Protoplasmic Action and Nervous Action

^ For a recent account of these phenomena, see the review by Seifriz, Annals of Botany, LXXXVIII (192 1), 269; cf. also Botanical Gazette, at the surfaces of cuts made with microdissection needles, or around isolated portions of protoplasm.^ Apparently the nearest inorganic analogies to these protoplasmic film-structures are the thin surface-films deposited at the boundaries between mutually insoluble liquids, one or other of which contains surface-active (especially colloidal) materials in solution; good illustrations are the films of soap or other material surrounding the droplets in an oil-water or other emulsion, or the solid films formed about globules of oil, mercury, chloroform, or other insoluble liquids immersed in proteincontaining solutions, or the haptogen membranes formed at the surfaces of warm soap solution, milk, or solutions of protein, peptone, saponin or other surface-active colloidal substances. Under certain conditions these films may acquire a solid consistency and exhibit considerable structural density, and thus limit diffusion between the two phases; the resulting system may then be described as triphasic, the three phases being the internal medium, the external medium, and the intervening thin phase or membrane. The ''artificial cells" described by Harvey,^ made by breaking up a chloroform solution of lecithin in dilute egg-albumin, may be cited as examples of structures formed under these conditions.

In living protoplasm it is to be presumed that condensation-films of protein and other surface-active substances are formed at all of the surfaces bounding ^ Chambers, American Journal of Physiology, XLIII (19 17), i; Proceedings of the Society of Experimental Biology and Medicine, XVII (1919), 41; Jour. Gen. Physiol., V (1922), 189. For plant cells cf. Prowazek, Biol. Zentr., XXVII (1907), 737. Drops of protoplasm from Vaucheria form membranes about their surfaces.

the different phases; and the physical resemblances between protoplasm and emulsions are apparently referable to the presence of these surface-films. Both systems are examples of what may be called filmpervaded or film-partitioned systems. By the formation of these films a certain structure is imparted to the whole system; this structure is largely the expression of surface-forces, in which chemical, electrical, and mechanical (surface-tension) factors all enter.

It should be noted that according to this conception of protoplasmic structure no essential distinction is to be drawn between the intracellular surface-films or membranes (alveolar membranes, vacuole membranes, nuclear membranes) and the surface-films inclosing entire cells (plasma membranes). In plant cells it can be shown experimentally that vacuole membranes and plasma membranes are similar in osmotic properties, and Hamburger has shown the same for the nuclear membranes and plasma membranes of animal cells.'

In forming a general conception of the physicochemical characteristics of protoplasmic membranes, the properties of colloidal gels, especially in their relation to diffusion-processes, may be taken as a starting-point. In a gel, i.e., a solid mixture of colloidal material (such as gelatine) and water, diffusion is hindered only slightly if the concentration of the colloid is low. When the gel is made denser — as the proportion of water is decreased — diffusion becomes slower and more restricted. At a sufficiently high density certain solutes, especially colloids, can no longer diffuse through the gel, although water and cr^-stalloids may still pass. If the density

be still further increased, crystalloids and finally even water fail to pass/ This last condition is exemplified in water-proof organic membranes like the external skin of most animals and the membranes of certain eggs (the Fundulus ^gg). From an elementary or purely physical point of view a membrane may be regarded as essentially a thin sheet consisting of a gel of the kind above. Such a gel has a large surface-area in proportion to its total volume, and by virtue of its diffusion-hindering property it prevents or retards the transfer of material (colloidal particles, molecules, ions) between the two solutions which it separates.

Connected with this diffusion-hindering property, which conditions the rate of interchange and hence the rate of chemical activity at the surface between the solutions separated, are certain electrical properties C' membrane-potentials"), resulting from the influence of the membrane on the distribution and transfer of ions between the two solutions.^ These properties are apparently of fundamental importance to the bioelectric processes, and their conditions will be considered more fully later.

Independently of its structural density, a membrane of complex chemical composition may exhibit a selective ^ The conditions may be compared with those presented by a series of "ultra-filters" of graded densities, as described by Bechhold {Colloids in Biology and Medicine); the permeability decreases as the density increases. 2 Cf. Lewis, A System of Physical Chemistry, II (1920), 320, for an account of membrane potentials. The type of equilibrium investigated by Donnan, in which solutions are separated by a membrane which is impermeable to some but not all of the ions, plays an important part in many membrane processes, as shown especially by Loeb in his recent work. Cf. Proteins and the Theory of Colloidal Behavior, Part 2.

permeability with reference to substances which are unequally soluble in its different chemical components. Substances soluble in a given component (lipoid-soluble substances) may thus pass a protoplasmic membrane, while chemically similar but lipoid-insoluble substances may not. Overton's experiments on the differences between the rates of diffusion of various organic compounds and salts into living cells illustrate selective permeability of this type.^

It is important to recognize that the plasma membrane is not a dead structure or a purely passive partition, but represents in reality a portion of the living protoplasm, characteristically modified in its structure and physical properties by surface conditions.^ Hence it is the seat of metabolic and other activities which influence its physical properties. Evidently it is that part of the cell which comes into the most direct relations with the surroundings. Hence many changes in the surroundings influence the cell primarily through their action on the plasma membrane, and there is evidence that in irritable cells this structure plays the part of a specially sensitive and reactive intermediary between the living protoplasm and the external world; it thus exerts a far-reaching control over the metabolic and other processes occurring in the cell-interior. The relations of the plasma membrane to stimulation will be considered in detail later.

* Cf. my paper in American Journal of Physiology, XLV (19 18), 406, for a more complete discussion of this phase of the problem of permeability. The normal semi-permeability of the plasma membrane is a function of the living state of the cell, i.e., is dependent upon the continuance of the normal constructive metabolism. When metabolism ceases, as at death, the membrane soon loses its insulating or semi-permeable properties, and free interchange of diffusible substances then occurs between the cell and the medium. The loss of turgor in plant cells after death is the most familiar example of this type of effect; leaves and other parenchymatous parts then wilt because of the diffusion of the osmotically active substances through the now permeable plasma membranes. The osmotic tension which during life keeps the cell walls in their normal stretched and rigid condition disappears, the tissue becomes soft and flaccid, and the protoplasm shows other evidences of increased permeability (increased electrical conductivity, loss of diffusible materials to the surroundings, ready entrance of dyes and other substances). Similarly in animal cells various substances, such as pigments and other compounds, normally confined within the protoplasm, diffuse rapidly into the surrounding medium on death, and the plasma membrane admits substances such as alkalis and salts, to which previously it was impermeable.

It is a remarkable and apparently paradoxical fact that the living plasma membranes usually show themselves highly impermeable to many dissolved substances which are essential to the cell as foods or otherwise (sugars, amino-acids, and neutral salts), and some general explanation of this peculiarity seems required. If we were to express the matter teleologically we might say that the advantage to the cell consists in the insulation of the living protoplasm from its environment,

which has an entirely different composition. The maintenance of the normal vital properties requires that the essential diffusible constituents of protoplasm should not be lost to the surroundings; it is also evident that too ready an entrance of substances from the outside would interfere with the stability of protoplasmic composition. The presence of a semi-permeable boundary layer appears thus to be a necessary condition for the preservation of the normal chemical organization of the cell. It is readily seen, for example, that the existence of a simple diffusion equilibrium between surrounding medium and protoplasm would prevent the latter from acquiring the special crystalloidal content which is characteristic of it. Hence, living cells are enabled to survive and develop largely by virtue of being inclosed by surface-films which are impermeable to crystalloidal compounds of the foregoing classes.

But since these compounds do in fact gain entrance to the cell, at least at certain times, it is clear that the problem of cell-permeability is not a simple one. Apparently we must conclude that the entrance or exit of substances by simple diffusion is in most cases a different phenomenon from their entrance or exit under physiological conditions. The processes of absorption and secretion are in fact special activities, requiring the performance of work by the cell. The distinction between a passive or purely physical permeability and an active or physiological permeability thus seems a necessary one.

The conditions of passive permeability are of interest chiefly because of the light which they throw upon the physical and chemical nature of the substances composing the plasma membranes. The most significant general fact is that apparently all substances with solubilities or solvent properties characteristic of organic compounds (rather than of water-soluble compounds or water) enter cells with special readiness. A relation between the presence of organic solvents in protoplasm and the permeability of the plasma membrane is thus indicated. Overton, who first investigated in detail this connection between organic solubility and power of penetration, drew the conclusion that the plasma membrane consisted essentially of the so-called "lipoid" compounds, especially lecithin and cholesterol, which are universally present in protoplasm. He showed that all members of homologous series, such as alcohols, ethers, esters, normal and substituted hydrocarbons, ketones, aldehydes, amides, and similar compounds, readily enter living cells; such compounds dissolve, or are dissolved by, the lipoids or solutions of lipoids in organic solvents; and their ready entrance is a result of this solubility. On the other hand, sugars and polyatomic alcohols (pentites, hexites, etc.), with molecules containing many hydroxyl groups, are highly soluble in water, but not in lipoids, and do not enter cells readily. In Overton's original experiments, acid and basic dyes also showed a relation between lipoid-solubility and power of penetration; but the conditions are complex in this case and many exceptions to this rule are now known. ^ In the case of neutral salts of alkali and alkali earth metals (especially Na, K, and Ca) there is also Kttle or no evidence of penetra-

^Cf. Ruhland, Jahrh. wiss. Botanik, XL VI (1908), i; cf. also Hober's discussion, Physikalische Chemie der Zelle und der Gewebe, pp. 426 ff. tion in balanced solutions, a fact corresponding to the lipoid-insolubility of these compounds. In pure solutions (pure NaCl solution) the salts alter the properties of the plasma membranes and secondarily may penetrate;^ but this fact is in no way inconsistent with Overton's view, which applies to the unaltered membrane. A definite correlation between lipoid-solubility and power of penetrating the living plasma membrane may be said to have been established by Overton's work and succeeding studies of the same kind; and this generalization is an important one, since it indicates (as do many other facts) that the lipoids play an essential part, apparently in association with the other chief colloidal compounds of protoplasm, the proteins, in the formation of membranes and probably of the other solid structural elements of cells.

Three chief methods have been employed in determining the permeability of cells to dissolved substances: (i) the plasmolytic or osmotic method,^ (2) the partition method,^ and (3) the electrical conductivity method.'* To these may be added methods dependent on the use of indicators, either those normally present in cells,^ or dyes like neutral red^ which may be introduced from outside. These methods are especially valuable in studying the permeability to acids or alkalis. Permeability to dyes may usually be studied by direct observation.

The method of plasmolysis, first employed systematically by Overton, is based on the production of osmotic effects (entrance or exit of water) when the cell is placed in solutions having a different osmotic pressure from that of the cell-contents (aniso tonic solutions). In hypertonic solutions of substances which do not readily traverse the plasma membrane the cell shrinks, in hypotonic solutions it swells, while in isotonic solutions its volume remains unchanged. The problem is to determine the relative degree of permeability to different substances, some of which may traverse the membrane, but with unequal readiness. To do this the behavior of the cell is observed in a hypertonic solution of the substance under examination. It is clear that when the dissolved substance in the external medium is quite unable to penetrate the membrane, it exerts pressure against the latter (by the continued impacts of the molecules), and since this pressure is greater than that exerted by the dissolved molecules within the cell, water is extracted (or expressed) from the latter until a permanent state of equilibrium is reached in which the osmotic pressure of the cell-contents equals that of the surrounding solution. The cell is then permanently shrunken. When, however, the dissolved substance

penetrates gradually, its water-abstracting effect is only temporary, since its concentration, at first greater outside than inside the cell, is eventually equalized by difi'usion; the cell then tends to resume its original water-content. In the case of a solute which penetrates rapidly (with a readiness like that of water) no effective inwardly directed pressure can be exerted against the membrane and no osmotic effect is produced. Using plant cells (Spirogyra and others) and moderately hypertonic solutions of various substances, Overton found that in solutions of sugars, amino-acids, and neutral salts the plasmolysis was permanent; in solutions of glycerine, glycol, urea, and similar compounds the degree of plasmolysis was less than in sugar solutions, and after the initial shrinkage, water gradually reentered the cell; while in solutions of alcohols and many other organic substances (of the same osmotic pressure as the effective sugar solutions) plasmolysis was entirely absent. Similar differences were typical of a large number of other compounds. He therefore divided soluble substances into three groups, according to their ability to penetrate the living plasma membrane: (i) Those to which the plasma membrane is completely or nearly impermeable, including sugars, polyatomic alcohols (from erythrite up), soluble amino-acids, neutral salts of alkali and alkali earth metals; (2) those which penetrate the membrane, but slowly and with var^dng degrees of resistance, including glycol, glycerol, and certain amides such as urea; and (3) those which enter without encountering any evident resistance; here belong a variety of organic compounds of the groups cited above. These general conditions were found by

Overton and other investigators to be characteristic both of animal cells (blood corpuscles, muscle cells, egg cells, etc.) and plant cells of the most varied kinds. In the partition method, as used by Hedin and others, the distribution of dissolved substances between the cells and the solution is measured directly (by cryoscopic determinations), and its results agree closely with those of the plasmolytic method. In general it has been found that the above-cited conditions of permeability are highly characteristic if not universal in living protoplasm.^

Since in general the lipoid-soluble substances which penetrate living protoplasm are also highly surfaceactive, the conclusion may be drawn that either lipoidsolubiKty or surface-activity (or both) is a property favorable to penetration (Overton, Traube). The penetration of one substance through another may depend on mutual solubility; the cases of the rubber membranes used in Flusin's experiments, the watersoaked bladder partitions employed by Nernst, and the palladium partitions of Ramsay's experiments with nitrogen and hydrogen may be cited as illustrations.^ Overton explains the permeability of the plasma membrane to lipoid-soluble substances as an expression of the solubility of these substances in the lipoids of the membrane; in general he finds a parallelism between the lipoid-water partition-ratio of a given substance and its ability to penetrate cells; this is illustrated by the behavior of substitution-products and of the members of

^ Cf. Hamburger's Osmotischer Druck ti. lonenlehre for an account of comparative investigations in this field. homologous series. The contrast between NH4OH (lipoid-soluble) and KOH (Hpoid-insoluble) in their abiHty to enter Hving cells is a striking one;^ and most investigators who have studied the penetration of substances (like acids and bases) whose behavior can be determined with accuracy, have found a general parallelism between the rate of penetration and the lipoidsolubility of the various compounds. This parallelism, however, is not exact, indicating the presence of other factors.^

Impermeability to neutral salts in balanced solution is an especially important property of the plasma membrane, since it renders possible permanent differences of salt-content between protoplasm and surroundings. This means differences of ionic content, and is probably a condition of the normal electrical polarization (''physiological polarization") of the membrane, since any partition separating two solutions of different ion-content is typically the seat of an electrical potential difference. The results of the mineral analysis of cells, as well as those of the plasmolytic and partition methods just described, show clearly the inability of neutral salts to

* Cf. Warburg, loc. cit.; Harvey, loc. cit.; Gray, Proceedings of the Royal Society, B, XCIII (1922), 104 (cf. p. no); Jacobs, Jour. Gen. Physiol., V (1922), 181. A similar contrast exists between organic acids (lipoid-soluble) and strong acids; cf. Loeb, Biochem. Zeitschrift, XV (1909), 254; Bethe, loc. cit.; Gray, loc, cit. Jacobs' results especially show the extraordinary ease with which carbon dioxide and ammonia penetrate Hving protoplasm. 'Harvey, Intern. Z. physik. chem. Biol., I (1914), 463; Crozier, loc. cit. Miss Collett's studies on the toxicity of acids to infusoria show a similar general relationship between organic solubility and toxicity {Journal Experimental Zoology, XXIX [19 19], 443, and XXXIV [192 1],

penetrate the unaltered living cell. Paine^ has shown experimentally that NaCl, (NH4)2S04, and Na2HP04 in n/io solution do not enter yeast cells appreciably even after hours of immersion; and numerous analyses have shown that the specific salt-content of many living cells (blood corpuscles, muscle, etc.) is quantitatively or even qualitatively entirely different from that of the medium. For example, sodium salts seem to be almost completely absent from vertebrate muscle cells.^ This result seems incompatible with more than a very limited permeability of the plasma membrane to these substances. Either the salts do not diffuse across the membrane, or some active physiological factor is at work which opposes or compensates the effect of diffusion and maintains the salt content of the protoplasm at a certain norm. The nature and proportion of the salts present in any species of cell are characteristic of that cell and apparently represent a constant feature of its chemical organization; this is illustrated, for example, in the analyses of the salt-content of mammalian blood corpuscles by Abderhalden and others.^ In the corpuscles of the horse, pig, and rabbit there is little or no Na and an abundance of K; in the ox, sheep, goat, dog, cat, the amount of Na is greater than that of K. Inorganic phosphates are always much more concentrated in the corpuscles than in the serum, which is always rich in Na and poor in K. Voluntary muscle cells are rich in K salts and phosphates and poor in Na salts.

Hober {op. cit., p. 371) cites certain observations of Warburg made in 1911^ indicating that the erythrocytes of the goose are influenced in their oxygenconsumption by alkaH-earth salts only after the plasma membrane has been destroyed by freezing and thawing. Apparently these salts cannot pass the intact plasma membrane. On the other hand, alcohols and urethanes check oxidations in the intact erythrocytes; these substances can penetrate. It may be noted that these observations are consistent with the view that the nuclear surface is a chief factor in the oxidation-processes of these cells.^

The relative or complete impermeability of blood corpuscles to the ions of the surrounding salt solution is also indicated by the low electrical conductivity of these cells. Stewart, Tangl, Bugarszky, and others have shown that the electrical conductivity of blood is due almost entirely to the plasma. Low electrical conductivity is in fact now known to be a highly constant and characteristic peculiarity of cells during life, and the evidence indicates that the high resistance is chiefly if not entirely a property of the plasma membranes. The conductivity of the cell as a whole appears to vary directly with the permeability of the plasma membrane to crystalloidal solutions. All conditions that increase general permeability (action of cytolytic substances or unbalanced salt solutions or of poisons, high temperatures, or other lethal agents) also increase electrical conductivity. According to Osterhout, the most exact

'In the nucleated erythrocytes of the frog, the indophenol test shows active oxidation at the nuclear surface; cf. R. S. Lillie, Journal of Biological Chemistry, XV (1913), 237. measurements of permeability are those given by electrical conductivity. The conductivity of a living tissue is a measure of its permeability to ions; and while it is conceivable that the permeability to other substances, such as non-electrolytes like sugar, may vary independently (within certain limits) of the permeability to ions, the advantages of estimating permeability quantitatively are such that the conductivity method must be regarded as the one to be preferred wherever it can be applied.

Osterhout has shown that by means of the conductivity method the permeability of plant tissues under varying external conditions can be readily and accurately determined; also that permeability can be varied at will, reversibly, in either direction, especially under the influence of neutral salts and lipoid-solvent compounds.^ In pure NaCl solutions the permeability of Laminaria fronds is increased, to a degree depending on the duration of exposure and the temperature; on replacing the tissue in sea water the permeability returns to or toward the normal, the degree of possible recovery depending upon the extent of the change produced by the NaCl solution.^ If the permeability has been increased beyond a certain limit, its reversal is impossible and the plant is dead. He has suggested, therefore, that the property of "vitality" may be measured by determining the electrical resistance.^ Isotonic CaClz solutions have the opposite kind of effect and at first decrease permeability; the antagonism between Na and

Ca salts is thus explained; when the two salts are present in appropriate proportions (20 NaCl+i CaCla) conductivity is unaltered, and the tissue remains living for days, while in the pure solution of either salt toxic action and death soon result. The changes of permeability accompanying normal physiological processes have also been measured by the conductivity method in certain cases, especially by McClendon and Gray in the fertilization of sea-urchin eggs.^