Protoplasmic Action and Nervous Action

^ The data in my paper {American Journal of Physiology, XX (1907), 127) show a relatively great depressant action on osmotic pressure for low concentrations of salts and a curve corresponding to the adsorption type. (influence of salts on surface tension, solubility, viscosity, catalytic action, etc.)/ The difference between the adsorbability of the two ions of an electrolyte implies an influence on the potential difference at the phase-boundary; the surface receives the charge of the more adsorbable ion and the adjacent layer of solution the opposite charge. Potentials arising in this way have been called ''adsorption potentials'' by Freundlich,^ and they undoubtedly play an important part in colloidal phenomena, since the alteration of the surface charge is a chief factor in the changes of aggregation-state and other phenomena characteristic of colloids. Adsorption potentials may also be of great importance in influencing the character of the chemical changes occurring under the catalytic influence of surfaces.

There are many indications that the ions of water, OH and H, are among the most readily adsorbed ions. This property is of special biological importance, since most forms of protoplasm are highly sensitive to variations in the concentration of these ions; and in certain cases this sensitivity has become a regulatory factor of the utmost delicacy. Free organic acids and bases are ^ For a summary of the physical effects of salts showing the lyotropic series cf. Hober, op. cit., p. 310.

* According to Freundlich the adsorption potential is not the potential between the interior of the solid phase and the adjoining solution, but that between an adhering immobile layer of solution and the mobile layer adjoining. Cf. Kapillarchemie, p. 243; also Report on the Physics and Chemistry of Colloids by the Faraday Society and the Physical Society of London (1921), p. 146. as a rule more readily adsorbed than their salts. Many organic acids (the lower members of the fatty series) are highly effective in lowering surface-tension, while their salts have little influence; this effect, however, may be in large part attributable to the undissociated molecules. As a class, acids have a marked effect on the surface charges of indifferent solid substances, tending to make these surfaces positive; similarly bases make them negative. Both effects appear in very low concentrations (Perrin)^ and are undoubtedly due to H and OH ions, respectively. The fact that adsorbent surfaces of the most widely varying chemical composition (carbon, hydrocarbons, silicates, oxides, metals) are thus affected indicates that adsorption rather than chemical combination in stoichiometric proportions lies at the basis of the effect. A slight change in H and OH concentration may thus have a very marked effect upon the potential difference across a surface; this is well shown in the curves given by Haber and Klemensiewicz,^ and in the results of Perrin.^ The great effectiveness of the H-ion as a precipitant for suspensions of indifferent substances probably depends on its high adsorbability as well as on its high velocity and special chemical properties.

At a certain concentration where OH and H-ions are adsorbed in certain proportions the surfaces will be electrically neutral; this condition defines the isoelectric 3 See especially Perrin's curve for naphthalene, reproduced in Freundlich's KapUlarchemie, p. 236. Ellis also describes this effect in the cataphoresis of oil droplets, Z. physik, Chem., CXXVIII (1911), 321, point. The H-ion concentration corresponding to this neutral position varies according to the special chemical character of the material and its adsorbent properties.

Various chemical effects apparently depending on unequal adsorption of ions are described in Freundlich's book; frequently these appear to be consequences of the displacement of one ion from an adsorbent surface by another which is more readily adsorbed. For example, many basic dyes (crystal violet or basic fuchsin) are chlorides of organic color-bases; when these dyes are adsorbed by charcoal, a large part of the combined chloride goes into solution as inorganic chloride. This result is explained by Freundlich as due to the high adsorbability of the color cation (dye-HCl = dye-H+ and Cl~) which displaces from the adsorbent the adsorbed cation;^ according to this view the chemical splitting is due to the unequal adsorbability of the two ions. The phenomenon may, however, be regarded as a consequence of the high adsorbability of the free base, which is present in the solution in consequence of partial hydrolysis; this base is removed, leaving the chloride in solution.^ It is known that many organic free bases are more highly surface-active (adsorbed) than the salts; thus Traube points out that the surface-tension of solutions of hydrochlorides of the alkaloids, cocain, atropin, and quinine is lowered by adding a little alkali, an effect due to the liberation of the free base;^ the latter will tend to be adsorbed and the hydrolysis will be promoted. The separation of

dyes on adsorbing surfaces in an insoluble form and the ''denaturation" of proteins on surfaces are apparently phenomena of the same general kind; adsorption will promote hydrolysis or other chemical alteration in a compound if any reaction-product is more readily adsorbed than the original compound. From this point of view the general kinetics of adsorption-catalysis appear in a clearer light. The heavy metal ions and the ions of trivalent metals appear to be adsorbed with especial readiness in many cases, and the solutions of their salts have a correspondingly great influence on the surface charge of colloidal particles and of porous partitions. In the case of suspended oil drops, Ellis^ found the following concentrations of three chlorides to be equally effective in removing the charge; i.e., in rendering the particles electrically neutral.

The relative actions of Na, Cu, and Al are approximately as I to 40 to 1,600, indicating a rapid increase of action with increase of valence. The same rule is found in the precipitation of suspensoid colloids by electrolytes (rule of Schulze and Hardy), and indicates 'f Ridsdale Ellis, Z. physik. Chem., LXXVIII (1912), 321. Loeb's recent study of the effects of ions in altering the charge on suspended collodion particles gives a similar result (/. Gen. Physiol., V [1922J, 109). In this case the collodion particles undergo precipitation when the P.D. against the medium falls below 16 millivolts. Bacteria are agglutinated at 15 rnillivolts according to Northrop and de Kruif (loc. cit.).

that a relatively great adsorption of polyvalent ions is a general rule. The relative effectiveness of polyvalent cations in certain characteristic physiological effects, e.g., ion-antagonism, is of a corresponding order, indicating that in such cases the ions act by adsorption at the structural surfaces of the living system. Thus Al and Cr ions greatly prolong the activity of cilia in isotonic NaCl solution when present in concentrations of less than M/ 100,000.^

It should be pointed out that even if the adsorption constants of mono-, di-, and trivalent ions were equal, the trivalent ions should be effective in less than a third of the concentration of the monovalent ions, because, on account of the characteristic form of the adsorption curve, the ratio between the quantity adsorbed and the quantity remaining in solution is much higher in dilute than in concentrated solution; hence sufficient trivalent ions to neutralize the surface charge may be adsorbed from extremely dilute solution. This is Freundlich's explanation of Schulze's rule."^ Whetham's explanation, based on chances,^ is probably insufficient, although purely mathematical considerations would indicate that the statistical conditions to which he calls attention must play a part in the total effect. The ratio between the effective concentrations of Al and Na in the above-cited experiments seems, however, too great to be accounted for on this ground alone, and a high degree of adsorption of polyvalent cations must apparently be assumed.

The importance of adsorption-potentials is shown most clearly in the phenomena of electrical endosmose, which are of great physiological importance and will be considered briefly below. The marked influence which slight quantities of acid and alkali have in altering the phase-boundary potentials, especially in the neighborhood of the isoelectric point, is a fact of special biological interest. Near this point the effect of variations in the H-ion concentration upon the phase-boundary potentials is at its maximum; farther from the isoelectric point the difference caused by a given change in the H-ion concentration is relatively slight.^

The reactions of the tissue fluids in higher animals are slightly on the alkaline side of neutrality, and the reaction of the living protoplasm (because of the higher tension of CO2 within the cell) is presumably somewhat less alkaline and is probably not far from the isoelectric point of some of the structural proteins. Hence slight variations of the H-ion concentration within the cell should have a correspondingly great effect upon the boundary-potentials of the corresponding cell-structures. Variations in the H-ion concentration of the cell-medium would affect first of all the boundary-potential of the cell as a whole; i.e., that existing across the plasma membrane, and this is probably the chief reason why living tissues are frequently so sensitive to changes in the external H-ion concentration. As already pointed out, this sensitivity has in some cases become the controlling factor in regulatory processes of vital importance to the whole organism, as in the respiratory nerve, cells of vertebrates, which show an accelerated rhythm in

response to an extremely slight rise of acidity due to increase in the CO2 of the blood. Certain physical phenomena of great biological interest, having similar relations to the charged character of the interphasic surfaces, are those classed as electrical endosmose. The essential phenomenon is the transfer of fluids with or against an electric current traversing a porous partition immersed in an electrolyte solution. The fundamental conditions of this transport are the same as those determining the electrical convection of colloidal particles or emulsion droplets, except that in electrical endosmose the continuous fluid phase is the mobile one, the solid phase which forms the substance of the partition being fixed in position. Colloidal gels interposed in the path of a current exhibit this phenomenon, and it is well known in protoplasmic systems.^ There is no question but that effects of the same kind must occur normally in living matter, since the latter, during functional activity at least, is continually being traversed by the currents of the bioelectric circuits; and it seems probable that electrical endosmose plays a special role in the processes of secretion and adsorption, as suggested by various physiologists.''

^ For example, Hermann describes experiments showing the transport of water through tissues (muscle and nerves) in the direction of the positive stream when a constant current is passed through the tissue {Arch. ges. Physiol., LXVII [1897], 240). At present, however, our knowledge is insufficient for any final estimate of its physiological importance. Since the effect depends upon the potential difference between the walls of the pores and the mobile fluid layer adjoining, all conditions that influence this potential difference affect the character of the movement. The rate of the movement or its direction or both may be thus affected. The effects of salts, acids, and alkalies are well illustrated in Perrin's investigations published in 1904.^ The following table gives the results of a typical experiment. The diaphragm consisted of naphthalene, and varying solutions of HCl and KOH were used:

This experiment and others of a similar kind show that when a current of given intensity is passed through a partition, the rate and direction of transport are determined by the charge of the partition substance, and that this varies in a definite manner with the nature and concentration of the ions present. Perrin also showed that the isoelectric point varied with the nature of the diaphragm; thus partitions of iodoform and glass were persistently negative, while those of BaC03 and CrCl3 were positive. The electro-

positivity of any surface, whatever its chemical composition, was invariably increased by adding monovalent acid to the solution, and decreased by adding monovalent base. Polyvalent ions were very effective, reversing the sign of the partition-charge in extremely low concentrations; e.g., for a partition of chromic chloride these results were found. Freundlich^ calculated from Perrin's data the concentration of salts (millimols per litre) required to reduce by 50 per cent the endosmose through positive and negative partitions, respectively, and obtained the following results.

The effect thus increases very rapidly, in accordance with Schulze's rule, with increase in the valence of the active ion, which is always that having the opposite sign to that of the partition. EUissafoff^ in an investigation with glass and quartz capillaries also found the Perrin-Schulze valence rule to hold for alkali and alkali earth salts, but heavy metals (Hg, Ag) were more active than corresponded to their valence; organic cations (morphine, neufuchsin) were also highly active. Reversals with Th(N03)4 and methyl violet were observed, but not with acids up to n/iooo; no higher concentrations were used.^

^ A general review by T. B. Briggs in the Journal of Physical Chemistry for 191 7 (XXI, 198, and XXII, 256) gives a full review of the subject of electrical endosmose. Recently its bearing on physiological Dhenomena and its relation to cases of anomalous osmosis have been considered in an important series of papers by J. Loeb (/. Gen. Physiol., II [1920], p. 557, and later papers in the same journal). For the action of ions see further, A. Gyemant, "Elektroendosmose und lonenadsorption," Kolloid-Z., XXVIII (1921), 103.

We have seen that the stability of emulsion systems and many of their essential properties are determined by the presence of thin interfacial films. In the formation of these films, and in the determination of their special properties, electrical and other factors enter, of a kind characteristic of boundary surfaces in general. In general, it may be said that heterogeneous mixtures containing substances that influence the surface-tension or the electrical polarization at the phase-boundaries tend to reach a state of equilibrium in which they have what may be called a ''structure"; i.e., a more or less orderly and definite distribution of the components. The gathering of surface-active compounds at the phase-boundaries, in conformity with the principle of Gibbs and J. J. Thomson, is a chief factor determining the character of this distribution; and the latter secondarily determines the character of the chemical changes occurring in the system. Living protoplasm is an example of a heterogeneous system in which the control of chemical change by structural conditions has reached perhaps its highest development.

Not only are the structural elements of protoplasm (alveoli, nuclei, colloidal particles, fibrils) bounded by surfaces at which adsorption and chemical change occur, but the whole mass of protoplasm, the living cell, is similarly bounded. Suspensions of living cells — suspensions of blood corpuscles, eggs, spermatozoa, bacteria, or yeast — in their normal aqueous media may be regarded as similar in many respects to emulsions. Each cell is a small discrete particle with a surfacetension and an electrical potential-difference against its medium. The existence of this potential is shown in convection experiments; just as oil droplets migrate to the anode in an electrical field, so do suspended living cells; each cell, although living and highly differentiated internally, behaves in this respect like a negatively charged colloidal particle. As in the case of a suspended oil droplet, the sign of the charge carried by the living particle may be changed by acids or polyvalent ions; and like colloidal particles in general the cells may be precipitated from suspension (or agglutinated) under various conditions.^

Red blood corpuscles travel in neutral media like isotonic sugar solution to the anode, thus showing the presence of a negative surface charge; by passing CO3 through the media or adding weak acids, the sign of the charge may be reversed; the corpuscles then become positive or cathodic."^ At an intermediate concentration the charge is abolished (isoelectric point); and it is interesting to note that at this point the corpuscle tends to break down or undergo •haemolysis.^ This fact is of

^ For data and references in this field, cf. Hober's Physik. Chemie d. Zelle, pp. 247, 300, 483, and 599. More recently the cataphoresis of bacteria and their agglutination by electrolytes and other substances has been studied by Northrop and De Kruif (/. Gen. Physiol., IV [1922], 629, 639, and 655). special interest as indicating the importance of electrical factors in the structural stability of the protoplasmic surface layer, and suggests a reason why depolarizing influences have in general a stimulating action on irritable cells. The relations between structural change and stimulation will be considered in more detail later.

Hober and Kozawa^ found this isoelectric point to vary for different species of corpuscles and to be characteristic for a particular species; i.e., certain corpuscles are more readily made positive than others by H ions and polyvalent ions; thus the corpuscles of rabbits and guinea pigs were found to require the least H-ion concentration for reversal and those of the ox and pig the highest; those of the dog, cat, goat, and man were intermediate.

The relative order with La ions was the same. Such differences are to be referred to the specific peculiarities of structure, composition, or permeability characteristic of each kind of corpuscle. In their general or common features, these phenomena indicate that the behavior of the protoplasmic free surface with reference to the ions present in the medium resembles that of the surface of an individual oil-droplet in an emulsion. It may be assumed that

the external surface of the cell is not peculiar in this respect, and that other protoplasmic interfaces (those within the cell) have similar properties. Film-formation, variations of electrical polarization, and dependent phenomena are thus to be regarded as constant features of the processes at such interfaces; e.g., at the surfaces of contractile fibrils, neuraxones, vacuoles, nuclei, alveoH, and other protoplasmic structures. Secondary effects peculiar to living protoplasm and dependent on metabolism may be superposed on these simple physical effects.

An important difference between the structure of protoplasm and that of a typical artificial emulsion, like oil in water, is that the two phases separated by a protoplasmic film are not necessarily aqueous and nonaqueous, respectively, but may both be aqueous. Any suspension of living cells, such as blood, illustrates this condition; both the interior of the corpuscle and the serum are complex aqueous solutions, and the film bounding the corpuscle is too thin to be optically detectable as a separate structure. Yet it can readily be shown by osmotic methods that a semi-permeable membrane is there present. Similarly within the limits of a single cell various structurally distinct regions, sometimes optically distinguishable, sometimes not, are to be regarded as separated by thin films. The details of this film structure will naturally vary in different types of cell.

The surface layer of the entire cell, the plasma membrane, is the modified surface-film which separates the internal protoplasm from its surrounding medium and through which the necessary interchange of material is effected. During life it behaves as if it were semipermeable and water-insoluble; yet it is evident that it allows passage, at least at certain times, to water and many dissolved substances; e.g., food-substances. Because of its special character as an intermediary structure or organ, its properties require special consideration in any study of the properties of protoplasm. In some respects the properties of the plasma membrane appear difficult to explain on known physical grounds or even paradoxical. If a diffusible foodsubstance passes through the membrane into the cell it may there undergo metabolism, but it is difficult to see how any such substance can enter without other important substances leaving. If simple diffusion alone is the chief factor in the entrance and exit of dissolved materials, we should expect that any increase of permeability sufficient to allow the entrance of substances like sugar would also involve a loss of diffusible cell constituents to the exterior. It seems necessary to assume that the conditions determining the normal transport of substances across the cell boundary are of a special kind, and that diffusion is only one of a number of factors. In fact we know that in secretion and absorption special physiological mechanisms of transport are concerned, by which water and dissolved substances are actively conveyed into and out of the cell, frequently against concentration-gradients. This process requires the performance of work by the cell, just as does the process of muscular contraction, although its exact conditions are at present unknown. Many substances (sugar, salts, amino-acids, urea) may thus be transported from regions of lower to regions of higher

concentration, a process necessarily involving osmotic work; in this work O2 is consumed and energy set free. The problem of the nature of this physiological transporting mechanism is often distinguished as the problem of "physiological permeability" from the simpler problem of the conditions of simple diffusion across the cell surface, or '^ physical permeability." The majority of recent studies on protoplasmic permeability have had reference to this latter problem.

It will be apparent that the problem of the nature and conditions of cell-permeability is not a special or limited one but underlies the whole problem of the essential nature of protoplasmic structure and activities. Apparently the most fundamental property of the plasma membrane is semi-permeability; water can pass, although with considerable resistance in many cases; but the normal water-soluble constituents of the protoplasm and its surroundings do not diffuse across the cell boundary, or only under special conditions. This property of "semi -permeability" is associated with a high degree of electrical resistance (signifying impermeability to crystalloidal ions) and water-insolubility. The physicochemical conditions of these properties of the plasma membrane have been much investigated of recent years, and since the results of this work have an important bearing on our general problem, a somewhat detailed review will be given.

The general importance of membranes in organic structure and activities has long been recognized, and much study has been devoted by physiologists to various artificial types of membrane, such as precipitation membranes, membranes of parchment and collodion, and various impregnated types of membrane, in the hope of throwing light upon the pecuKarities of hving membranes. The resemblance of these artificial structures with the protoplasmic membranes seems, however, in many cases remote. The conditions observable in living protoplasm indicate that membranes of a different and somewhat special type, viz., the interfacial films formed between the separate phases in emulsions and other polyphasic systems, have a closer affi.nity with protoplasmic membranes than any other simple type of physical structure.

The property of forming thin films at structural surfaces of all kinds is highly characteristic of living protoplasm. The entire cell body is separated from the surrounding medium by the plasma membrane; within the cell the nuclear area is sharply delimited by a membrane which usually disappears only during mitosis; structurally distinct membranes are also formed about fibrils, vacuoles, alveoli, chromatophores, spheres, chromosomes, and various cell-inclusions. Apparently the conditions required for the formation of thin solid films are present everywhere in living protoplasm; at any structurally well-defined surface, continuous sheets of material may be deposited; the formation of membranes at cut surfaces and about extruded portions of protoplasm is simply one illustration of this general property.^ The fertilized eggs of marine animals (sea-urchin, starfish) are especially favorable objects for showing this property; new films are rapidly formed