococcus Iysodeikticus (NCTC 2665) and Sarciizn Iutea (labora- tory strain),ll and methods (iii) and (vi) to Bacillus megateriunz (strain KM).12, 13 In Sfaph. auyeus, M. Zysodeikticus, and S. Zutea the permeability of the plasma- membrane in media buffered at neutral pH with 0-02 M(Na2HP04 + NaH2P04) at 20" C is very low for sucrose, D-glucose, D-fructose, D-mannose, D-galactose and260 PERMEATION MECHANISMS D-sorbose. The permeability of B. megaterium to sucrose is also very low. The half equilibration times for the diffusion of some other non-electrolytes across the plasma-membrane of Staph. aweus, M. lysodeiktus and S. lutea are approximately as follows : D-sorbitol 60 min, L-arabinose 30 min, D-ribose 5 min, erythritol 20 sec, glycerol 3 sec.Urethane equilibrates rapidly across the membrane of B. megaterium. It will be noted that in all the organisms studied, the pentoses, arabinose and xylose, penetrate more slowly than ribose although the only difference between the sugars is in the configuration of the OH groups. Also, while ribose penetrates at about the same rate in all the organisms studied, glycerol, erythritol, arabinose and sorbitol penetrate the plasma-membrane of the cocci more than an order of magnitude faster than they penetrate the membrane of B. coli, the ratio of the surface area to volume of these organisms being about the same (cn. lop-1). A half equilibration time of 7 min corresponds to a permeability coefficient of ca.1 Ajsec. There is no doubt that the hexoses, glucose, mannose and galactose, which can be metabolized by all the organisms studied at a rate of about 1 mole/l. wet cell volume per hour, nevertheless do not approach osmotic equilibrium across the plasma-membrane to a significant extent (when present initially in the external medium at a concentration between 0.1 and 0.5 M) either in normal cells or in cells treated with the metabolic inhibitors, dinitrophenol, iodoacetate, cyanide or mer- curic chloride. The passage of the hexoses through the plasma-membrane (either without chemical change or after chemical transformation on the outer surface of the plasma-membrane) must therefore be linked in some way to carbohydrate metabolism and cannot be due to an independent passive permeability of the plasma- membrane as appears to be the case for D-ribose.We can suggest two distinct types of mechanism whereby linkage between the penetration of the hexoses and their metabolism might occur. (a) The membrane may be impermeable to the hexoses themselves, but permeable to a metabolic product of the hexoses which is formed on the outer surface of the plasma-membrane. (b) The membrane may be permeable to the hexoses through a specific carrier mechanism of the type studied by Rosenberg and Wilbrandt 14 in the red blood corpuscle membrane. If the sugar molecules pass through the membrane only as a specific carrier complex, as the concentration of the sugar on the side of the mem- brane towards which it is diffusing rises across the dissociation constant of the carrier complex, the net rate of diffusion drops rapidly and becomes effectively zero when the carrier is saturated on both sides of the membrane.If thedissociation constant of the carrier complex were in the region of that often observed for enzyme- substrate complexes, namely ca. lO-4M, the entry of sugar might readily be con- trolled by its rate of metabolism within the cell. We should, perhaps, point out that in the permeability experiments described earlier, one measures the osmotic equilibration of solutes at a concentration of 0.1 to 0-5 M across the plasma- membrane. If specific carrier mechanisms exist for the diffusion of solutes across the plasma-membrane, we would expect to be able to observe them only when the concentrations of the solutes used were not much higher than the effective dissocia- tion constants of the carriers.This poses a difficult practical problem. It is, however, an important step to have become aware of it (see 5 2). Methods (i), (ii) and (vi) applied to B. C O Z ~ , ~ and methods (iv), (v) and (vi) applied to Staph. azcreus,9. 10 S. lutea and M. lysodeikticcrs 11 have shown that in media buffered at neutral pH with 0.02 M(Na2HP04 + NaH2P04) at 20" C, the plasma-membranes of these organisms are practically impermeable to the following salts : NaCl, KCl, NH4C1, MgC12, KBr, Na acetate (pH 9), K acetate (pH 9), (NazHP04 + NaHzP04), (K2HP04 + KHzPO~), Na2S04. We have observed that 0.01 M glucose causes a limited rise in the internal osmotic pressure of cells suspended in saline media containing 0.007 M K+.When bacteria have been reported to be permeable to salt solutions on the grounds that deplasmolysis occurs in the course of about an hour, the " permeability " was ELECTROLYTES.P . MITCHELL AND J . MOYLE 26 1 probably due to the presence of low concentrations of Kf and glucose or other metabolites. B. coli is practically impermeable to NaCNS, KCNS and NH4CNS, but Staph. aureus is permeable to these salts. For KCNS at pH 6.8, pH 6-1 and pH 5-0 the times for half equilibration across the plasma-membrane are respectively about 10 min, 5 min and 1 min. The dependence of the rate of penetration of NaCNS on pH is similar to that of KCNS, the rate scale being reduced by a factor of 0-8. These observations show that the plasma-membrane of Staph.aureus is permeable to the cations Na+ and K + in the presence of CNS- . The pH dependence of the permeability to the alkali thiocyanates suggests that the anion and cation permeate separately and that the rate of penetration of NaCNS and KCNS is determined by the rate of CNS- penetration through a titratable, charged plasma-membrane. However, although it is certain that the CNS- ion does not cause a general rise in membrane permeability, there being no loss of internal inorganic phosphate under the conditions of our experiments, it is possible that the presence of CNS- may cause an increase in the permeability to cations, or alternatively that the alkali and ammonium thiocyanates may penetrate in the unionized state.Unlike Staph. aureus, B. coli is not significantly permeable to the alkali thio- cyanates. It is also impermeable to NH4C1 at pH 7 or 8 and to Na acetate and K acetate at pH 7 or 6, provided that traces of glucose or other metabolites are absent. On the other hand, NH4 acetate at pH 6 to 8 equilibrates across the mem- brane in a few seconds. The NH4 acetate does not damage the plasma-membrane, for the cells may subsequently be plasmolysed by the addition of NaCl to the NH4- acetate-containing medium. These observations suggest that the plasma-membrane of B. coli is impermeable to the cations K+ and Na+, and to the anions C1- and CNS-. One may perhaps ask why we have attempted to measure the permeability of the membranes of bacteria by observing net solute transport instead of making use of up-to-date tracer techniques.15 The reason for this is that the flux rates measured by tracers may include an exchange diffusion component of unknown magnitude.With phosphate, for example (with which we shall deal in more detail later) the flux rate measured with tracers is high, but the net flux of phosphate or the permeability of the membrane in the old sense is negligibly low. The net flux and the mutual exchange of solute molecules across a membrane may represent two quite different processes between which we wish to distinguish. In order to determine whether the net flux of an ion through a membrane is possible, it is necessary to allow an electric charge to pass across the membrane. For membranes which are accessible from both sides, this may be done as Ussing has shown 27 by making a suitable electrical connection between the media on either side of the membrane.As the medium on the inside of the bacterial plasma- membrane is not accessible to electrode systems, the permeability of the membrane to one ion can only be measured in conjunction with that of another ion. As pointed out above, isotope exchange methods will not yield the information which we require. One therefore searches for a salt to both ions of which the membrane is permeable so that either ion may then be used to investigate the permeability of the membrane to other ions of opposite sign. PARTICIPATION OF A ' LIPID ' PHASE IN THE PLASMA-MEMBRANE. The permeability properties of the plasma-membranes of the bacteria described above are generally in accord with the view that the obstacle to the free diffusion through the plasma- membrane may be a thin hydrophobic or ' lipid ' layer, for the rates of penetration of the solutes tend to decrease rapidly as the hydration is increased. However, there is no doubt from the results described that the configuration of the solutes also plays an important part in determining the rate of penetration, and that factors other than lipid solubility must be concerned in the interaction between solute and plasma-membrane during penetration.CHEMICAL coMPosinoN OF THE PLASMA-MEMBRANE. Determinations of the weight, morphology and chemical composition of fragments of mechanically262 PERMEATION MECHANISMS disintegrated Staph. aureus, segregated into morphologically homogeneous fractions by differential centrifugation, indicated that the plasma-membrane of this organism, which readily disintegrates into small particles, is a complex lipo-protein containing 41 % by weight protein and 22.5 % lipid.17 Semi-quantitative amino-acid analysis of the protein component, by paper chromatography, showed the presence of a high content of the non-polar amino acids, glycine and alanine, and of the acidic amino acid, glutamic acid.The lipid component contained 1.85 % P and 1.3 % N- about half the phosphorus content and rather more than half the nitrogen content of lecithin. The material of the “ small particle fraction ” has been found to account for some 10 to 15 % of the dry weight of the cells.17~ 18 Since the cells are ca.0 . 7 ~ in diameter and have a ratio of wet to dry weight of ca. 3, it can readily be calculated that the material of the “ small particle fraction ” would form a layer ca. 5 rnp thick if unhydrated. This would correspond to about a monolayer of lipid and a monolayer of protein. Work which is at present in progress has shown that the composition of morphologically intact protoplast membranes from Staph. aureus correspond fairly closely to that of the “ small particle fraction ”, and there can be little doubt that the material described above corresponds to that of the plasma-membrane of normal intact cells. 2. THE PLASMA-MEMBRANE AS AN OSMOTIC LlNK It has been suggested that exchange diffusion might represent the trans- location reaction of active transport uncomplicated by the activity of the coupled reactions which normally drive i t ; and that the study of the characteristics of exchange-diffusion might shed light upon the mechanism of active transport.19 We would like to add that when the permeability of the membrane to a particular solute is caused by a specific carrier, although net transport effectively ceases when the solute concentration is such as to saturate the carrier on both sides of the mem- brane, exchange diffusion would be expected to continue at its maximum rate, and to be very strictly coupled.Also, we suggest that if the passage of a solute through the plasma-membrane is facilitated by the mutual occupation of hydro- philic groups of a “ carrier ” component of the membrane and the solute, the move- ment of the “ carrier ” across the membrane may be as dependent upon the presence of the specifically carried solute as the movement of the solute is dependent upon the “carrier”.The study of exchange diffusion reactions may therefore be expected to play an important part in research on the mechanisms of membrane permeability and active transport. TRANSPORT OF PHOSPHATE. Some of the evidence for the participation of an exchange diffusion-like reaction in the exchange of phosphate across the plasma- membrane of resting Staph. aureus has already been reviewed.19 We shall summarize this evidence 8 9 20921 and discuss it critically in the light of recent observations.lo.22 (i) The plasma-membrane of resting Staph. aureus is apparently impermeable to the H2PO; and HPOj- ions when they are present in solution as the alkali salts.This might be due either to impermeability of the membrane to cations or to impermeability to phosphate ions or both. However, the phosphate ions do not exchange across the membrane with acetate, arsenite, azide, bicarbonate, bromide, chloride, chromate, cyanide, fluoride, glutamate, iodide, molybdate, nitrate, nitrite, oxalate, pyroantimonate, succinate, sulphate, thiocyanate, thio- sulphate, p-toluenesulphonate, tungstate or versenate : they exchange strictly with arsenate. The membrane may therefore be impermeable to all the above anions including phosphate, or be specifically permeable to phosphate and arsenate alone and impermeable to the associated cations. While there is a slow permeation of the membrane by NaCl estimated as chloride and a rapid permeation by NaCNS estimated as thiocyanate, there is no significant exchange of phosphate for C1- or CNS-.We may therefore consider two possibilities : (a) 2 he membrane is impermeable to HzPOi and HPOZ- but slightly permeable to C1- and very permeable to CNS-. (6) The strong electrolytes NaCl and NaCNS permeateP . MITCHELL AND J. MOYLE 263 unionized and the membrane may be specifically permeable to H2P0, and/or HPOZ- (but not to the unionized salts), and impermeable to the associated cations. (ii) By labelling the inorganic phosphate of the medium internal or external to the plasma-membrane of Staph. nureus with 32P, a mixing of the phosphate of the media on either side of the membrane is observed. The rate of this mixing varies with the salinity, the pH and the phosphate concentration of the external medium and with temperature.The dependence of the rate of phosphate exchange on pH and external phosphate concentration indicates that the H2POT.ion and not the HPO2- ion takes part in the exchange. The rate of exchange (P) may be described in terms of the external H2PO4- concentration ([HzPO, ] E ) by the equation in which ISmax. stands for the maximum value of 9 and Kis the value of [H2P0& at which P = Pma.J2. When P is expressed in ,u mole phosphate/g cell dry weight min, and [H2P04 ] E is expressed in mM, K has a value of 0.8 & 0.1 mM between pH 5.5 and 8-5 ; and PmaX. has a value of ca. 10 pmole/g min at pH 7 and a positive slope of some 5 pmole/g min pH unit.The hyperbolic form of eqn. (1)-which is formally identical to the enzyme kinetics equation of Michaelis and Menten 23-shows that a saturation phenomenon occurs in phosphate exchange and implies that one stage in the movement of phos- phate through the plasma-membrane involves a specific spatial or bonding relation between a component of the membrane and the phosphate molecule. (iii) The rate of phosphate exchange is very sensitive to certain inhibitors, notably phenyl-Hg+ and other compounds which combine with thiols of low re- activity. The relationship between the degree of inhibition of the exchange reaction and the concentration of phenyl-Hgf may be represented by (2) M being the amount of phenyl-Hg+, n the percentage activity of P and K' a constant. This indicates a reaction of the type M + X + MX, X representing the sites which when combined with inhibitor (as MX), cause inactivation of a corresponding number of units controlling phosphate exchange.The number of these sites can be estimated to correspond to not more than 4.4 p mole phenyl- Hg+/g cell dry weight. (iv) From the dependence of P on temperature it can be calculated that the total heat of activation for the exchange movement of phosphate across the membrane is 37,40Ocal/mole. Using the above estimate of the number of exchange sites, the absolute value of P gave a maximum value for the free energy of activation of 19,700 cal/mole, leaving an entropy of at least 17,700 cal/mole. The thermo- dynamic data suggest that the movement of phosphate across the plasma-membrane is accompanied by a molecular disturbance quite out of proportion to that which would be expected unless the phosphate moves in relation to some larger molecule or molecules within the plasma-membrane.21 The resemblance of the data to those of reversible protein denaturation is perhaps significant. There can be no doubt that whether the plasma-membrane is permeable to phosphate ions (H2POiions) or not, the movement of phosphate groups across it is dependent upon the existence of a highly specific reaction mechanism.There are three simple alternatives for this mechanism : (a) The membrane allows a net transport of H2POQ through the specific reaction mechanism, and the distribution of phosphate ions across the membrane is determined by the distribution of cations, to which the membrane is supposed to be impermeable.The exchange mechanism may, under these conditions, be identical to that proposed by Ussing for Naf exchange.16 (b) The membrane will not allow a net transport of H2P04. Since there is negligible net movement of phosphate across the membrane even when the concentration on the outside is as low as 0.08 mM (an order of magnitude less K' = Mn/(lOO - n),264 PERMEATION MECHANISMS than the apparent dissociation constant K of the carrier implicit in eqn. (l)), the distribution of phosphate ions across the membrane in resting cells must be main- tained by a strict one-to-one carriage of phosphate inwards and outwards by the exchange mechanism. This might be accomplished by the type of carrier considered by Ussing if it is assumed either that the carrier can move only with a phosphate passenger or that the carrier is always occupied with a phosphate group and that this group may exchange with phosphate ions in the media on either side of the membrane.(c) It is possible that no carrier is involved, but that the media on each side of the plasma-membrane are connected by a " pore " at either end of which there is an adsorption site accessible to phosphate ions in the medium on that side only. One of the adsorption sites is assumed to be always occupied by a phosphate group which is supposed to be able to move from one end of the pore to the other and to be able to exchange with a phosphate group in the medium at the appropriate end of the pore, but not to be able to leave the pore unoccupied.Analysis of the kinetics will not distinguish between the alternative mechanisms proposed above. The fact that the membrane is only one or two molecules thick encourages one to consider that the carriage of solutes across it may occur by a thermal rotation of protein, lipid or other component in a manner similar to that visualized by Langmuir 24 in surface films and by Lundegardh 25 in plant cell mem- branes. The formal similarity of the kinetics of the phosphate exchange reaction to enzyme-linked reactions does not, of course, show that enzymes are involved. There are, however, other circumstances which suggest that the phosphate exchange reaction may be coupled to enzyme reactions. When glucose is present there is a net transport of phosphate inwards through the osmotic barrier.This occurs, not as a result of an increase in the rate of influx but as a result of a decrease in the rate of outflux below that of resting cells. It has therefore been suggested that the phosphate exchange reaction of resting cells represents the active transport reaction operating reversibly because it is not being driven by coupling with carbohydrate metabolism. This concept has been supported by the observation that all the in- hibitors of phosphate exchange are also inhibitors of active phosphate uptake. Staph. aureus is not peculiar in possessing the phosphate exchange system, for a similar system has been demonstrated in B. coli. As might have been anticipated, exchange diffusion of phosphate does not occur across the plasma-membranes of the strict aerobes S.Iutea and M. lysodeikticrrs, under semi-anaerobic conditions. ENZYMES OF THE PLASMA-MEMBRANE. The material of the plasma-membrane of Staph. aureus, isolated as the " small particle fraction " described above, contains more than 90 % of the total activity of an acid phosphatase which acts at the outer side of the plasma membrane in intact cells. It also contains at least 90 % of the cytochrome, measured by the total extinction at a wavelength of 425 mp, and some 90 % of the total succinic dehydrogenase activity, as well as potent lactic dehydrogenase activity, the latter showing the characteristics of the cytochrome blinked enzyme.22 The cytochrome spectrum shows the presence of components with extinction maxima at 604, 558 and 528mp.The plasma-membrane of B. megnteriunz has also been reported to contain the cytochrome system.26 These studies show that part of the protein of the plasma-membrane of Staph. aureus is constituted of enzymes, and we are at present investigating the possible participation of these enzymes in the reactions causing phosphate exchange and accumulation across the plasma-membrane. We suggest that the occurrence of the cytochrome blinked enzymes in the plasma-membrane may be connected with the fact that they are the last members of the enzymic chain and deal with the end- products of metabolism, namely lactic, succinic and formic acids which must be specifically carried through the plasma-membrane. We are indebted to the Scottish Hospital Endowments Research Trust for personal grants in support of this work, and to the Rockefeller Foundation for a grant for equipment.P. MITCHELL AND J . MOYLE 265 1 Fischer, Vorlesungen iiber Bakterien (Fischer, Jena, 2nd ed., 1903). 2 Topley and Wilson, The PrincQles of Bacteriology and Immunity (Amold,London, 3 Knaysi, Efements of Bacterial Cytology (Comstock Publ. Co., Ithaca, New York, 4 Mitchell, The Nature of the Bacterial Surface, ed. Miles and Pirie (Blackwell, Oxford, 5 Weibull, Symp. SOC. Gen. Microbioi., 1956, 6, 111. 6 Mitchell and Moyle, Symp. SOC. Gen. Microbiol., 1956, 6, 150. 7 Mitchell and Moyle, in press, 1956. 9 Mitchell and Moyle, in press, 1956. 11 Mitchell and Moyle, J. Gen. Microbiol., 1956, 15, 512. 12 Weibull, Expt. Cell. Research, 1955, 9, 139. 13 Weibull, Expt. Cell. Research, 1955, 9, 294. 14 Rosenberg and Wilbrandt, Expt. Cell. Research, 1955, 9, 49. 15 Roberts, Abelson, Cowie, Bolton and Britten, Studies of Biosynth. in Escherichia coli 17 Mitchell and Moyle, J. Gen. Microbiol., 1951, 5, 981. 18 Mitchell and Moyle, unpublished. 19 Mitchell, Sywp. SOL Expt. Biof., 1954, 8, 254. 20 Mitchell and Moyle, J . Gen. Microbiol., 1953, 9, 257. 21 Mitchell, J. Gen. Microbiol., 1954, 11, 73. 22 Mitchell, J . Gen. Microbiol., 1954, 11, x. 23 Michaelis and Menten, Biochem. Z., 1913, 49, 333. 24 Langmuir, Science, 1935, 87, 493. 25 Lundegardh, Lantbr. Hogsk. Ann., 1930, 8, 233. 26 Weibull, J. Bact., 1953, 66, 696. 1929). 2nd ed., 1951). 1949), chap. 4. 8 Mitchell, J. Gen. Microbiol., 1953, 9, 273. 10 Mitchell and Moyle, in press, 1956. (Carnegie Inst., Washington, 1955). 16 Ussing, Nature, 1947, 160, 262. 27 Ussing, Physiol. Rev., 1949, 29, 127.

ISSN:0366-9033    

DOI:10.1039/DF9562100258

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

P. MITCHELL AND J . MOYLE 265 THE IONIC SELECTIVITY OF NERVE AND MUSCLE MEMBRANES BY R. D. KEYNES AND R. H. ADRIAN The Physiological Laboratory, University of Cambridge Received 10th February, 1956 The evidence is summarized for regarding nerve and muscle fibres as systems in which the high internal potassium and low internal sodium concentrations depend on the selective passive permeability of the cell membranes and on the activity of coupled sodium-potassium pumps. The propagation of impulses involves a temporary reversal of the membrane permeability from a resting state in which the membrane is as much as 100 times more permeable to potassium than to sodium, to an active state in which sodium enters over 30 times more easily than potassium, the permeability changes being controlled in a regenerative way by the membrane potential, and timed so that first sodium enters and then potassium leaves the fibres.These downhill ionic movements are subsequently reversed by a metabolically-driven active transport mechanism. Recent work on frog muscle has emphasized the close relationship between the resting potential and the potas- sium concentration ratio, and has provided support for the idea that the efflux of labelled sodium results partly from the operation of an exchange diffusion process. The per- meability of excitable membranes to other ions is discussed briefly. The tendency of most living cells to contain more potassium and less sodium than the surrounding medium suggests that many, if not all, cell membranes are capable of discriminating between sodium and potassium ions.This statement assumes, of course, that the differential distribution of the alkali metal ions in living tissues depends on the permeability properties of membranes enveloping each cell, and does not arise, as has sometimes been suggested,l because some I*266 IONIC SELECTIVITY constituent of the cytoplasm has a specific affnity for potassium. There does seem to be some potassium accumulation by ion binding in certain bacteria,2,3 but there is no compelling evidence that such a process is of major importance in the types of cell with which we are concerned (there is sodium binding in frog muscle, but at least part of this appears to occur in the extracellular connective tissue4). Many aspects of the behaviour of nerve and muscle fibres can be explained satisfactorily only on the supposition that the intracellular ions are freely ionized, but prevented from mixing with the extracellular ions by the presence of a selectively permeable cell membrane, as will be apparent from the following summary of some of the properties of excitable tissues.(i) When the tip of a 0 . 5 ~ microelectrode filled with KC1 penetrates a nerve or muscle fibre there is a very abrupt change in the recorded potential, the inside of the fibre being 50-95mV negative with respect to the outside.5 The whole of the potential drop seems to happen over a depth of less than 1 p, and the interior of the fibres is at a uniform potential. An electrode as much as 100 p in diameter can be inserted several cm longitudinally into a 500 ,LL squid axon without altering the potential, as long as it does not scrape the surface at any point.(ii) An internal negativity is still recorded if the microelectrode is filled with Ringer’s solution (or sea water for squid axons5) instead of KCl, though the potential is then smaller by a few mV.6 The system is now symmetrical (Ag/AgCl, Ringer, cytoplasm, Ringer, Ag/AgCl), so that even if the intracellular potassium is “ bound ” in some fashion, there must be a difference between the two Ringer- cytoplasm junctions. This difference evidently consists in the presence of a selectively permeable barrier at only one of the interfaces-the cell membrane. (iii) Impedance bridge measurements 8.33 and studies of the distribution of applied potentials 73 9 show that nerve and muscle fibres have a cable-like structure, a relatively low-resistance core being surrounded by a capacitative high-resistance membrane.The membrane behaves as if it had a capacity of about 1 pF/cm2 in non-myelinated invertebrate nerves, and about 5 pF/cm2 in frog muscle fibres. (iv) Considerations of osmotic balance and electrical conductance indicate that virtually all the potassium in the cytoplasm is in free ionic form. Observa- tions on the rate of movement of 42K inside Sepia axons 10 and frog muscle fibres 11 under the influence of a longitudinal voltage gradient show that the mobility and diffusion constant of potassium are almost the same inside the fibres as outside. It has also been shown in micro-injection experiments 12 that 24Na diffuses radially inside squid axons nearly as fast as in free solution.(v) The time course of the exchange of 24Na and 42K in nerve and muscle fibres 13-17 is reasonably consistent with that expected for a simple two-compart- ment system, and suggests that although the ions can move more or less freely inside and outside the cells, there is a barrier at the cell surface which is relatively difficult to cross. (vi) In the resting or quiescent state, the membrane behaves as if it were much more permeable to K+ than to any other ion, The measured membrane potential agrees well with that predicted by the Nernst equation for variation both of internal 6 and external [K].18-21 At high external potassium concentrations, ten-fold variation in [K], changes the membrane potential of frog muscle fibres by very nearly the theoretical 58 mV, while the corresponding figure for changes in [K]i is 49 mV 6 (this slight reduction in the slope may arise from changes in the activity of the internal potassium with concentration).At the normal plasma potassium concentration of 2.5 mM, both slop= are lower by about 10 mV, presumably because the membrane is not altogether impermeable to other ions, whose effects are liable to become increasingly important when the external potassium is lowered.R. D . KEYNES AND R. H. ADRIAN 267 (vii) During the rising phase and at the peak of the action potential, the membrane behaves as if it were much more permeable to Na+ than to any other ion. The size of the reversed membrane potential (inside positive) at the crest of the spike agrees well with the Nernst equation both when the external sodium concentration is altered22.23S20 and when the internal sodium is altered.249 12 (viii) Observations on squid axons by the " voltage-clamp " technique 25 show that for both sodium and potassium there are critical potentials above which the net currents of Na-+ and K+ respectively are inward and below which they are outward.263 27 These equilibrium potentials agree fairly well with the Nernst equations given above, when the external concentrations of sodium and potas- sium are varied.The equilibrium potential for potassium in Sepia axons, defined as the membrane potential at which the passive inward and outward fluxes of 42K are equal, also fits well with the value calculated from the Nernst equation.21 It is clear from this list that the basic problem is to account for the selectivity of nerve and muscle membranes towards sodium and potassium.But it must be emphasized that excitable membranes exhibit selectivity in more than one way. Not only does the selectivity vary in a subtle fashion during the propaga- tion of impulses, but there is also a mechanism for absorbing potassium and extruding sodium from the cells which seems to be separable from the conduction mechanism. The process by which the ionic concentration differences are buill up and maintained involves a forced movement of ions against the electrochemical gradients and therefore requires the expenditure of energy provided by meta- bolism, whereas the conduction process utilizes these same gradients as an energy source, allowing ions to flow downhill in the kind of way that has been termed " facilitated diffusion ".28 Although it may be an over-simplification to assume that the net uphill and downhill movements take place through entirely separate channels across the membrane, it is certainly possible to block movements in one direction without upsetting them in the other,29 and the relative rates of transfer of different ions in the two directions are by no means the same.The next step is to see how far nerve and muscle membranes do discriminate between sodium, potassium, and other ions. The idea that the conduction of impulses involved permeability changes was implicit in Bernstein's 30 membrane theory, and the occurrence of such changes was first demonstrated experimentally about 20 years ago.31-33 Impedance bridge measurements do not, however, identify the ions which carry the increased membrane current, and detailed information about the variation during a nerve impulse in the permeability towards individual ions awaited the voltage-clamp technique.349 359 25 This enables the net ionic currents flowing through the membrane of a squid axon to be measured at various predetermined potentials, the distinction between sodium and potassium currents resting partly on parallel observations in sodium-free solutions (choline being substituted for sodium), and partly on the results of combined membrane current and 42K efflux determin- ations,36 which showed unequivocally that the large outward current flowing during a sustained depolarization was carried almost wholly by K+ ions.Having analysed the factors affecting sodium and potassium permeability in squid axons, Hodgkin and Huxley 37 then showed in detail that both impulse propagation and certain other features of nerve behaviour could be explained quantitatively in terms of a sequence of permeability changes. As has already been mentioned, the resting membrane is selectively permeable to potassium. The action potential is generated by a transient increase in the sodium permeability, which is controlled by the membrane potential so that it rises in a regenerative manner once a critical level of depolarization has been passed, allowing Naf ions to rush inwards for268 IONIC SELECTIVITY a short while.After the peak of the spike has been reached the sodium permea- bility mechanism is " inactivated ", and the potassium permeability then rises well above its resting value, allowing a rapid outflow of K+ ions to repolarize the membrane. During the refractory period following the spike, both sodium and potassium permeabilities revert towards their original resting levels. It is important to appreciate the degree of selectivity achieved by excitable membranes. One way of arriving at a quantitative estimate is to compare the influxes of labelled ions per unit external concentration; such a calculation for the resting Sepia axon 14 makes PK/PN~ = 13/1. However, there are reasons for expecting this figure to be misleadingly low, and a consideration of the ability of small changes in external potassium concentration to alter the membrane potential in the presence of a much larger external sodium concentration suggests that under normal resting conditions P K / P N ~ may be as much as 100/1 in frog muscle ; 6 for squid axons this argument gives a value of 25/1.22 These figures are for the resting, potassium-permeable membrane ; in the active membrane the selectivity is reversed temporarily, and Hodgkin and Huxley's voltage-clamp results (fig.3 and 6, from ref. (37)) suggest that early in the spike PNJPK is at least 30/1. The figures for the fluxes of labelled ions during the spike 14, 37 give a lower ratio, but this only represents a lower limit, since it does not take into account the separation in time of the permeability changes.In accounting for the operation of the conduction mechanism we have therefore to ask : (i) How does the membrane discriminate so successfully between Na+ and K+ ions? (ii) How is the permeability to sodium so sharply increased by reducing the membrane potential? It should be observed that the initial sodium current increases very steeply with depolarization, the sodium conductance rising e-fold for only about 4mV drop in potential.37 The delayed rise in potassium conductance is not much less steep, an e-fold change requiring 5-6 mV.37 (iii) How is the timing of the permeability changes achieved? It is, of course, an essential feature of the mechanism that the sodium permeability should rise immediately the membrane is depolarized, whereas the potassium per- meability increases after a definite lag.These questions are still virtually unanswered, and there are very few clues to indicate what the answers might be. Hodgkin and Huxley have pointed out 37 that since the permeability changes depend so markedly on membrane potential they are likely to arise from the effect of the electric field on the distribution or orientation of charged molecules or dipoles, and have discussed one or two ways in which this might happen. Possibly bridges of sodium- or potassium-specific sites are formed across the membrane; this might explain the interaction of influx and efflux of labelled Kf ions which has been observed in Sepia axons 21 and frog muscle,lsl 38 and which suggests some kind of channelling in the passage of ions across the membrane.But very few compounds are known which would have the necessary high specificity, let alone any which might occur in living tissues. It would be helpful in this connection to know what sorts of chemical or physico- chemical differences there are between sodium and potassium which might be utilized by the membrane in order to distinguish between them. The greater hydration of sodium ions has often been put forward to explain the relatively low resting sodium permeability, but this seems inadequate to account for all the results. A possible approach is to consider what other ions can deputize for sodium and potassium; for what they are worth, the facts are these. In nerve fibres and frog muscle the only sodium-substitute is lithium,s which appears to behave, as far as excitation is concerned, almost identically to sodium In crus- tacean muscle certain quaternary ammonium ions seem able to replace sodium,R. D .KEYNES AND R. H . ADRIAN 269 but it is uncertain whether the processes involved are exactly analogous to those in other excitable tissues.39 Choline ions are of great experimental value because they behave quite unlike either sodium or potassium, and can therefore be used to make up sodium-free solutions in which the chloride concentration is unaltered. The other alkali metals behave more like potassium than sodium. In Sepia and crab axons, rubidium ions reduce the membrane potential more than the same external concentration of potassium.6.40.41 This seems consistent with the smaller hydrated size and higher mobility of the Rb-+ ion.However, in frog muscle rubidium is a less effective depolarizing agent than potassium.6, 429 43 Caesium ions depolarize all these tissues appreciably less than potassium. We must now turn to the reverse aspect, to consider the active transport of sodium and potassium. Again the experimental evidence concerning the mole- cular nature of the mechanism is meagre, and we do not know much more than that nerve and muscle fibres can undoubtedly bring about a net movement of ions in the uphill direction,44* 45,24 which can sometimes be interrupted by anoxia 46 or treatment with metabolic inhibitors.29 The movements of sodium and potas- sium are roughly equal and in opposite directions (as are the downhill movements during the impulse), and there is now some indication as to how they are coupled together.The two most obvious possibilities would be : (i) that Na+ ions are extruded as such, creating a potential across the mem- brane in the way that the inward stream of Na+ ions through frog skin has been shown to do by Ussing,47 and thus attracting K+ ions passively inwards ; (ii) that the outward passage of a Na+ ion is obligatorily linked to inward passage of a K+ ion in such a way that the system as a whole is electrically neutral. The uptake of K+ would then be coupled more closely with the output of Nat than via the membrane potential, and would properly be regarded as active rather than passive. If the first hypothesis were correct, one would expect interruption of the sodium efflux to result in an immediate drop in membrane potential.In isolated squid axons, poisoning with dinitrophenol, which is known to reduce the sodium efflux markedly, has very little effect on the resting potential; 29 in frog nerve, anoxia does cause depolarization,48 but this observation has not been made under con- ditions where an accumulation of potassium outside the fibres was excluded as the cause of the potential change. The first hypothesis also requires that during re-absorption of potassium the actual membrane potential should be greater than EK; there is no clear-cut evidence that this happens (but see ref. (44)). On the other hand, the fact that removal of external potassium causes an immediate reduction in sodium efflux, in frog muscle 15 and Sepia axons,29 as well as in ery- throcytes,4g9 50 seems to argue strongly in favour of the alternative hypothesis.Further evidence for a fairly tight coupling between sodium efflux and potassium influx are the observations that in Sepia axons these two fluxes have much larger temperature coefficients than those in the other direction, and that both are greatly cut down by metabolic inhibitors, again in contrast to a lack of effect on the potas- sium eHux and sodium influx.29 In frog muscle there is a similar, but much less obvious, difference between the temperature coefficients for the uphill and down- hill fluxes,38 and metabolic inhibitors have rather little effect (Keynes and Maisel 51 found no effect on the sodium efflux, but recent experiments with improved technique suggest that there may be a slight reduction 52).There has been some discussion about the nature of the efflux of labelled sodium from frog muscle. Levi and Ussing 53 considered that it was too large to repre- sent an active transport proc=ss, and proposed that part of it might arise from a non-energy-consuming sodium-sodium exchange which they called ‘‘ exchange diffusion ”. Keynes and Maisel 51 questioned the validity of their arguments, and Hodgkin and Keynes29 showed that in Sepia axons the sodium efflux was270 IONIC SELECTIVITY increased rather than decreased by complete removal of external sodium, which seemed to rule out the occurrence of a sodium-coupled sodium efflux. It now appears that Levi and Ussing were correct in their suggestion, because Swan has recently found that in frog muscle removal of external sodium by substitution with choline or lithium reversibly reduces the sodium efflux by rather more than half? It is not yet clear just how the sodium-coupled fraction of the sodium efflux in muscle is related to the potassium-coupled fraction, and there is still some efflux of sodium in the absence of both sodium and potassium from the external medium.Some idea of the selectivity of the active transport mechanism may be obtained from the flux figures for Sepia axons. The sodium efflux is about 0.5 pmoIe/cmz sec per mM internal Na, while the potassium efflux is about 0.1 pmolelcm2 sec per mM internal K.29 Virtually the whole of the sodium efflux seems, in this case, to be carried by the active transport channel, and practically none of the potassium efflux, since treatment with inhibitors reduces the sodium efflux by a factor of ten or more, but either does not alter or slightly increases the potassium efflux. The process moving sodium outwards therefore has an affinity for sodium which is more than five times greater, and probably much more, than its affinity for potassium.There is little helpful information about the behaviour of ions other than sodium and potassium towards the active transport mechanism. There is evidence that although lithium will enter erythrocytes, it is pumped out much more slowly than sodium,54 but we do not yet know whether this is also true for nerve and muscle. As was mentioned above, lithium appears not to substitute for sodium as far as the sodium-coupled sodium efflux from frog muscle is concerned, so that possibly it cannot be extruded by the sodium pump, A general inability of living cells to extrude lithium might help to explain the belief that this ion is poisonous.Another interesting point which has not yet been tested experi- mentally is the extent to which rubidium and caesium can be transported inwards by the mechanism responsible for the active uptake of potassium. For the sake of completeness, we must conclude this paper with a brief dis- cussion of the permeability of nerve and muscle fibres to anions and to divalent cations. There is surprisingly little information about the anion permeability of excitable tissues, probably because there is no obvious role for anions in the excitation process.Boyle and Conway 55 showed that frog muscle fibres are not impermeable to C1- ions, as had previously been assumed, and it can be calculated from Levi and Ussing’s 53 results for the exchange of 38Cl that the chloride. fluxes in frog muscle are of the same order as the potassium fluxes. From the recent work of Shanes and Berman 17 this also seems to be the case in squid axons. But the nature of the external anions does not greatly influence excitability, as Overton 56 and others have shown, and there is no appreciable net chloride movement during nervous activity.57 Negatively charged molecules much larger than Cl- seem unable to penetrate the membrane, and substances like aspartate and glutamate 58 and isethionate 59 play an important part in invertebrate nerves by providing an impermeable anion to balance the high internal potassium concentration.There is also little to be said about permeability to cations such as Ca2+ and Mg2+. In squid axons the intracellular calcium concentration is only 1/20 of the external concentration (10 mM),60 so that although the influx of 45Ca is very much smaller than the influxes of univalent cations,sl the nerve must possess some means for removing calcium against the gradient. There is, too, some acceleration of calcium inflwr during stimulation.61 But calcium is chiefly of interest because it exerts a profound effect on the membrane permeability to other ions, apparently by producing large shifts in the curve relating sodium permeability to membrane potantial.6zB 63 However, it is once again difficult to envisage what is happening in the membrane, on a molecular scale, when the external calcium concentration is altered.R .D . KEYNES AND R . H . ADRIAN 271 1 Ling, Phosphorus Metabolism, vol. 2, ed. McElroy and Glass (Johns Hopkins 2 Eddy and Hinshelwood, Proc. Roy. SOC. B, 1950,136, 544. 3 Roberts, Roberts and Cowie, J. Cell. Comp. Physiol., 1949, 34, 259. 4 Harris and Steinbach, J. Physiol., 1956, 131, 20P. 5 Hodgkin, Biol. Rev., 1951, 26, 339. 7 Cole and Hodgkin, J. Gen. Physiol., 1939, 22, 671. 8 Bozler and Cole, J. Cell. Comp. Physiol., 1935, 6, 229. 9 Hodgkin and Rushton, Proc. Roy. Soc. B, 1946, 133,444. 10 Hodgkin and Keynes, J. Physiol., 1953, 119, 513. 11 Harris, J. Physiol., 1954, 124, 248.12 Hodgkin and Keynes, J. Physiol., 1956, 131 (in press). 13 Harris and Burn. Trans. Faradoy SOC., 1949, 45, 508. 14 Keynes, J. Physiol., 1951,114, 119. 16 Creese, Proc. Roy. SOC. B, 1954, 142, 497. 17 Shanes and Berman, J. Gen. Pliysiol., 1955, 39, 279. 18 Ling and Gerard, Nature, 1950, 165, 113. 19 Curtis and Cole, J. Cell. Comp. Physiol., 1942, 19, 135. 20 Huxley and Stampfli, J. Pliysiol., 1951, 112, 496. 21 Hodgkin and Keynes, J. Pliysiol., 1955, 128, 61. 22 Hodgkin and Katz, J. Physiol., 1949, 108, 37. 23 Nastuk and Hodgkin, J. Cell. Comp. Physiol., 1950, 35, 39. 24Desmedt, J. Physiol., 1953, 121, 191. 25 Hodgkin, Huxley and Katz, J. Physiol., 1952, 116, 424. 26 Hodgkin and Huxley, J. Physiol., 1952, 116, 449. 27 Hodgkin and Huxley, J. Physiol., 1952, 116, 473.28 Danielli, Symp. Soc. Expt. Biol., 1954, 8, 502. 29 Hodgkin and Keynes, J. Physiol., 1955, 128, 28. 30 Bernstein, Elektrobiologie (Vieweg, Braunschweig, 19 12). 31 Blinks, J. Gen. Physiol., 1936, 20, 229. 32 Cole and Curtis, J , Gen. Physiol., 1938, 22, 37. 33 Cole and Curtis, J. Gen. Pliysiol., 1939, 22, 649. 34 Cole, Arch. Sci. Physiol,, 1949, 3, 253. 35 Marmont, J. Cell. Comp. Physiol., 1949, 34, 351. 36 Hodgkin and Huxley, J. Physiol., 1953, 121, 403. 37 Hodgkin and Huxley, J. Physiol., 1952, 117, 500. 38 Fluckiger and Keynes, unpublished. 40 Wilbrandt, J. Gen. Physiol., 1937, 20, 519. 41 Hodgkin, J. Physiol., 1947, 106, 319. 42 Sandow and Mandel, J. Cell. Comp. Physiol., 1951, 38, 271. 43 Netter, Pflug. Arch., 1928, 218, 310. 44 Steinbach, Symp. SOC. Expt. Biol., 1954, 8, 438. 45 Calkins, Taylor and Hastings, Amer. J. Plzysiol., 1954, 177, 211. 46 Shanes, J. Gen. Physiol., 1951, 34, 795. 47 Ussing and Zerahn, Acta Plzysiol. Scand., 1951, 23, 110. 48 Lorente de No, A study of nerve physiology (Rockefeller Institute, New York, 1947), 49 Harris and Maizels, J. Plzysiol., 1951, 113, 506. 50 Glynn, J. Physiol., 1954, 126, 35P. 51 Keynes and Maisel, Proc. Roy. SOC. B, 1954, 142, 383. 52 Swan and Keynes, unpublished. 53 Levi and Ussing, Acta Physiol. Scand, 1948, 16, 232. 54 Maizels, Symp. SOC. Expt. Biol., 1954, 8, 202. 55 Boyle and Conway, J. Physiol., 1941, 100, 1. 56 Overton, Pfliig. Arch., 1902, 92, 346. 57 Keynes and Lewis, J. Physiol., 1951, 114, 151. 5 8 Lewis, Biochem. J., 1952, 52, 330. 59 Koechlin, J. Biophys. Biochem. Cytol., 1955, 1, 51 1. 60 Keynes and Lewis, unpublished. 61 Fluckiger and Keynes, J. Physiol., 1955, 128,41P. 62 Frankenhauser and Hodgkin, J. Physiol., 1955, 128,40P. 63 Weidmann, J. Physiol., 1955, 129, 568. Press, Baltimore, 1952), p. 748. 6 Adrian, unpublished. 15 Keynes, Proc. Roy. SOC. B, 1954,142, 359- 39 Fatt and Katz, J. Physiol., 1953, 120, 171. vol. I, p. 114.

ISSN:0366-9033    

DOI:10.1039/DF9562100265

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

272 GENERAL DISCUSSION GENERAL DISCUSSION Dr. J. H. Schulman (Cambridge University) said: I am afraid that I quarrel with the interpretation given by Eley and Hedge to their results in that they claim they are measuring adsorption of various proteins on to cholesterol, stearic acid and lecithin * monolayers at the airlwater interface. The technique they are using is that of spreading a monolayer of protein at the liqid air/water interface by a microsyringe injection method. The protein evidently rapidly spreads at this lipid interface and does not appear at the other side of the Langmuir trough boom. This novel technique has interesting implications. Unfortunately they are spreading the protein under lipid films which are com- pressed to pressures well below the collapse pressure of a normal protein mono- layer, e.g.16-1 8 dynes/cm. Consequently, the equilibrium pressures, if there is no association and if there is sufficient protein present in the underlying layer, will reach that of 16 dynes/cm, e.g. the collapse pressure of the protein monolayer. This is observed in most cases. Deviations will occur from this principle due to non-ideal conditions. If the lipid film is in the vapour or expanded states, Raoult’s law will hold and we shall have the addition of the surface pressures of the protein at the air/water interface and the lipid monolayer. In the condensed states deviations and non-ideal conditions are exaggerated, especially with films such as cholesterol or stearic acid which are present in the condensed phases. Very small diminutions in the area of the available surface due to penetration of the protein molecules or portions of the protein molecules, such as hydrocarbon side-chains, will greatly increase the overall surface pressure of the combined film, since the cholesterol and stearic acid monolayers are being used over steep portions of the force against area curve.In previous work on penetration or adsorption phenomena, in order to avoid these physical effects on injection of a soluble com- ponent underneath an insoluble monolayer, I always worked above the collapse pressure of the soluble component at the injection concentration. I showed that this then had direct analogies to adsorption at lipid interfaces by measuring the adsorption on to oil droplets in oillwater emulsion systems where the same lipid as used at the air/water interface was present or dissolved in the oil phase of the emulsion.It can be readily seen that when penetration of lipid monolayers takes place at an air/water interface above the collapse pressure of the protein, such as occurs when the protein solution is at a pH giving the protein molecules an opposite charge to an ionic lipid at the interface, strong adsorption in monolayer or multi- layer form occurs on the oil droplets stabilized by the lipid. When the pH is changed, such that the charge of the protein molecule and the lipid interface are of the same sign, the protein leaves the emulsion droplet interface and is at the same time ejected from the lipid monolayer at the airlwater interface above the collapse pressure of the protein, e.g.16 dyneslcm but not below this pressure. This has been fully established, both for a range of serum proteins and various enzymes (catalase and trypsin). It can be further shown that when penetration by the lipid takes place only below the collapse pressure of the protein and generally not above, no adsorption of this protein takes place on a suspension of the lipid or of that lipid stabilizing an oiljwater emulsion. In fact, adsorption which readily takes place at an oil/water interface, such as Nujol/water emulsion droplets with proteins, is prevented by the presence of a lipid which is not penetrated at the air/ water interface. This protein when adsorbed is strongly denatured in the process of adsorption such that when an enzyme adsorbs at a free oil/water interface complete and irreversible inactivation of the enzyme takes place.If now to this oil a lipid, such as lecithin, which is not penetrated at the airlwater interface, is added the protein is prevented from adsorption and denaturation. A simple * private communication with Prof. D. D. EIey.GENERAL DISCUSSION 273 experiment to demonstrate this phenomenon can be achieved with an enzyme such as trypsin or catalase. These enzymes adsorb readily at a free oil/water interface up to a condensed protein monolayer concentration. The enzyme is completely inactivated by this process. If now lecithin is added to the oil phase no adsorption of the protein takes place at the interface as shown by simply filtering or centrifuging the oil phase droplets away from the aqueous phase and no diminu- tion in activity of the protein in the free aqueous phase can be measured.AS stated above, this monolayer is penetrated by proteins below the collapse pressure of the protein but not above. Experiments of the above nature with other lipids using electrophoresis tech- niques as well are being carried out by M. J. Fraser working with me in Cambridge and are in the course of publication. Dr. Freda Bowyer (King’s College, London) said : Several workers in the field of active transport and facilitated diffusion are at present discussing the possibility of a relationship between these phenomena and reversible denaturation of proteins in the membrane. Have Prof. Eley and Mr. Hedge any information on the phenomenon of reversible protein denaturation in their lipo-protein films ? Secondly, I should like to ask if there is any evidence of discontinuities in the lipid layer and if so whether they are of a temporary or a permanent nature.Finally, is there any information about the amplitude of vibrations normal to the surface of the films? Dr. B. A. Pethica (Cambridge University) said: I would like to raise the question of the effect of calcium ions on stearic acid films. Spink and Sanders 1 showed that very small amounts of Ca markedly affect the properties of stearic acid films. Dr. Betts has confirmed this in our laboratory, and we conclude that the calcium impurities in A.R. NaCl are sufficient to alter the monolayer properties. I think that this consideration will only slightly alter the general interpretation of the results given by Eley and Hedge, but the calcium effect is interesting in view of the well-known place of this cation in biological membrane structures.Prof. D. D. Eley (Nuttingham University) (partly communicated) : In reply to Miss Bowyer, reversible denaturation would certainly be limited to the protein mole- cules adsorbed in the second layer since the molecular areas here are close to those expected for the natural protein (in fact slightly larger), but we have no evidence as to whether these molecules are at all denatured and if so whether the effect is reversible. Presumably if these molecules in the second layer are denatured the bond-breaking and associated configuration changes must be small but we have no evidence of reversibility.We do not consider that discontinuities or holes in the lipid layer are larger than the side chain cross-sections of the protein. To decide the amplitude of the vibrations of the film normal to the surface we should require measurements of the entropy of the surface films, and even then there may be great theoretical difficulties of interpretation. In connection with Dr. Schulman’s comments it seems to me that he is suggesting that the protein molecules penetrate the lipid as complete molecules, giving true mixed molecular films. Our view is that the protein is largely under the lipid film and that a certain fraction (perhaps 50 %) of the hydrophobic chains penetrate the lipid, which is the view I believe Dr. Schulman put forward (ref.(9), (10) in our paper) for the case when the lipid is held above 16 dynes/cm. The time- effects support this view. To spread a protein at the air/water interface under our conditions takes only about 5 min, but to spread it at the lipid/water interface takes about 50 min. This suggests we are not concerned simply with the spreading of protein on “ macroscopic ” gaps in the lipid/water interface. In connection with the nature of the interactions we claim these are hydrogen- bond interactions. A priori this seems essential as the protein-lipid interaction in the primary sublayer will have to compete with protein-water and lipid-water 1 Trans. Faraday SOC., 1955, 51, 1154.274 GENERAL DISCUSSION interactions, which are hydrogen-bond type interactions.It may be, however, that our interaction energies in many cases are weaker than those observed by Dr. Schulman working above the collapse pressure of the protein. It may be desirable that the word “ interaction ” be more strictly defined in further work in this field. In reply to Dr. Pethica it seems to me reasonable that one should have a thermo- dynamic distribution of side chains between the lipid layer and the protein sub- layer, and that similar methods are applicable as when one has a thermodynamic distribution of penetrating molecules between a lipid layer and the whole of the substrate solution. Mr. Hedge (Nottingham University) said: In answer to Dr. Schulman’s criticism that we were not working above the collapse pressure of the proteins I should like to point out that in several systems the collapse pressure was exceeded by the lipo-protein film.Since every system followed the same pattern, it would appear that whether or not the collapse pressure is reached, is of little consequence. This is reasonable, since the layer here is not simply a mixed film, but a complex consisting of a lipid film with a protein film spread immediately beneath it. The surface pressure, then, is borne not by the protein molecules as a whole, but by the lipid film and the penetrated non-polar side chains. This has been supported by a study of the energetics of the systems. Replying to Dr. Schulman’s inter- pretation of the results, that the isotherms obtained are simply force/area curves of the protein, an examination of the surface pressures attained shows this to be incorrect.Taking the case of B.P.A. and stearic acid, at the first discontinuity in the curve the specific area of protein is 1.2 m2/mg and the surface pressure has been increased by 5.2 dyne/cm. However, at the air/water interface, B.P.A. films have a surface pressure of only 0.8 dyne/cm at that area. The discrepancy is even more marked with cholesterol, where the specific area of B.P.A. at the first layer is 2.4 m2/mg. Dr. R. J. Goldacre (Chester Beatty Res. Inst., London) said: In view of the interest in the films of Eley and Hedge as models for biological membranes, I would like to draw attention to some interesting physical properties of various lipo-protein films after they have been collapsed between barriers on the Langmuir trough.1 They form fibres which can be lifted off the water surface on a micro- scope slide and examined under high power.The fibres appear as hollow tubes, about 10-100 microns in diameter, i.e. about the same diameter as living cells. The hollow tubes enclose the underlying aqueous medium, as can be shown by injecting small flageIlates under the film just before collapse. Under the micro- scope they can be seen swimming up and down the tubes, unable to get out. The tubes also respond to changes in osmotic pressure of the external medium, showing that there are no leaks. Such a preparation might be useful in permeability studies on films which would otherwise be difficult to handle. The hollow tubes are separate and this is probably due to the sticking together and pinching off of adjacent folds in the film during the compression.Two points of biological interest arise : (i) This seK-adhesiveness resembles that of the inner surface of the amoeba’s membrane ; when one cuts off an amoeba’s pseudopod, the cut ends immediately join together like sticky rubber-thus making micrurgy possible without killing the cell. (ii) In considering the origin of life, from pools of protein lying around, each covered with a natural surface film, the collapse of such films by the wind would provide a mechanism for wrapping up the underlying soup into little bags by natural forces. Dr. R. D. Keynes (Cambridge University) said: Presumably the ability of yeast cells to retain K+ depends partly on their impermeability to anions.Is it 1 Goldacre, J. Animal Ecology, 1949, 18, 36.GENERAL DISCUSSION 275 possible that the effect of mercuric ions is to allow an outward leakage of anions, accompanied by Kf, i.e. that their action is on the anion impermeability rather than on the cation permeability? And secondly, has Dr. Rothstein any evidence as to how the movements of anions and cations are kept in step, as they must be? He has said, I think, that when phosphate is taken up rapidly there is a simultan- eous acceleration in K entry. How is an exact balance between anion and cation entry achieved? One would suppose that there are two fundamentally different ways of ensuring that electrical neutrality is maintained-either that anion and cation movements are chemically linked so that they are obligatorily transported together with no net transfer of charge, or that the active transport of one species of ion creates a potential difference across the membrane which builds up until it is just large enough to attract the other species inwards at an equal rate.On the first hypothesis, the passive permeability of the yeast cell membrane could be negligibly small to all ions; on the second, the membrane would be relatively permeable to some ions, and variation in the membrane potential would have functional importance. Has Dr. Rothstein any reason for preferring one of these alternatives to the other ? Prof. A. Rothstein (University of Rochester) (commcrrzicated) : The mechanism by which mercuric ion reduces the ability of the yeast cell to retain K+ has not been clearly established.The rapid efflux of K+ is accounted for, at least in part, by an exchange for Hf. The possibility of an increased anion permeability must also be taken into account. There is no appreciable phosphate efflux, but other anions have not been studied. With regard to the balance of anions and cations during the uptake of K+ and of phosphate, the studies from Conway’s laboratory as well as our own, would suggest the following : (i) In the absence of phosphate, K+ is taken up in exchange for HI- derived ultimately from metabolic reactions. (Conway suggests redox reactions.) The increased K+ in the cell is balanced by metabolically produced anions such as bicarbonate and succinate. (ii) In the presence of phosphate, both K+ and phosphate are taken up, pro- vided a substrate is present. However, the time course for the two ions is quite different.The Kf is taken up rapidly and reaches a maximum value early. The phosphate lags behind and in some cases does not start until K+ uptake has ceased. Nor is there, in most cases, a strict stoichiometric relationship between the total uptake of Kf and the total uptake of phosphate. The complete cation- anion balance during uptake of both K+ and phosphate has not been studied. However, differentials in rates of K+ and phosphate uptake presumably are balanced out by metabolic anions and by Hf. Dr. R. J. Goldacre (Chester Beatry Res. Inst., London) said : Dr. Rothstein has listed a number of enzyme activities which are localized at the cell surface. I would like to add to that list some further examples which occur in Amoeba proreus, evidence of which can be seen in the living cell in the microscope : (i> In responding to a touch stimulus, the amoeba reacts only if the cell mem- brane is pushed into contact with the plasma gel, from which it is normally separated by a thin hyaline layer.It is as if an enzyme on the membrane reacted with its non-diffusible substrate in the plasmagel, to produce some substance which causes a contraction. 1 (ii) In digesting a phagocytosed organism, a careful scrutiny of the process under high power reveals that digestion does not begin until the contracting walls of the food vacuole actually touch the prey and press it out of shape--e.g. with paramecium, into a concertina shape. Arresting the progress of the contraction (which can be done by enucleating the amoeba or slightly flattening it, or cutting 1 Goldacre, Symp. SOC.Expt. Biol., 1952, 6, 128.276 GENERAL DISCUSSION off the pseudopod containing the vacuole), prevents further digestion-showing that this is not due to a secreted enzyme solution. The digestive enzyme appears to be only on the walls of the vacuole, and since this is formed by an invagination of the cell membrane, it is probably located on the cell membrane itself. (iii) If one watches the advancing edge of an amoeba, films can be seen peeling off the inner surface of the cell membrane, at the rate of about 12 per min. This I have suggested represents protein synthesis on the cell membrane acting as a template. At any rate it indicates a film-forming enzyme on the membrane since hundreds of films may peel off the same place on the membrane, which is only about 4 molecules thick.This process of stripping-off illustrates one of the advantages of having an enzyme on a membrane, for it makes it possible to drive a reaction in a direction in which it would not naturally occur J by removing one of the components of the reaction from the site of synthesis on the membrane, the reaction may be driven to completion, and synthesis, rather than hydrolysis, brought about. The synthesis of macromolecules is particularly suitable for this mechanism, as they alone would be capable of being stripped off the membrane in sheets. These examples show that the cell membrane is not only a barrier to diffusion but a highly active organelle.Dr. R. N. Robertson (C.S.I.R.O., Australia) said : The structure of the cells of higher plants may be relevant to work on yeast. In plant cells the cytoplasm, which makes contact with the cell wall, behaves as a cation-exchange medium with no high resistance membrane on the outside. This cytoplasm seems to con- sist of a gel containing a high concentration of non-mobile anions which are balanced by the common cations of the plant cells' environment (K, Na, Ca and Mg). High-resistance membranes seem to occur on the inner surface of the cytoplasm adjoining the vacuole (the tonoplast) and on the mitochondria in the cytoplasm. Active transport by a mechanism which is not understood results in the accumulation of ions in the vacuole.While not all plant cells necessarily have the same sort of structure, the properties of most of those investigated are consistent with this. The evidence for this interpretation will be reviewed by Briggs and Robertson in an article to appear in the Ann. Rev. Plant Physiul., 1957. Prof. H. J. C. Tendeloo, Prof. D. MacGillavry, Dr. G. J. Vervelde and Dr. A. J. Zwart Voorspuy (Netherlands) (communicated): During the last 15 years we have investigated root potentials. It was thought that these should be inter- preted primarily as Donnan membrane potential differences,l although other factors may contribute to the total potential difference across the root as a link in the electrochemical chain. The potentials were found to attain apparently steady values and to vary reversibly with concentration changes in the test solutions.The effective membrane is probably one of the outer cell membranes. It may be assumed that the interior volume of the effective Donnan equilibrium system coincides with the AFS. The essential properties of this region are Wac- tically unhindered diffusion and ion exchange, e.g. with the cytoplasm. In the mathematical theory, it was first assumed that only one particular kind of anions at constant concentration was the effective non-permeating ions. Next it was realized that the concentration of non-permeating anions was a function of the pH in the AFS. Furthermore, H+ ions play a role since they are always present and their internal and external concentrations should attain a definite ratio as is the case for other univalent cations.:! If these other ions are present cP.Hope and Robertson, Nature, 1956, 177, 43. Hope, Austral. J. Sci. Res., B, 1953, 6, 396. Hope, Austral. J. Sci. Res. B, 1951, 4, 265. 2 Tendeloo, Vervelde and Zwart Voorspuy, Rec. trav. chim., 1944, 63, 97 ; 1946, 65, 539 ; Versl. Konink. Ned. Akad. v. Wet., 1944,53,169. van der Molen and Tendeloo, Proc. Konink. Ned. Akad. v. Wet., 1947, 50, 763. Hope and Stevens, Austral. J. Sci. Res. B, 1952, 5, 335.GENERAL DISCUSSION 277 in only very small concentrations, the pH of the solutions may become the deter- mining factor. The theory for Donnan systems with a complex mixture of non- permeating ionizing materials 1 has also been developed. On mixing a few buffer- ing systems a solution may be obtained with a buffer capacity showing little vari- ation over a definite pH range.One may therefore define an ideal buffer solution as one with a constant buffer capacity in the appropriate pH range. A solution containing many ionizing substances all at the same concentration and character- ized by a continuous set of titration constants would give such a solution. If a certain component is present in the AFS in a considerably increased amount, the buffer capacity shows a peak at the corresponding pH. From earlier measurements Vervelde deduced estimates for the effective iso- ionic point and the effective buffer capacity for several plant species. Hope gives a corresponding order of magnitude. More recent studies by him and others have given surprisingly high estimites of the AFS volume fraction of the total root volume.Unless these are high one must conclude that the AFS extends a number of cells deep into the root. This shows that cells of the AFS have mem- branes permeable to ordinary inorganic ions (cp. Hope). and in KC1 + HCl solutions (to be published elsewhere) of varying concentrations has given further estimates of the effective iso-ionic point (pH = 3) and for the effective buffer capacity (in the range 3 to 10 x 10-3 M). Values of the iso-ionic points appear to be fairly reliable. The confidence interval of the buffer capacity estimates, unfortunately, is still very wide. Further work in this field is in progress. Dr. W. B. Hugo (Nottingham University) said: The action of phenol and 2-phenoxyethanol on the oxidation of single substrates by washed suspensions of Bact.coli (NCTC 5934) was markedly influenced by the nature of the respir- atory substrate. At concentrations of 0.1 % to 0.2 % w/v the rate of oxygen uptake on mannitol, lactose and glucose was stimulated (10-20 %) but using the same concentration with lactate, succinate, pyruvate or acetate the uptake was inhibited (10-15 7 9 . 3 These differences could not be attributed to changes in the viable population, nor was the stimulation associated with an uncoupling effect as found with certain nitrophenols. It was considered possible that the observed differences in response occasioned by the nature of the substrate resulted from differences in the cellular location of the respiratory enzymes mediating oxygen uptake.Treatment of the cell-free glucose-oxidizing system with phenol or 2-phenoxyethanol over the full range of concentrations did not stimulate oxygen uptake. Cell-free lactic-oxidizing system was more resistant to the action of the antiseptics than the glucose system, a result in accordance with the behaviour of the intact cells in presence of inhibitory concentration of these antiseptic^.^ The results with the cell-free preparations are considered to support the hypothesis that enzyme location (accessibility) as well as sensitivity are both opera- tive in determining the response differences of intact cells to phenol and 2-phen- oxyethanol. Dr. G. Manecke (Berlin, Germany) said : We are investigating a new possi- bility of transfer of H+-ions through membranes.We have prepared membranes from redox-resins. Our redox-resins are insoluble swelling high-polymers which A further investigation of root potentials of lucerne clover in KCl solutions 1 Vervelde, Proc. Konink. Ned. Akad. v. Wet., 1948, 51, 308. Vervelde and Tendeloo, MacGillavry and Tendeloo, Rec. trav. chim., 1954, Rec. trav. chim., 1953, 72, 62. 73, 15. 2Tendeloo and MacGillavry, Proc. Konink. Ned. Akad. v. Wet., 1954, 57, 509. MacGillavry and Tendeloo, Proc. Konink. Ned. Akad. v. Wet., 1954, 57, 513. MacGillavry and Tendeloo, Bull. Central Electrochem. Res. Inst., Karaikudi, India, 1955, 2, 46. MacGillivray, Proc. Konink. Ned. Akad. v. Wet., 1956, 59, 73. 3 Hugo and Street, J. Gen. Microbiol., 1952, 6, 90. 4 Hugo, J. Gen. Microbiol., 1956, in press.c 7 OH OH i -0- - The oxidation-reduction -& n # + 2nHf + 2ne When the resin in the hydroquinone form is oxidized it loses H+-ions and electrons.The electrons are taken up by the oxidizing agent. The Hf-ions are either taken up by the oxidizing agent or they can accumulate in the solution. When a membrane prepared from such a redox resin is in contact on one side with a suitable oxidizing agent then the linked hydroquinone molecules on the surface are oxidized, the electrons are taken up by the oxidizing agent and the H+-ions pass into the solution. The oxidized surface layers of the membrane react with the layers beneath, which are still in the reduced form. The quinone reacts with the hydroquinone and semiquinone is formed. The intermediate state of this reaction can involve a hydrogen bonding between vicinal quinone and hydroquinone molecules.The semiquinone formed in the surface layer of the membrane can be oxidized again by the oxidizing agent which itself cannot penetrate the resin so that again electrons and H+-ions pass into the solution. By such a process the oxidation may proceed into the interior of the membrane while electrons and Hf-ions proceed to the surface and then into the solution. Now if we have an oxidizing agent on one side of such a redox membrane and a reducing agent on the other we can transfer by such oxidation-reduction reactions electrons and H+-ions through an organic membrane. We have evidence that such a process takes place in our redox resin. Dr. P. Mitchell (Edinburgh University) said : We have reached a point in this discussion where we need to raise the old problem as to whether we should attempt to describe natural membranes as perforated solid films or as liquid films.Some years ago I pointed out that there seems to be only one difference between the liquid lipid membrane and the solid molecular sieve models when the holes in the latter are considered to be very small.1 In the liquid membrane model, it is supposed that thermal movements cause the material of the membrane to yield before a penetrating molecule and close behind it, while in the solid membrane model the path of the penetrating molecule is already traced and is maintained by the rigidity of the substance of the membrane. In either case, the specificity of permeability would depend upon the specificity of bonding between the per- meant and the material of the membrane.I suggest, however, that the liquid membrane may be capable of greater specificity than the solid one. In the paper by Mr. Stein and Prof. Danielli we are asked to consider the move- ment of glycerol through pores of high specificity in the red cell membrane, and to accept the concept of competitive inhibition for this process. We must pre- sumably visualize the glycerol molecules adsorbing specifically at the mouths of the pores and moving through the membrane by thermally activated steps between which recombination with the pores must take place. It would seem that, for such a process, high specificity would militate against rapid movement. More- over, since one does not distinguish between the specificity for adsorption in and for diffusion through the pores, one would expect that the so-called competitive inhibitors would be capable of passing through the pores as well as being able to enter them, and that they should therefore be regarded as alternative substrates. If this is so, the specificity of the facilitated diffusion of glycerol must be very low.1 Mitchell, The Nature of the Bacterial Surface, ed. Miles and Pirie (Blackwell, Oxford, 1949), chap. 4.GENERAL DISCUSSION 279 In enzyme reactions-from which the concept of competitive inhibition stems -the overall process is generally regarded as being divided into three steps : the reactant becomes adsorbed on the enzyme molecule with a specificity that in- dicates a three-dimensional shape relationship.This adsorption is followed by a thermal movement which causes an electronic rearrangement by which the re- actant changes to the resultant; and the resultant may then desorb. I suggest that specific permeation reactions of high specificity may occur by a process quite analogous to that of enzyme reactions and that the thermal movement by which the reactant is converted to the resultant in enzyme reactions may, in specific permeation reactions, cause a change of configuration or position of a protein carrier in the membrane such that the accessibility of its adsorbed passenger changes from one side of the membrane to the other. I think that for membrane reactions of high specificity, this type of model has the merit of conforming more closely to what is known of the kinetics of biochemical reactions than the simple pore model of Stein and Danielli.At all events, these considerations show how important it is to discover more about the physical state of the components Of plasma-membranes ; in particular, to determine the extent to which components of the membrane are free to make rotational and translational movements. Prof. J. F. Danielli (King’s College, London) said: That there are certain formal similarities between diffusion through liquids and diffusion through a molecular sieve was of course appreciated by earlier workers, e.g. Michaelk. However, we differ from Dr. Mitchell when he suggests that a liquid membrane may be capable of greater specificity than a solid membrane.It is our view that the solid structure in which the pores are of the same order of magnitude as the permeating molecules will be capable of a much higher degree of stereo- chemical specificity than will any liquid membrane. In the model which we have suggested, high specificity is based simply on a suitable distribution of hydrogen- bonding groups, which will permit the breaking of hydrogen bonds one by one for diffusing molecules of an appropriate structure, but not otherwise. In theory the pore of this type should permit rapid passage of stereochemical appropriate molecules, but not of other molecular species. The suggestion that there may be a change in the configuration of a protein component of a membrane is of course within the range of possibilities.In fact, it has been suggested elsewhere that the transition from facilitated diffusion to active transport may be carried out by the supply of energy to a molecule which has just this potential, i.e. it is quite possible that facilitated diffusion is based upon the protein pore of the type we mentioned in this paper, and that active transport is based upon an energized version of a similar molecule. However, Dr. Mitchell is wrong in thinking that this more complicated model has any present demon- strable superiority to the simple protein pore which we have already suggested. It seems undesirable to consider hypotheses which are more complicated than the simplest which will explain experimental data. In fact, the urgent need at the mment is not to produce new hypotheses but to find appropriate methods of isolating from membranes the macromolecular components which are responsible for membrane specificity, and to find out upon which properties their specificity depends.Prof. J. F. Danielli (King’s College, London) said: I should like to draw attention to the recent work of Prof. J. Monod and his colleagues at the Pasteur Institute in Paris on the genetical control of permeation into bacteria. They have shown that for a sugar to be metabolized it may be necessary to have both an appropriate enzyme set-up within the bacteria and an appropriate membrane component which specifically admits the metabolite to the cell. Many compounds will be metabolized by a cell only if it has both the enzymic system and the specific membrane system, which Monod calls a permease. Bacteria can be obtained which have the permease only, i.e.will admit a particular metabolite into the cytoplasm, but cannot metabolize it; or may have the metabolic system only,280 GENERAL DISCUSSION in which case the sugar would be metabolized if it could get into the cytoplasm, but is unable to do so because of the absence of permease. The genes controlling the permeases and metabolic enzymes appear to be different, though perhaps closely related. Dr. R. D. Keynes (Cambridge University) said : Any mechanism designed to allow a particular type of molecule to penetrate a cell must not at the same time constitute an excessive leak for other molecules. Will not pores of the kind pro- posed by Prof. Danielli and Dr.Stein be somewhat unselective towards those inorganic ions which are much smaller than sugar molecules? Or would they suggest that the pores act as a selective leak favouring anions, to which the erythro- cyte is enormously more permeable than to cations? Prof. J. F. Danielli (King’s College, London) said : Davson has suggested that small inorganic anions, which penetrate red cells very rapidly, do in fact penetrate by facilitated diffusion. It does not necessarily follow however, that they would be able to diffuse through pores which were specially selective towards sugar molecules. It does on the other hand follow that pores which are selective towards sugar molecules may be unselective towards small molecules such as formamide, methyl alcohol and water. It has in fact been shown that many cell membranes, including the red cell membranes, are abnormally permeable to such small molecules.On the whole, therefore, the evidence which is available suggests that the unselectivity towards small molecules which might be predicted from our theory does in fact exist. Dr. J. B. Finean (Birmingham University) said : If “ pores ” through the cell membrane result from the type of molecular organization illustrated in fig. 8b of Stein and Danielli’s paper then there is a possibility that they could be detected in high-resolution electron micrographs of thin sections through cells. The avail- able electron micrographs of the red cell membrane are probably not sufficiently clear to show such features but the resolution obtained in electron micrographs of other membrane structures is certainly great enough to reveal such structural detail providing it takes up osmium tetroxide. These membranes seem to take up osmium tetroxide in a specific way which results in their being visualized in the electron microscope as two dense lines of osmium separated by a comparatively light space.The general feeling is that the osmium is probably associated with a protein layer and, if this is so, a “ pore ” of the type suggested might be expected to appear in the electron micrograph as a dense line crossing the light space of the “ double ” membrane. If the osmium deposition is with a component other than protein then the “ pore ” might appear as a gap in the otherwise continuous lines of the “ double ” membrane, but this would be difficult to distinguish from gaps produced by the preparative procedure.If the “ pores ” are few in number the chances of cutting a section through one of them might be small, but electron micrographs of serial sections through a cell may provide direct evidence for the existence of such “ pores ”. I wonder if the authors have considered this possible source of confirmatory evidence. Prof. J. F. Danielli and Mr. W. D. Stein (King’s College, London) said : We have considered the possibility of using electron microscopy to obtain direct evidence of the structure of the areas of the cell membrane which have special permeability properties. The techniques which are at present available are not satisfactory but Dr. M. M. Coombs of our laboratory is engaged in developing cytochemical methods which could be used with the electron microscope.If his experiments are satisfactory we may be able to make satisfactory studies of the type you suggest. Dr. R. J. P. Williams (Oxford University) (contributed) : With regard to the biological systems I would like to make one or two comments which relate to the selectivity of cation function. Stein and Danielli note that cupric and mercuric ions inhibit glycerol transport. Prof. Gregor has suggested that this inhibition might well be compared with the ability of the cations to combine with carboxylateGENERAL DISCUSSION 28 1 groups of polyacrylic acid. Prof. Gregor went on to suggest that this com- parison lent support to the views of Stein and Danielli on the mechanism of transport of glycerol through the membrane.Is it not as likely that the cupric ions, and mercuric ions, inhibit by combining with sulphydryl groups whence their action is to prevent phosphorylation reactions? Cupric ions have a very high affinity for sulphur groups. While many biological membranes have a high selectivity with reference to the sodium/potassium ratio they also are selective with regard to magnesium and calcium. Peculiarly magnesium tends to enter the cell with potassium while sodium and calcium are excluded. Now as far as I know simple membranes have a higher affinity for sodium and magnesium relative to potassium and calcium (carboxylate resins) or for potassium and calcium relative to sodium and mag- nesium (sulphonate resins) and none of them show the selective behaviour common to many biological systems.On the other hand, it is worth noting that ethylene- diamine tetracetic acid combines more readily with calcium than with magnesium and more readily with sodium than with potassium. Selectivity of biological membranes appears to be related to some types of complex ion formation rather than to the simpler properties of the hydrated cations. Dr. B. A. Pethica (Cambridge University) said : Mr. G. V. F. Seaman and I wish to report some electrophoretic studies made on the normal and sickle human red cell which have some bearing on the properties of the cell membrane We. have shown that the negative charge of the normal and sickle cell is predominantly due to phosphate groups as judged electrophoretically.These findings confirm some similar observations made on the pig erythrocyte by Winkler and Bungenberg de Jong,l which led them to postulate a membrane consisting essentially of oriented phospholipid molecules. Normal red cells do not in general adsorb proteins, with the exception of a few whose isoelectric points are > ca. pH 10. Sickle cells show certain abnormal- ities in this respect. Of particular interest is the fact that normal cells remain unaltered electrophoretically after sphering by passage over glass beads and being converted back to discs with serum albumin. This confirms the findings of Dr. E. A. Browne (private communication) using radio-iodinated serum albumin and the original experiments of Furchgott and Ponder.2 These findings are not in accord with the view that the lipid membrane is coated to any extent with protein.Dr. W. F. Widdas (King’s College, London) said : I should like to draw attention to some additional evidence of the complex nature of DNFB inhibition of hexose transfer in erythrocytes. Dr. Bowyer and I have studied the inhibition produced in cells incubated with DNFB in the presence and absence of glucose and find that in the presence of glucose inhibition develops more rapidly. The removal of DNFB from the medium as judged by testing the supernatant with a new suspension of cells is comparable in both experiments. The amount of DNFB used up during incubations at different temperatures bears no direct correlation with the inhibition produced and it would appear that a substantial amount of DNFB is used up by a non-inhibitory reaction.The occurrence of this secondary process is a factor in the complexity of the overall reaction with the cell. The acceleration of the inhibitory reaction of DNFB by glucose is the opposite to what one might expect if there was competition between glucose and DNFB for the same sites but it is possible that glucose forms hydrogen bonds with the membrane component and makes groups more available for reaction with DNFB by a process akin to reversible denaturation. Preliminary results with urethane and guanidine show a similar though smaller effect than glucose. 1 Arch. Neerl. Sci. Ext., 1940-41, 25, 431. 2 J . Gen. Physiol., 1941, 24, 447.282 GENERAL DISCUSSION It should be emphasized that the specificity and special type of kinetics of facilitated transfers fit several molecular mechanisms.The kinetics do not appear to have a parallel in any of the work on artificial membranes described at this Discussion. I should also like to apologize for a mis-statement regarding the contracting protein. The extension of this idea to specialized transfer through membranes is due to Danielli.1 Dr. P. Mitchell (Edinburgh University) said : It has become very clear in the course of this Discussion that in order to decide between hypotheses proposed to explain kinetic data on membrane permeation more information about the com- position of the membrane is required. I think it may therefore be appropriate to summarize some recent observations that we have made on the enzymic com- position of the plasma-membrane of Staphylococcus aureus.Controlled autolysis of Stahp. az~rcus in a medium of high osmotic pressure, followed by dilution with water, causes the external cell wall-which is normally a spherical shell of considerable tensile strength-to break into two hemispherical parts. At the same time, the protoplast, which is normally contained within the cell-wall7 swells and bursts. Phase-contrast microscopy shows that the plasma- membrane does not disintegrate at this stage, and the intact membranes may be collected by differential centrifugation. On washing with distilled water, however, the plasma-membranes progressively break down until the material consists en- tirely of small particles, indistinguishable from the so-called " small particle " lipoprotein fraction which can be isolated from mechanically disintegrated bacteria and which, some years ago, we suggested might represent the material of the plasma-membrane.2 Table 1 shows a comparison of the weights of cell-wall and " plasma-membrane material " produced by the autolytic and by the mechan- ical disintegration techniques from the same batch of washed bacterial suspension.The agreement is very satisfactory. Table 2 shows the distribution of some enzyme activities between the plasma-membrane fraction and the protoplasm. The activities in the cell-wall fraction were too small to be measured. Since the end-products of metabolism of this organism are lactic, formic and succinic acids, it seems legitimate to suggest that the corresponding dehydrogenases which are concentrated in the membrane may be concerned with the movement TABLE 1 .-QUANTITATIVE SEPARATION OF CELL ENVELOPE FRACTIONS OF Stuphylococcus uureus - mechanical weight morphology material autolytic weight morphology intact cells 100 100 cell wall 1 6 .6 spherical shells 1 2 . 5 hemispherical shells plasma membrane 9-6 small particles 10.2 spherical shells TABLE 2.-DISTRIBUTION OF ENZYME ACTIVITIES IN StUphYlOCOCCUS UUrr?US enzyme succinic dehydrogenase lactic dehydrogenase malic enzyme malic dehydrogenase formic dehydrogenase a-glycerophosphate dehydrogenase glucose-6-phosphate dehydrogenase glucose-6-phosphatase acid phosphatase protoplasm plasma-membrane or " small particle " fraction > 90 80-95 > 90 > 90 > 90 50-70 3 10 > 90 < 1 0 5-20 < 10 < 1 0 < 1 0 30-50 97 90 < 1 0 1 Danielli, Symp.SOC. Expt. Biol., 1954, 8, 502. 2 Mitchell and Moyle, J. Gen. Microbial., 1951, 5, 981.GENERAL DISCUSSION 283 of their substrates outwards through the cell membrane. The presence of the cytochrome system in the membrane also lends some support to hypotheses in which the cytochrome system is supposed to be implicated in the movements of ions across the membrane, and in the last analysis to provide the source of the membrane potential. Dr. Freda Bowyer (King’s College, London) said: Dr. Mitchell and Miss Moyle state that in the organisms studied hexose transfer must be linked to carbo- hydrate metabolism. Their suggested mechanism (6) seems to be equivalent to saying that the postulated carrier is 100 % saturated at low concentrations of hexose and seems rather unlikely.Could it be therefore that the organisms are like adult rabbit red cells which have a very low hexose permeability-about the same order of magnitude as the rate of metabolism. Morgan et aE.1 have shown that inhibition of metabolism causes slow accumulation of glucose. I should like to ask Dr. Mitchell if his experiments with metabolic inhibitors were continued over a long period of time. One of the inhibitors used by the authors to inhibit metabolism, i.e. mercuric chloride, is itself a powerful inhibitor of glucose transfer in the red cell. Dr. Mitchell and Miss Moyle state that 0.01 M glucose causes a rise in internal osmotic pressure when 0.007 M K+ is present. i should like to ask if they have any explanation of this and any knowledge of the penetrating species.i t would be very interesting to know if it is (a) entry of K+ requiring metabolic energy, or (6) the entry of glucose accelerated by Kf in a manner similar to the acceleration of the hexokinase reaction described by Rothstein. Dr. P. Mitchell (Edinburgh University) said : Dr. Bowyer is correct in saying that our suggested mechanism (b) for hexose transfer requires the carrier to have a high affinity for hexose, comparable to that of an enzyme for its substrate. This does not seem to us to be particularly unlikely. As to whether a low affinity facilitated permeability to hexose might account for the normal rate of entry : our results, both with the cocci and with Bncteriunz coli indicate that, under the conditions of our experiments, the rate of entry of hexose into cells suspended in potassium-free or inhibitor-containing media was not greater than one-fifth of the normal rate of metabolism.The concentration of glucose used in the experiments on the cocci was very high and might have affected the membrane permeability, but this was not the case in the experiments on Bact. coli. We agree that mercuric chloride might inhibit a specific hexose transfer, but cyanide, dinitrophenol and iodoacetate would not be expected all to do the same. We have not yet studied the cause of the limited rise in the internal osmotic pressure of Bact. coli in glucose and potassium-containing media. It might be due to either of the mechanisms which Dr. Bowyer suggests or to other mechanisms such as anion uptake, or perhaps more likely to a combination of mechanisms.Our comments on the similarity of the thermodynamics of the movement of phosphate across the membrane of Sraphyfcoccus aureus to the thermodynamics of reversible protein denaturation were intended only to indicate the type of system which might be involved. We should be cautious about suggesting that these data are evidence for an unfolding of polypeptide chains accompanying phosphate transfer-indeed we are not certain to what extent the protein denaturation which we are considering (e.g. of trypsin at pH 6.5) involves the unfolding of polypeptide chains. In view of the known complexity of the membrane, the thermodynamic data can only be said to indicate that the movement of phosphate through the membrane is dependent upon a thermal movement of part of the membrane of low probablity and not involving a large free energy change.We are glad that Prof. Danielli has brought up the question of the so-called permeases recently studied by Prof. Monod and his collaborators in Paris. in 1 Morgan, Kalman, Post and Park, Fed. Proc., 1955, 14, 103.284 GENERAL DISCUSSION spite of their name, the permeases are active transport systems which cause the accumulation of sugar derivatives and of amino acids in or on the cells of Bacterium coli during the active metabolism of glucose. These systems closely resemble the amino acid accumulating systems extensively studied by Gale in Gram-positive bacteria.1 The interesting work of Prof. Monod and his collaborators, of which we have only just learned since it has not yet been published, appears to agree with our own observations and to help substantiate the view that the movements of metabolites into bacteria is controlled by enzymes or by carriers of equivalent specificity and affinity.Dr. J. H. Schulman (Cambridge University) said: I do not think that the question of the structure (porous or continuous) of the membrane around the cell is of vital importance in the permeability of ions or molecules through the cell wall. I question whether “ holes ” are in any way necessary. It can be easily shown that water will evaporate from an air/water interface comparatively quickly through oil layers up to a millimetre thick, whilst evaporation of water can be drastically restricted by a condensed monolayer of a long chain alcohol only lOA in thickness.If the hydrogen bonding between the hydroxyl groups in the alcohol molecule is broken down and the polar group replaced by an ester group, unim- paired evaporation takes place. Similarly, if the solid structure of the monolayer is broken down by the insertion of double bonds in cis form into the hydrocarbon portion of the molecule making the film liquid, normal evaporation again proceeds. Therefore, control of the permeability of oil layers can be easily made by structure in the non-polar portion and hydrogen bonding in the polar portion of the mole- cules making up the interface. It can also be shown in work done recently by C.S. Hocking, that potassium and sodium ions can readily diffuse through a several millimetres thick layer of amyl alcohol if a Teorell potential is held across the oil layer with water on both sides. The quantity of potassium and sodium dis- solving into the amyl alcohol layer is greatly influenced by surface-active anionic and cationic ions spread at the two interfaces or dissolved into the oil layer. The diffusion of potassium and sodium through these layers was determined by flame photometer techniques. Dr. R. J. Goldacre (Chester Beatty Res. Inst., London) (communicated) : The question of how a living cell discriminates between sodium and potassium ions need not involve properties other than those with which we are familiar. Perhaps the well-known high solubility of most sodium and potassium salts suggests that they can never be bound to a carrier ; but it is difficult to see how selective uptake of ions of similar charge and little difference in size could occur without a binding process being involved at some stage or other.The question then becomes, can this binding or adsorption ever be very different for sodium and potassium, and is adsorption ever strong enough to have any marked effect ? The actual values of the solubilities of the same salt of potassium and sodium sometimes differ by a factor of 10 or more (and much higher in those salts used in gravimetric analysis, e.g. cobaltnitrite, chlorplatinate, etc.) ; for example, the ratio of the solubilities of potassium carbonate to sodium carbonate is 12.7 at 0” C.Such differences should be reflected in differences in adsorption of these ions on surfaces, and there is no reason to think that adsorption of these two species will be exactly the same. A concrete example where this difference in adsorbability has been exploited in industry, is in the separation, by the froth flotation process, of crystals of sodium chloride from crystals of potassium chloride, suspended in a saturated solution of mother liquor.2 It is possible to find conditions (of pH and concentration) under which a hydrophobic monolayer of sulphonated castor oil will be adsorbed 1 Gale, Adv. Protein Chem., 1953, 8, 285. 2 Kuzin, J. Appl. Chem. (U.S.S.R.), 1939, 12, 836, 843 ; see also Weiner, U.S. Pat. 2,382,310, and Weinig, U.S. Pat. 2,188,931.GENERAL DISCUSSION 285 on one species of crystal and not on the other, Such a complete discrimination indicates that it is not impossible for selective adsorption to be a basis for biological discrimination, for example in an adsorption-desorption cycle as suggested by Goldacre.1 Dr. E. Glueckauf (Harwell) said : The retardation of the migration of cations by simultaneously moving anions mentioned by Dr. Keynes is apparently quite a general phenomenon. We have also observed the converse effect, namely the acceleration of Na-cation migration by a fast moving cation such as Hf ions. This is obviously a field deserving more detailed investigation. Dr. G. S. Adair (Cambridge University) said : Fig. 1 represents a hypothetical “membrane protein”. The letters Y and Z represent part of two long poly- peptide chains or helices on the surface of the cell.by reversible cross-links. A is a cross-link which is not easily broken (possibly covalent) ; E is a labile cross-link in contact wth the liquid outside the cell wall; G and H are additional cross-links and I is a labile cross-link, in contact with the liquid inside the cell. Fig. 1 b represents the “ membrane protein ” after the labile link E has been broken by thermal movements or by contact with a molecule X present in the liquid surrounding the cell. The letter S represents a slit or pore, formed by the separation of the chains after the rupture of the link E. The dimensions of the slit depend on the distance between A and G, the thermal movements of Y and Z, and the charges of the amino acid side-chains and other groups between A and G.The properties of the slit must depend on the groups on the chains near E, represented by the letters B and C, and the groups D and E, released by the rupture of link E. Four examples can be given. A slit S1 per- meable by molecules soluble in organic liquids may be formed if B and C represent leucine or isoleucine side chains or lipid groups associated with the chains Y and Z . The link E might represent van der Waals’ attractions between organic groups. A slit S2 permeable by glucose These chains are connected B flC D E F U C 0 H ,nZ I la U C 0 H y n z I I b FIG. 1 .-Diagram representing hypothetical “ membrane pro- tein ”. ~~ might be formed if B and C-represent carbohydrate groups and E a Iink between sulphydryl groups at D and F.A slit S3 permeable by anions might be formed if B and C or (D and F) be positively charged side chains (arginine, lysine or histidine). A slit S4 permeable by potassium ions might be formed if D and F represent carboxylic groups and E a calcium ion. Arrangements of slits of type S4 may have possibilities for the spread of a region of high permeability. Mech- anisms of this type may prove useful in studies of nerve and muscle. It seems possible that membrane proteins on the cell surface might be folded to give a pattern, so that the opening of a slit in the membrane protein MI may be correlated with the closing of a slit in the adjacent protein M2. In reply to a question put to me by Dr. Stein, active transport of a substance across a “ membrane ” may be possible, if the cross-link E (in contact with the external surface of the cell) be broken by contact with an ion or molecule XI.If the slit S5 opens for a short interval of time (10-6 sec) the ion or molecule X1 may diffuse across the slit Sg, and enter the cell. 1 Goldacre, Int. Rev. Cytology, 1952, 1, 135.286 GENERAL DISCUSSION The rupture of the link E may be associated with the hydrolysis of a molecule X2, which forms part of the link E (enzyme-substrate). The system may be re- stored to its initial state when a second molecule of the substance X2 combines with the groups D and F. It is possible that a steady state may be reached where the probability that a molecule XI enters the cell is equal to the probability that a molecule of X1 is lost by diffusion. Much detailed work is required before the idea that the surfaces of living cells are partly composed of “ membrane proteins ” can be compared with hypotheses described or referred to by Danielli and Stein, Bowyer and Widdas, Mitchell and Moyle, and Keynes and Adrian.(1) The cell surface may be a bimolecular film of lipid molecules, with adsorbed layers of protein. (2) The cell membranes may contain polar spores, hydrogen bonded. (3) The cell mem- branes may contain mobile membrane components or carriers. At the moment, the hypothesis (4) that the cell surface is partly composed of “ membrane proteins ” seems attractive, but there is no direct evidence for the existence of such proteins, and it is possible that evidence against this hypothesis has been overlooked. Mr.W. D. Stein (King’s College, London) said : I was very interested in the remarks of Prof. Gregor and agree that the copper inhibition that we have been studying may have a basis similar to the inhibitions he found in the synthetic polymer systems. Dr. Adair’s detailed model of a pore will, I am sure, be of much value in our future thinking. I feel that Prof. Teorell’s comments on the spectrum of pore sizes in biological systems provide a stimulating and necessary perspective in which to place the pore model for facilitated diffusion. With regard to the points raised by Dr. Mitchell, I agree that the simple pore model we have been considering here is not sufficient to explain certain properties of facilitated diffusion. We have been considering some extensions of the pore model and I have been tempted to postulate a guard molecule at each end of the pore which controls the selectivity, progress through the interior of the pore occurring along a simple system as in fig.8. Dr. Adair’s model may perhaps be recalled here. Dr. Mitchell asks whether the competitive inhibitors also take advantage of the facilitated diffusion system and penetrate through the pores. We have considered this question but it is one which we are not yet able to answer. The reason is that the competitive inhibitors studied so far, all have higher oillwater partition coefficients than glycerol (see table 1) and hence penetrate the cell membrane far more rapidly than glycerol. Thus, even if they did penetrate by the pore system, the facilitation that would result would make little difference to their already rapid rate of penetration. I have calculated that for even the slowest of these competitive inhibitors (ethylene glycol) the increase in permeability that the facilitated diffusion system would give would be less than 10 % of the measured permeability, which increase could not be detected by our present experimental technique.Thus the competitive inhibitors may or may not make use of the facilitation system. Dr. B. A. Pethica (Cambridge University) said: Dr. J. H. Schulman and I have suggested1 that the breakdown of red cell membranes by detergents is associated with the collapse of a membrane lipoprotein structural component or grouping removed from the membrane when the surface pressure exceeds 34 dyneslcm.Dr. P. Mitchell 2 has reported that butanol causes phosphate leakage from Micrococcus pyogenes above a critical concentration of butanol of about 0-4 M. This is almost identical with the critical concentration for haemolysis and is the concentration of butanol giving a surface pressure of 34 dynes/cm. Dr. T. Mann (Cambridge) has recently shown (private communication) that a similar concentration of butanol causes breakdown of the ram spermatozoa mern- brane. Similarly it is reported by Kaplan 3 that yeast catalase is “ altered” 1 Progress in Biophysics, 1955, 5, 41. 3 J . Gen. Physiol., 1954, 38, 197. 2 J. Cen. Microbiol., 1953, 9, 273.GENERAL DISCUSSION 287 (removed from intracellular surface structures) by a range of non-ionic detergents at concentrations giving surface pressures near 34 dynes/cm.Dr. P. Mitchell has informed me that he has confirmed his findings on the butanol effect on bacterial cell leakage with Micrococcus lysodeikticus, Sarcina lutea and Eschericia coli. With all these various cells, therefore, we have evidence of a non-specific membrane breakdown that may indicate a common basic structure. In view of the important place of cholesterol in the erythrocyte membrane, it should be pointed out that any common structure for all these cell membranes will not involve cholesterol which is not found in most bacteria. Dr. D. Reichenberg (C.R.L., Teddington) said : Any satisfactory theory of ion uptake by biological systems must be capable of accounting for: (i) the ac- cumulation of ions against a concentration gradient: (ii) the kinetics of the process; (iii) the specificity of biological systems for certain ions.been shown, both theoretically 1 9 2 and experimentally 2 that transport of an ion A from a solution of lower concentration to one of higher concentration can readily occur across ion-exchange membranes. The necessary driving energy can be supplied in a number of ways, e.g. from an externally applied electrical potential across the membrane or from the passage of another ion B (of the same sign) across the membrane in the opposite direction from a solution of higher to one of lower concentration. The latter case, that of ion-exchange diffusion, is of particular interest in that inside the membrane itself, the ion A flows down a concentration gradient even though it is passing from a solution of lower to one of higher concentration.The only necessary conditions for this accumulation are (a) that [A1]/[B1] > [A2]/[B2], where 1 and 2 refer to the solutions on the two sides of the membrane. This can obviously be the case even though [All < [Az] and in this event [B2] > [Bl]; (6) that the total concentrations in the solutions should not be too high compared with the fixed charge concentration. This is necessary to ensure that ion-exchange diffusion predominates over " free " difTusion in the membrane. (ii) THE KINETICS OF ION UPTAKE.-The concentrations of A and B at points just inside the membrane surfaces are dependent, as indicated above, on the ratios [A]/[B] in the two solutions in contact with these surfaces, rather than the absolute values of [A] and [B] in these solutions.3 Hence if these ratios [A]/p] in the solutions are kept constant, the concentration gradients of A and B inside the membrane will remain constant, even though we may vary considerably the total concentrations ([A] t [B]).The same will apply, if in one or both of the solutions [A] is very much larger than B (or vice versa), since then the fixed charges at the membrane surface will be saturated with A (or B) ions. Hence the ionic flux will be constant and independent of the total concentration under these conditions. That this in fact occurs has been shown experimentally both with ion-exchange resin beads 49 5 and with ion-exchange membranes.6 There are, however, two limiting conditions: (a) The rate of exchange must be controlled entirely by concentration gradients inside the membrane (or resin particle). This demands that the solution be stirred and that the total concentrations exceed a certain value (which is very low in most relevant cases.5 (b) Excess electrolyte must be excluded from the membrane (or resin) by Donnan effects. This demands that the solution concentration should be appreciably less than the fixed charge (i) ACCUMULATION OF IONS AGAINST A CONCENTRATION GRADIENT.-It has 1 Reichenberg and SutclifFe, Nature, 1954, 174, 1074. 2 Teorell, Transport Processes and Electrical Phenomena in Ionic Membranes in 3 Hale and Reichenberg, Faraday SOC. Discussions, 1949, 9, 79. 4 Reichenberg, J . Amer. Chem. SOC., 1953, 75, 589. 5 Conway, Green and Reichenberg, Trans. Faraday Soc., 1954, 50, 51 1. 6 Neihof, J . Physic. Chem., 1954, 58, 916. Progress Biophysics Biophysical Chem. (Pergamon Press, London), 1953, 3.288 GENERAL DISCUSSION concentration of the membrane. Thus there is both an upper and a Jower con- centration limit for the " constant flux '' condition to hold but the range between is usually fairly wide. A similar independence of total concentration over a range of concentration has been demonstrated in the rate of uptake of ions by certain plants.12 logical systems often show a marked preference for certain ions. Thus Fago- pyrzcrn can take up 40 times as much potassium as sodium from a culture solution containing equal amounts of both ions.3 It is not correct, as has been sometimes alleged, that ion exchange materials are incapable of showing a similar high selectivity. Thus a simple strong base anion-exchange resin (containing quaternary ammonium groupings) with a cross-linking of 2 % DVB prefers perchlorate ion to nitrate ion by a factor of about 20.4 It must be admitted that a very high selectivity as between potassium and sodium has not yet been demonstrated in synthetic ion-exchange materials. However, some progress in this direction has been made and Skogseid 5 has prepared a resin (with a dipicrylamine structure) with a preference factor of 9 in favour of potassium at low potassium loadings. 1 Olsen, Conipt. rend. Carlsberg Skrie Chimique, 1950, 27, 291 ; 1953, 28, 477, 484, 2 Sutcliffe, personal communication. 3 Collander, Plant Physiol., 1941, 16, 691. 4 Reichenberg, unpublished work. 5 Skogseid, Diss. (Norges Tekniske Hogskole, Trondheim, 1946). (iii) THE SPECIFICITY OF BIOLOGICAL SYSTEMS TOWARDS CERTAIN IoNS.-Bio- 488.

ISSN:0366-9033    

DOI:10.1039/DF9562100272

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

288 GENERAL DISCUSSION AUTHOR INDEX * Adair, G. S., 285. Adrian, R. H., 265. Barrer, R. M., 138. Bergsma, F., 61, 125. Bonhoeffer, K. F., 217, 219. Boterenbrood, E. I., 141. Bowyer, F., 251, 273, 283. Danielli, J. F., 239, 279, 280. Despit, A., 121, 150, 204, 205, 207. Eley, D. D., 221, 273. Elton, G. A. H., 205, 212. Finean, J. B., 280. Glueckauf, E., 129, 285. Goldacre, R. T., 274, 275, 284. Gregor, H. P., 162. Hale, D. K., 208. Hedge, D. G., 221, 274. Helfferich, F., 70, $3, 122, 125, 133,213,216. Heller, H., 101. Hermans. J. J.. 141. Hill, T. S., 31,’117. Hills, G. J., 121, 123, 150, 204, 205, 207. Hugo. W. B., 277. Hufchings, D., 192. Jacobs, P. W. M., 198. Kagawa, I., 52, 119. Keynes, R. D., 136, 265, 274, 280. Kitchener, J. A., 207. Kressman, T. R. E., 185, 215. Krishnaswamy, N., 127, 215.Lorimer, (Miss) J. W., 141, 198, 202. MacGillavry, D., 276. McCanley, D. J., 208. Mackie, J. S., 111. Manecke, G., 101, 118, 127, 277. Meares, P., 111, 119, 126, 129, 203. Mitchell, P., 258, 278, 282, 283. Moyle, J., 258. Nagasawa, M., 52, 11 9. Neihof, It., 94, 135, 136. Peers, A. M., 124. Pethica, B. A., 117, 139, 273, 281, 286. Reichenberg, D., 138, 287. Robertson, R. N., 276. Rothstein, A,, 229, 275. Runge, F., 128. Salmon, J. E., 122, 210. Scatchard, G., 27, 70, 117, 118, 138, 203, Schlogl, R., 46, 118, 128, 133, 211. Schmidt, G., 202. Schulman, J. H., 272, 284. Sollner, K., 94, 120, 123, 127, 132, 214. Spiegler, K. S., 174, 199, 213. Staverman, A. J., 61, 125. Stein, W. D., 21 1 , 238, 280, 286. Stock, D. I., 205. Straub, 1. J., 117, 213. Tendeloo, H. J. C., 276. Teorell, T., 9. Tye, F. L., 121 , 128, 200. Ubbelohde, A. R., 137. Vervelde, G. J., 276. Voorspuy, A. J . Z . , 276. Wetstone, D. M., 162. Widdas, W. F., 251, 281. Williams, R. J. P., 123, 192, 216, 280. Wolf, F., 128. Wyllie, M. R. J., 174. Yoest, R. L., 174. 211. * The references in heavy type indicate papers submitted for discussion.

ISSN:0366-9033    

DOI:10.1039/DF9562100288

出版商:RSC    

年代:1956

数据来源: RSC