Protoplasmic Action and Nervous Action

Pure salt solutions (NaCl) or solutions of acid or alkali alter the membrane and allow penetration. Characteristic antagonisms are shown in these effects;^ e.g., the penetration of heavy metal salts (cobalt and nickel) is retarded by CaClz and their toxic action is thus prevented.^ Eggs placed in pure strongly hypertonic NaCl solutions float at first, but soon sink and die, indicating the penetration of salt and water as the membrane is altered; but in the same solution to which CaCL has been added, they may float and remain living for several days.^ Recently Loeb and Cattell have studied the penetration of K salts and acids into the Fundiilus egg in the presence of other salts (alkali and alkah earth) in varying concentrations.'' As index of the penetration of KCl the paralyzing action of the salt on the heart was used; this action is reversible, so that if eggs containing embryos whose hearts have previously been arrested in KCl solution are placed in sea water, the heartbeat after a time revives, indicating outward diffusion of KCl through the membrane. The remarkable fact is that this recovery does not occur in distilled w^ater or in solutions of non-electrolytes; in order that the potassium shalf penetrate the membrane, a treatment of the latter with salt solutions is necessary. Thus (typically) all of the poisoned hearts resume beating within a day when the eggs are placed in sea water or

* Cf. the article by J. Loeb in Oppenheimer's Handhiich der Biochemie, II, 104, for a general account of antagonisms in Fundnlus eggs. isotonic NaCl solutions, while none revive in sugar solution. A ''salt action " is necessary to the penetration of KCl in either direction through the membrane; all salts, however, are not equally effective; in the case of Na salts the effect shows an increase with increase in the valence of the anion. These effects apparently indicate that the penetration of ions through a membrane consisting of colloidal material with which the ions form compounds (proteinates) requires the presence of other free ions with which the penetrating ion can form alternative combinations. Otherwise it is held in position by chemical forces and unable to move freely. Similar conditions have been observed in the diffusion of dyes (neutral red) from stained eggs; this dift'usion is also facihtated by the presence of salts in the outer medium. In order to obtain the maximum diffusionfacilitating salt-effect a certain medium concentration of the effective salt is required; higher concentrations and lower concentrations have a retarding influence on diffusion; this latter effect appears to form the basis of the usual salt-antagonisms or protective effects shown by calcium and other bivalent salts. ^

It is evident that changes in the physical properties of membranes must alter the readiness with which diffusing materials penetrate, and that such changes may result from changes in the chemical composition of the membrane-forming compounds as well as from changes in their distribution or state of subdivision. In a membrane consisting of protein, the formation of proteinates with varying physical properties will change the permeabiHty and the other properties of the membrane. Similarly

if lipoids or other ester-like compounds are membranecomponents, the formation of soaps becomes a possibihty; these vary in their solubilities and watercombining properties; and variations in their proportions, e.g., the substitution of Ca soaps for Na or K soaps, will entail corresponding changes of permeability. The question of whether the protein or the hpoid components of the living plasma membranes are chiefly concerned in the variations of permeability underlying toxic effects or normal physiological processes is an open one ; but in all probability both compounds play a part, although the present evidence seems to favor the view that the proteins are physically the more stable of the two and form the permanent structural substratum, while variations of permeability depend chiefly on changes in the Hpoid components, especially the soaps. Parallels to the physiological salt antagonisms are wxll known in non-Hving systems containing both classes of compounds. Of chief biological interest is the antagonism between Na salts and Ca salts, which is apparently universal in Kving organisms. Protein systems containing salts of these two metals show variations in properties when the proportions of the salts are varied; such properties as water-combining power (swelling or hydration), viscosity, osmotic pressure, susceptibility to alteration by organic compounds (precipitation by alcohol), and electrical polarization of particles are affected; Loeb's recent investigations afford instances of salt-antagonisms affecting all of these properties.^ The inference from such facts would be that Ca and Na

proteinates must co-exist in certain definite proportions in the cell-structures, including the membranes, if certain biologically necessary physical properties are to be preserved. Substitution of Na for Ca compounds would occur in pure Na salt solutions, with resulting structural changes which might well be injurious or fatal. Clowes^ has recently shown that salt-antagonisms having an even closer resemblance to biological salt antagonisms may be demonstrated in oilwater emulsion systems by varying the proportions of Na and Ca soaps in the interfacial films. The number of separate drops formed when a slightly alkaline salt solution is allowed to flow through a stalagmometer into oHve oil is found to vary in a remarkable manner with variations in the proportions of NaCl and CaCla in the solution; thus with an alkaUnity of .ooin NaOH, the following number of drops per minute were observed with different solutions :

The pure NaCl solution greatly promotes the tendency toward fine subdivision of the drops; the CaCL decreases this tendency; at a certain ratio of the two salts (Na: Ca about 50 or 100:1) the two tendencies counteract each other. The result depends on the different solubilities of the Na-oleate and the Ca-oleate in the two phases, the former being soluble in water but not in oil, the latter soluble in oil but not in water. The surface-tension conditions at the interface are accordingly oppositely affected by the two salts, with correspondingly opposite effects on the state of dispersion; at a certain ratio of concentrations the two opposite effects are balanced. Clowes shows that when equal volumes of oil and alkaline salt solution (mixtures of n/io Na OH and n/io CslCU) are shaken together, the effect varies according to the proportions of Na and Ca in the solution; when Na is present in excess, the oil is dispersed as droplets in a continuous water phase, while when Ca is in excess the water is dispersed as droplets in a continuous oil phase. A reversal of phase-relations may thus be accomplished in an emulsion system by changing the salt-content; and this conclusion has highly important biological applications, since it bears closely on the problem of the relations between the lipoid and the aqueous components in the living protoplasmic system. Any system in which the oil (organic solvent or lipoid) phase is the continuous one is permeable to oil-soluble substances, but not to water-soluble substances which are oil-insoluble; and the general correspondence of this condition with that observed in living protoplasm suggests the possibility that the external layer of the living plasma membrane consists (at least during the greater part of its existence) of a continuous layer of lipoid material, the continuous condition depending on the presence of compounds with properties like those of Ca soaps. The importance of Ca to the semi-permeability and water-resisting properties

of the protoplasmic surface-film, on this hypothesis, is evident. The possibility of a reversibility of phase-relations under salt action^ implies the possibility of inducing reversible changes of permeability under these conditions, and such reversible changes are, as we have seen, a characteristic feature of the living plasma membranes. Clowes has in fact constructed a model in which an emulsion consisting of equal volumes of oil and a mixed salt solution (containing NaCl, KCl, and CaCla in proportions similar to those of sea water and slightly alkaline) is supported in the interstices of a paper partition (sheets of filter paper) fixed in position by rings of rubber in the interior of a U-tube.^ The electrical conductivity of the emulsion-permeated paper is then found to vary, when the solution in contact with it is changed, in a manner resembling in general that shown by the partition of living plant-tissue in Osterhout's experiments. In pure NaCl solution the conductivity is rapidly increased; in pure CaCla it is decreased; and the changes of conductivity are reversible if they are not allowed to proceed too far. It is assumed that in the capillary interstices the emulsion undergoes reversible changes of phase of the above-described kind, the conversion of the

' Clowes, Proceedings of the Society of Experimental Biology and Medicine, XV (1918), 108. oil from the continuous to the discontinuous phase corresponding to an increase in conductivity and vice versa. It will be remembered that water-insoluble colloidal material (Cua FeCye) held in the interstices of a supporting structure forms the essential composition of the semipermeable membranes used in the osmotic pressure determinations of Pfeffer and Morse; in such membranes, also, permeability is changed by the action of salts. The conditions in living membranes, while probably similar, in the above broad sense, to those in such structurally composite membranes, are vastly more complex, chiefly on account of the constant presence of the metabolic factor. Clowes propounds the general hypothesis that ^'variations in the permeability of the protoplasmic membranes are attributable to the action of electrolytes and metabolic products on delicately balanced interfacial soap-films and emulsion systems, and that proteins may play no part in the valve-like mechanism controlling permeability other than to afford a supporting filamentous or mesh-like structure."^

Such an arrangement need not be taken too literally as an exact model of the conditions in the protoplasmic surface-films. It seems probable, however, that in the plasma membrane there is a combination of a relatively permanent supporting structure, presumably protein, with a variable emulsion-like component whose waterinsoluble phase is normally so disposed as to block, with a considerable degree of completeness, the interstitial spaces or capillary channels of the membrane. Variability in this emulsion, under the influence of salt solu-

tions or other factors, involves variability in the permeability of the partition; and as a secondary consequence other conditions depending on that permeability, such as electrical polarization, are affected. Such a conception implies that changes in the supporting protein structure may also influence permeability; but it regards the normal or physiological variations in this property as dependent chiefly on variations in the state of the most external water-insoluble or lipoid portion of the protoplasmic complex. Such a conception allows for a wide range of variability in the permeability of the membrane. The latter is not to be regarded as a continuous lipoid sheet under one set of conditions, which changes without transition, under other conditions, into a state in which the lipoid becomes the discontinuous and the aqueous phase the continuous phase. A balanced state, with fluctuations on either side of a mean, which is regulatively maintained by metabolic processes, is rather the one to be conceived as representing the condition actually existing during life.

A simple case of salt-antagonism, which I have recently studied in the starfish egg,^ appears to throw light upon the more specifically chemical conditions of these phenomena. The fully mature starfish egg (ca. 1 60 /x in diameter) is surrounded with a layer of jelly-like substance, of 15 to 20 fj, in diameter, consisting of water-swollen material (of undetermined nature) separated or secreted from the egg-protoplasm. This layer is rendered visible by mounting the eggs on a slide, with cover-glass, in a suspension of India ink, to which the jelly is impermeable; it then appears under the micro-

scope as a clear halo surrounding each egg. When the egg is washed in pure isotonic NaCl solution (0.54 m), preferably with the aid of centrifuging, the jelly swells and dissolves; but in NaCl solution containing a little several other metals (Mg, Mn, Co, Ni, Al) have been found to have a similar effect.' The jelly layer, in the presence of the salts of sea water and in mixtures of NaCl and CaCU, thus possesses a certain waterinsolubility and physical consistency; these properties are apparently dependent on the presence of a calcium compound (possibly proteinate) with definite physical properties (especially water-insolubility and lack of tendency to swell) which preserves the characteristic structure and consistency of the whole layer. In pure NaCl solution this constituent is replaced by the corresponding sodium compound which is water-soluble and swells readily, hence the coherence and insolubility of the layer as a whole are lost.

This process may be regarded as a model of the essential kind of change occurring in the plasma membranes of living cells in pure NaCl solution. The surface of the starfish egg is in fact physically altered by NaCl solution in a characteristic manner; unfertilized eggs placed in the pure solution cohere in clumps or agglutinate; many eggs also form fertilization-membranes (when returned to sea water) and show evidence of partial activation by cleaving and in some cases developing to the blastula stage. All of these effects of the pure

^ Unpublished experiments in the Nela Research Laboratory at the Marine Biological Laboratory at Woods Hole. solution are prevented in the presence of a little CaCla. The toxic action of the pure solution and also its membrane-forming action are correlated with a permeability-increasing action, as in the other cases cited above; and the correspondence with the behavior of the jelly seems to imply that the increase of permeability in the pure solution is to be referred also to the replacement of water-insoluble Ca compounds (e.g., Ca proteinates or soaps), on which the properties of the plasma membrane depend, by soluble Na compounds. There is much independent evidence that this role of calcium compounds — i.e., of determining the properties of the surface layers of cells — is a general one; we may thus understand the importance of Ca to all normal cell-processes (stimulation, etc.) which depend on changes in the surface layers. The peculiar relation of Ca to the coherence of blastomeres and of plant cells has already been mentioned; according to the earlier work of Mangin, Ca compounds (''Ca-pectate") in the middle lamella of plant cells are essential to the structural coherence of cellular tissues; when the Ca is replaced by Na, the cells tend to fall apart. Similar phenomena are also well known in the epithelial tissues of animals; e.g., in the ciliated epithelium of Mitylus the cells swell and fall apart in pure solutions of many Na and K salts, and this effect is prevented by Ca.^ Recently the changes occurring in plant tissues in pure NaCl solutions have been studied in much detail by Hansteen.* The

general effects are swelling, loss of consistency or turgor, and disintegration of the tissue, accompanied by loosening of intercellular coherence; these effects are all prevented or greatly decreased in the presence of a little Ca. Hansteen believes that water-insoluble Ca compounds are essential constituents of living protoplasm in general, and that they are present especially in the protoplasmic surface layers and other solid structures; he also cites evidence indicating that these compounds are lipoid in nature. According to his conception, pure alkali salt solutions attack the surface layers of cells because they alter the water-insoluble Ca-lipoid compounds there present, converting them into watersoluble compounds and secondarily inducing absorption of water and structural breakdown. Such effects are apparently of the same nature as the disintegration of the cilia of marine animals (Arenicola) and the general loss of semi-permeability of plant and animal cells in pure salt solutions, already described. They are consistent with the general view that in the formation of the solid or permanent (water-insoluble) protoplasmic structures, Ca compounds play an essential part. Hansteen found that the presence of more than the normal concentration of Ca salts in culture solutions favored profuse branching and the formation of an abundance of root hairs in seedlings; Wiechmann, in Hober's laboratory, has recently confirmed this result, and has found further that Sr, Ba, and a few heavy metals (Mn, Ni, Co) act similarly to Ca, while Mg is ineffective.^ It would seem that certain necessary physical properties of protoplasmic structures, such as rigidity, water- ^ Wiechmann, Arch. ges. Physiol., CLXXXII (1920), 99.

insolubility, and impermeability to water-soluble substances (like sugar and salts), require the presence of Ca compounds. These compounds impart the necessary structural stability to the whole protoplasmic complex. A fact of interest in relation to this question is Meigs's recent observation that an approach to semi-permeability can be imparted to certain artificial colloidal membranes, otherwise highly permeable, by depositing insoluble Ca salts (phosphate) in their substance.^

These facts and considerations are consistent with the view that Ca compounds, e.g., Ca soaps, play a similar part in the surface layers of protoplasm, and the recent interesting experiments of Clark^ with the frog's heart lend support to this general conception. He finds that the heart, after being weakened by prolonged perfusion with Ringer's solution, rapidly regains its vigor if perfused with Ringer's solution to which serum, serum-Hpoids, lecithin, or soaps of higher fatty acids, have been added. During perfusion with pure Ringer's solution the heart loses to the solution some material which has a similar reviving action when perfused through other exhausted hearts. In order that these substances, or soap, should exhibit this beneficial action, calcium must be present. He concludes that the beneficial action of the soaps is associated. with the adsorption of a water-soluble Ca soap or similar compound upon the surface of the muscle cells, and puts forward the hypothesis ''that the activity of the heart is dependent upon the semi-permeabihty of the cell to electrolytes, that this is

dependent on the presence of Ca and lipoids at the surface of the cells, and that during perfusion the heart loses Hpoids and becomes more permeable to electrolytes." This conception of semi-permeability as dependent on the presence of lipoids, and of the state of the Hpoids as determined by the salts of the medium, is in harmony with much recent work.^ It also affords a point of view from which it is possible to understand why the physiological effects of salts and of lipoidsolvent compounds should have so much in common.

^Blackman, "The Plasmatic Membrane and Its Organization," New Phytologist, XI (191 2), 180; Clowes, loc. cit; Czapek, Oberfldchenspannung der Plasmahaut, Jena (1911); McDougall, Science, LV (1922), 653; Hansteen-Cranner, loc cit. See also Stiles's review of the subject of cell permeability in New Phytologist, XX, XXI, XXII (1921-23). From the physiological point of view the reversible forms of salt action are the important ones; the properties and activities of living protoplasm may thus be modified by changing the salt-content of the medium, and return to the normal when the original salt-content is restored. Such reversible effects are of special biological interest, since their essential conditions are in all likelihood similar to those controlling the normal variations of activity. Substances are continually being formed in metabolism (e.g., CO2 and other acids) which directly influence protoplasmic action. It is, therefore, of fundamental interest to note the existence of another large class of substances, many of which are chemically indifferent, i.e., not readily oxidized or reduced (hydrocarbons and their substitution-products), which have a profound influence on protoplasm, completely reversible within wide limits. These substances are those organic compounds, varying widely in their chemical nature, which have in common two general physical properties : (i) a solvent action on, or solubiHty in, the water-insoluble organic constituents of protoplasm (fats, lipoids, etc.); and (2) a high degree of surface-activity, i.e., influence on the surface-tension at the boundary between water and non-aqueous phases. These compounds appear also to produce their physiological effects by altering the

structural substratum of protoplasm in a manner which does not permanently change its properties or physical state. Hence in their presence physiological processes are modified temporarily in rate or character, and resume their former conditions when the substance is removed. The special affinity of these compounds for substances having fat-like properties indicates that their primary action is on the lipoid constituents of protoplasm; their possible action on proteins, however, is also to be considered. In general they include the substances comprised in Overton's first group (alcohols, ethers, esters, normal and substituted hydrocarbons, etc.).

We distinguish, therefore, two chief groups of compounds which by means of their reversible influence on the structural substratum of protoplasm may modify vital processes without affecting them permanently or injuriously: (a) neutral salts or other electrolytes (acid and alkali), and (b) lipoid-solvent or surface-active organic compounds. These two groups may be characterized respectively as general colloid-alterants and lipoid-alterants. The compounds of these groups differ somewhat sharply in their physiological action from those compounds whose chemical effects tend to be irreversible; the latter include most of the strong oxidizing and reducing agents and the salts of heavy metals; usually these are not capable of modifying physiological processes without permanent injury; hence they are toxic or poisonous in small doses. Recovery from the effects of this '' poisonous" group depends upon the reparative activity of the living protoplasm, just as does recovery from mechanical injury, and not upon a simple reversal of the chemical or other action of the compound.

The most remarkable general physiological effect produced by the lipoid-alterant substances is a reversible suppression of irritability or spontaneous activity. This effect always appears in certain definite, not too high, concentrations of these compounds, and constitutes the phenomenon of narcosis or anaesthesia, which is universal in living matter.^ The power of inducing this state seems to be independent of the special chemical nature of the narcotizing compound; evidently this power is connected in some manner with the general physical properties just named; and the question first arises whether the lipoid-solubility of these compounds or their surface-activity is the property primarily responsible for this characteristic action.

The existence of a relation between the solubility of chemical compounds in fats and their narcotic action was early noted, first by Bibra and Harless in 1847, ^^^ later by Claude Bernard, Hermann, Richet, Ehrlich, and others.^ Richet propounded the rule that any narcotizing compound has the stronger action as a narcotic the lower its solubility in water. This is similar to the rule of Overton and^ Meyer that the narcotic effectiveness of a compound runs parallel with

^"We may say that everything living is sensitive and can be anaesthetized; whatever is not sensitive is not living and cannot be anaesthetized." — Claude Bernard, address, "La Sensibilite" (delivered in 1876), published in his book. La Science Experimentale, Paris (1890). ^ Cf. Overton, Studien iiber die Narkose, Jena (i 901), for a historical account of the earlier work on narcosis. its oil-water partition-coefficient, a relation first studied systematically by these investigators.'

In his investigations on permeability, Overton had reached the conclusion that solubility in lipoids was the chief factor determining the entrance of compounds into cells; such readily penetrating compounds belong for the most part to the narcotizing group; and a detailed study of the phenomena of narcosis in tadpoles demonstrated a close parallelism between the relative solubilities of a large number of organic compounds in oil and water and their narcotizing action.^ The ratio according to which any compound is distributed between these solvents (when the solvents are in contact and equilibrium is reached) is a measure of its relative solubility in the two; this ratio is known as the ^'partitioncoefficient." In the early members of any homologous series of compounds the ratio of oil-solubility to watersolubility increases progressively as the molecular weight increases, and the same is true of the narcotizing properties of the compounds. For example, with the ethyl esters of the first five fatty acids, Overton found the concentrations required for the complete narcosis of tadpoles to be as indicated in the table (p. 191).

The narcotic effectiveness of the ester increases regularly as its water-solubility decreases; and in general each member of the series is from two to three times as effective as its immediate predecessor. Relations of a similar kind were found with other series, including hydrocarbons, alcohols, aldehydes, ketones, ethers, and various substituted compounds. Any increase in the oil-water partition-coefficient was associated with increase in narcotic effectiveness. Overton accordingly drew the conclusion that v the narcotics act by dissolving in certain oil-like or fatty constituents of the irritable cells (in this case nerve cells); and he identified these constituents with the lipoids, especially lecithin and cholesterol, which appear to be always present in protoplasm. The essential determining condition of anaesthesia, according to his view, is the solution

in all parts oil Ethyl valerianate. . . .0019 m In 500 parts water; of the narcotic compound in these cell constituents; when the hpoids are charged or impregnated with the compound, they undergo a change of physical properties, entailing corresponding alterations in the irritability of the cell. Meyer drew independently^ a similar conclusion;' he pointed out that the narcotizability of cells seems to be related to the nature and proportion of the lipoids present in the protoplasm; e.g., the high susceptibility of nerve cells to narcosis is a correlative of their high hpoid-content. Different narcotics act unequally because they are distributed in unequal ratios between the

lipoid and the aqueous phases of protoplasm; in general, the greater the relative lipoid-solubility, the larger the proportion of the anagsthetic compound which is in solution in the lipoids when the partition-equilibrium is reached. Hence when a compound has a very high Hpoid-solubiHty it may exert narcotic action in extremely dilute solution; phenanthrene, for example, was found by Overton to narcotize tadpoles in dilutions of one part in 1,500,000 of water.

The general conception known as the '^Overton- Meyer theory of narcosis" may be defined as follows. The solubility of narcotizing compounds in the cellhpoids forms the basis of their narcotic and presumably other pharmacological properties. By dissolving in the lipoids, such compounds alter the physical properties of these essential components of the protoplasmic system, and hence all properties of the system, especially irritability, which are dependent on the state of the Hpoids. Since simple solution without chemical combination is the basis of this effect, the latter is readily reversed by allowing the compounds to diffuse away.

The general conclusion that selective solubiHty is the essential basis of narcotic action does not, however, necessarily follow, since the same reasoning would apply to other physical effects which are reversible under similar conditions; e.g., effects dependent on surfaceactivity, involving a lowering of surface-tension at the protoplasmic phase-boundaries and a concentration of the narcotizing compound at these surfaces. Probably, however, the case is not one of alternatives; if lipoids are present in the system, any substances which are soluble in these com_ponents must inevitably dissolve