Protoplasmic Action and Nervous Action

^ See my series of papers on antagonisms between salts and anaesthetics, American Journal of Physiology, XXIX (1912), 372; XXX, i, andXXXT (1913), 255; dlso Journal of Experimental Zoology, XVI (1914), 591. Hober's observations (above cited) on the effects of anaesthetics in retarding the production of injury-currents in muscle by salts furnish other examples of this phenomenon. this action in a manner resembling that of an antagonistic salt like CaCla. Clowes has shown that in the physical drop-systems which he studied — alkaline NaCl solution flowing from a stalagmometer through oil — anaesthetics produce an effect closely comparable with that of CaCla; both actions are to be referred to changes in the properties of the interfacial films formed between the oil and the aqueous solution.^ The presence of a compound (Ca soap, or fat-solvent compound) which is more soluble in the oil phase than in the water phase, modifies the conditions at the boundary in the same manner in both cases, and produces the same physical effect in the system. The parallelisms observed by Clowes between the biological and the physical phenomena may be interpreted as indicating that the conditions in living protoplasm are of a closely analogous kind.

The foregoing effects of anaesthetics on the physical properties of protoplasm do not, however, enable us to decide whether the solvent action or the adsorbent action is the chief factor in the narcotic effect, or whether both are equally concerned. Certain widely general biological phenomena do not appear to be entirely consistent with Traube's theory that surface-activity is the essential factor in all cases of narcosis. These are: (i) the great differences observed between the narcotizing concentrations of the same compound in different cells, tissues, and organisms; (2) the fact that weak solutions of many narcotic compounds have a sensitizing or accelerating influence on many cell-processes;^ (3) the

* Cf. the instances cited in my review of the theory of anaesthesia {Biological Bulletin, loc. cit.). general relation, pointed out by Overton and Meyer, between the Kpoid-content of tissues and their susceptibility to narcotic action; and (4) the anaesthetizing action of compounds, such as Mg and K salts, wliich are without surface-activity in the foregoing sense. Traube's theory in fact, in emphasizing the importance of a single physical factor, seems to disregard other possible factors, and on the whole to underestimate the complexity of the physiological conditions.

Hober,^ Vernon,^ and others have pointed out various exceptions to the rule that isocapillary solutions have equal physiological action. This rule cannot be true in any precise sense, since an organic compound may affect the conditions at the various interfaces in a complex system Kke protoplasm, and at an air-water interface, quite differently; and other factors, including viscosity, solubiUty in the protoplasmic phases, and specific chemical affinities enter to modify the simple surface conditions. In cases where the conditions are simple, the increase from compound to compound in a homologous series may be very regular; a good example is seen in Fiihner's observations on haemolysis by solutions of alcohols;^ the critical haemolytic concentrations for the first five alcohols are as follows:

In this case, the ratio of about 3 : i between successive members is well shown. In other cases this relation is obscured by chemical and other factors; an instructive instance is the relatively high toxicity of methyl as compared with ethyl alcohol, apparently a consequence of the special properties of the former's oxidation products, formaldehyde and formic acid. Fiihner cites other cases showing a similar regularity; he finds, however, that in tissues with high lipoid-content, such as the vertebrate central nervous system, the higher members of certain series (alcohols) are more effective than would be expected from this simple rule. This discrepancy he ascribes to the larger proportion of organic solvent (Hpoid) in such tissues; thus, in the adult frog the divergence from the 3 : i ratio is greater ( = ca. 4:1) than in the tadpole ( = 2.9:1); this difference is apparently referable to the increase in lipoid constituents as the central nervous system develops.^ It seems probable that in lipoid-rich tissues the lipoid-solvent factor becomes relatively important in comparison with the capillary constant factor.

The relatively great solubility of many anaesthetic organic compounds in protein-containing systems deficient in Hpoid (serum, finely divided muscle, etc.) has been attributed by Moore and Roaf^ to the formation of chemical combinations with the protein; but since all such systems are undoubtedly polyphasic, and since chemical combinations (in the stoichiometric sense) of hydrocarbons (like CHCI3, benzol, etc.) with proteins are 2 Moore and Roaf, Proceedings of the Royal Society, B, LXXIV

difficult to conceive, it seems more likely that an adsorption-effect is involved, similar to that observed in finely divided suspensions of charcoal. It is well known that the catalytic effect of such suspensions is markedly influenced by surface-active compounds;^ this effect indicates an alteration in the character of the surface, probably resulting from adsorption. Other phenomena of a related kind, e.g., the precipitation produced by organic solvents in protein solutions (as observed by BattelH and Stern and Moore and Roaf),^ the liquefying action of these compounds on gelatine gels (Traube and Kohler),^ the solidifying action on lecithin suspensions, and the interference with the precipitation of lecithin suspensions by electrolytes (Koch, Hober and Gordon, and others) ,'' are similarly referable to surfaceconditions, although the special nature of these conditions is not clear in all cases. It is noteworthy that most of these effects are observed at concentrations far in excess of those required to produce reversible narcotic effects in living protoplasm. To characterize the organic anaesthetics as negative catalyzers, as Traube does, may place them in a class, but does not explain their characteristic action on living matter.

It seems certain from the physical peculiarities of these substances that they must undergo adsorption 3 Traube and Kohler, Internal. Zeilschr. f. physik.-chem. Biol., II at the structural surfaces in protoplasm, and also dissolve, in accordance with their partition-coefficients, in the organic solvents of protoplasm. It is not a case of two incompatible processes; both occur simultaneously, and each contributes to the total effect. Possibly the reversible effects characteristic of low concentrations of anaesthetic substances are the expression of solution in the lipoids, while with higher concentrations the specific structural compounds of the protoplasm, the proteins, are affected through coagulation or other changes due to adsorption, and irreversible effects result. The reversible effects — stimulation or sensitization in very weak, inhibition or anaesthesia in stronger, solutions — are the expressions, respectively, of facilitation and hindrance of the normal metaboHc processes underlying stimulation and automatic activity; i.e., the influence of the cell-structure on the metabohc reactions is modified, and the whole behavior of the protoplasmic system is altered correspondingly.

If the influence of structural conditions on cellmetabohsm is to be included under the class of heterogeneous or contact catalysis, as many of the foregoing facts indicate, it is evident that a consideration of this type of catalysis and of the manner in which it is influenced by substances of the foregoing kind becomes essential in the further analysis of the conditions in living protoplasm. The chemical reactions in protoplasm are under the control of structure, as we have seen, and their velocity is decreased and, in the case of the most characteristically vital reactions, the specific syntheses, is reduced to zero when protoplasmic structure is destroyed. If we class this influence of structure as catalysis, the case becomes one of heterogeneous catalysis, in which the reacting substances are predominantly substances in aqueous solution. Organic solvents, however, are also present, represented chiefly by the lipoids; and, as in all cases of heterogeneous catalysis, the interfacial relations are undoubtedly of primary importance. The permanent structural elements are chiefly protein in composition, probably associated with lipoid; and this fact favors the inference that the interfaces between the solid protein structures of the cell and the adjoining more fluid phases are the site of the biologically essential reactions, and especially of the syntheses. The fact that surfaceactive substances as a class interfere so strongly with these reactions favors this interpretation.

The distinctive syntheses of living matter are those of proteins. These are the syntheses on which specific growth depends, and growth is the fundamental vital activity. It thus appears probable that growth is chiefly a result or expression of synthetic reactions occurring at such interfaces; and the general susceptibility of protoplasmic processes, including growth, to electrical influences seems to imply that the electrical conditions existing at these interfaces are an essential factor in the control of these reactions.

The general subject of the catalysis of substances in aqueous solution in heterogeneous systems has thus an intimate bearing on the fundamental biological problem which we are considering; and a brief review of the more relevant facts in this field is essential to the further analysis of the conditions in living protoplasm. It must be remembered, however, that the theory of catalysis is still in many respects incomplete, and that many reactions in living protoplasm appear to be determined by other than purely catalytic conditions — using the word catalysis in the accepted sense of an acceleration in which the catalyzer undergoes no permanent change in the reaction. Induced reactions probably play an important part; and there are apparently also cases where the catalyzer acts by introducing a factor necessary to those special physical conditions' — e.g., flow of electric current through the bioelectric circuit — which control the reaction.

The general parallel between the conditions determining chemical reactionvelocities in general, and those determining the flow of an electric current through a circuit, has often been dwelt upon.^ The quantity of material transformed in a reaction, or of current flowing through the circuit is determined: (i) by the intensity of a physical condition, called electrical or chemical "potential," whose expression is a furtherance of the change in question; and (2) by the resistance to this change. The general formula C = P/R describes the general conditions, where C signifies either the rate of chemical change (under determined conditions of concentration, temperature, etc.) or the intensity of the current flowing through the circuit, P the potential, signifying a function of "chemical affinity" in the one case, or the electrical pressure or "voltage" of the circuit in the other, and R the resistance to either the chemical change or the flow of current. In the case of a chemical reaction occurring at an electrode (electrolysis), where the quantity of chemical change, e.g., of copper deposited as metal at the cathode, is proportional to the quantity of current flowing through the circuit (Faraday's Law), the factors determining the flow of current are the same as those determining the rate of chemical change, and chemical resistance becomes identical with electrical resistance. In such a case any condition decreasing the electrical resistance or increasing the electrical potential increases the velocity of the purely chemical change at the electrode.

At present it is customary to describe a catalyst as a substance which in some manner, without itself undergoing permanent alteration, decreases the resistance to the interaction of other substances in the reactionsystem. On such a definition any substance which decreases the electrical resistance in a circuit would catalyze the chemical reactions occurring at the electrodes. Such an effect might not ordinarily be classed as catalytic; but since our interest is not in defining the significance to be attached to terms, but in ascertain-

ing the physico-chemical conditions under which chemical reactions actually are accelerated in systems of the kind under consideration, we must note as especially significant the fact that reactions occurring under electrical influence at surfaces (especially metallic surfaces) may be influenced in their velocity by the contact of materials which change locally the electrical state of the surface. The rusting of iron in water or salt solution is a good instance of this type of effect; the reaction may be greatly accelerated by placing another metal, e.g., copper or platinum (which itself does not undergo change) in contact with the iron. The apparently catalytic effect in this case is due to the formation of an electrical circuit between the two metals, the iron becoming anodal and hence freeing Fe ions with increased rapidity; these can then react to form carbonate or hydrate with the anions present in the solution. Another simple and striking demonstration of a '^catalytic" action of this kind is made by placing in a solution of KjFeCye (containing a little NaCl to allow a soluble zinc salt to be formed) two similar strips of metallic zinc, one of which is marked with a lead pencil or bound with a small piece of copper or platinum, while the other is free from such contact. In a few hours a luxuriant ^'growth" of filaments and tubules of zinc ferricyanide is formed from the first strip, while the second remains almost unaltered. The carbon, or the noble metal, acts ''catalytically" in this reaction because it furnishes a surface of lower solution-tension, which formes the cathode of the local electric couple; and since these two areas are in metallic connection and immersed in the electrolyte solution, a current flows which enables

Zn ions to enter the solution more rapidly and hence accelerates the formation of the structure-forming precipitate of zinc ferricyanide.^ There is reason for believing that the remarkable chemical activity of living matter, as well as its susceptibility to electrical influence and to stimulation, is largely dependent on physical conditions which are fundamentally of the kind just described. For example, during stimulation the excited and the unexcited areas of the reactive protoplasmic surface — the surface of the irritable cell, neurofibril, or other structure concerned — are at different electrical potentials; apparently the current flowing between these two areas produces chemical effects which secondarily determine the propagation of the state of excitation and hence the distinctively physiological effect or response. There is a close analogy here to the case of local circuits in metals immersed in electrolyte solutions; these circuits also form the condition for the transmission of chemical effects. This general condition will be considered more fully under the subject of stimulation; at present it is sufficient to call special attention to it as probably forming a highly important factor in the catalytic or quasi-catalytic action of Hving protoplasm. Here we use the term "catalytic" simply as a designation for the remarkable property shown by living protoplasm of enabhng reactions to occur, at a relatively high velocity, which are absent or inappreciable in dead protoplasm.

The most familiar form of catalysis observed in living organisms, and the one showing the closest parallels with the usual inorganic types of catalysis, is that dependent on the activity of enzymes. Enzymes are constant constituents of protoplasm, and their presence accounts for many characteristic features of its chemical performance. Enzyme-action, however, is obviously responsible for only a portion of the metaboHc reactions, especially the destructive or disintegrative ones, which are largely hydrolytic. Although certain types of synthesis are accelerated by enzymes under certain conditions (dehydrolytic synthesis of esters, carbohydrates, and apparently polypeptides), others cannot be thus accounted for; e.g., photosyn theses, synthesis of fat from protein or carbohydrate, or other syntheses involving the expenditure of much energy. Moreover, the responsiveness of protoplasm to stimulation is not thus explained, since enzymes show no such instantaneous and marked acceleration of their action, under electrical or mechanical influence, as is shown by Hving protoplasm. As we shall see later, changes in protoplasmic structure seem to be primarily responsible for the immediate chemical effects following stimulation.

Enz3anes are simply colloidal catalyzers of complex and specific chemical constitution. A part of their catalytic activity presumably depends on their colloidal state; i.e., a state of subdivision making surface-conditions of preponderant importance in their chemical behavior. The general conditions of heterogeneous catalysis thus apply to enzyme action; in addition there are special conditions referable to the specific stereochemical configuration of the enzyme molecule.

It is well known that finely divided material of various kinds (material with large surface-extent) is often very active in catalyzing chemical reactions; this is especiallytrue of certain forms of carbon and of metals like platinum. The action of platinum is especially well known; it increases with the state of subdivision, i.e., the extent of surface, hence colloidal platinum is a very effective catalyzer. Other metals have similar properties, although usually less marked.

Usually in such heterogeneous catalyses the acceleration of reaction-velocity is regarded as a result of increased concentration at the surfaces. Faraday (1839) suggested this explanation for the action of platinum in catalyzing the combination of hydrogen and oxygen. In general, when considering any special case of heterogeneous catalysis, three independent processes with different rates are taken into account: (i) the rate of diffusion of the dissolved substrate to the active surface; (2) the rate of adsorption at the surface; and (3) the rate of chemical combination. The rate of reaction is Limited by the rate of the slowest of these interdependent processes. In most cases the reactionvelocity (F) is regarded as determined by the concentration (C) attained at the surface and by the specific velocityconstant (K) of the reaction (i.e., V = KC), since adsorption is rapid and also diffusion (when the distances are small). The rate of chemical change is increased (catalytic effect) because the concentration of the reacting molecules is increased in this part of the system.^

It is evident, however, that other factors frequently if not usually enter dependent on the special chemical nature of the reacting compounds. Many inorganic ^ Cf. Hober's Physik. Chemie der Zelle, pp. 702 ff., for an account of the general conditions of catalysis in heterogeneous systems. catalyses are referable to the formation of intermediate compounds, and the same is undoubtedly true of many enz3ane-reactions. The specificity of enzymes and other facts in their behavior indicate that chemical union often occurs between the enz3rme and the substrate molecules, and that it is the combination thus formed which breaks down rapidly, yielding the products of hydrolysis and the free enzyme, which then repeats the chemical cycle of combination and hydrolysis with fresh molecules of substrate.

Apparently many cases of heterogeneous or contact catalysis are referable to simple increase of concentration due to adsorption. But in the case of metals and other conducting substances the possibility of an additional factor, the formation of local electrical circuits between different portions of the active surface, is also to be considered. In either case the essential condition is some form of surface-action. Enzymes are colloidal in their condition, indiffusible, precipitable, readily adsorbed by indifferent adsorbents, and, according to Bayhss, their mode of action is also a surface-action. The clearest proof of this is that emulsin, lipase, urease, and trypsin exert their action in alcohoHc media of such a strength that the enzyme is insoluble and can be filtered off.^ BayHss regards the adsorption of the substrate on the enzyme phase as the first step in the process; the chemical reaction then follows. In some cases the adsorptioncompound of enzyme with substrate is separable; e.g., starchamylase, fibrinpepsin, and trypsin with caseinogen.^ A close union or

adhesion of the enzyme to the substrate is characteristic; and the influence of electrolytes on enzyme-processes is probably in large part to be referred to their influence on adsorption.^ In the simple adsorption type of catalysis the accelerating effect depends on the concentration attained at the interface. Close adhesion of the reacting compound to the adsorbent surface is important since this impHes a high concentration at the surface. Hence a correspondence between the molecular configuration of the adsorbing surface and of the adsorbed compound is favorable to adsorption as well as to chemical combination. The importance of such conditions is seen in the growth of crystals, in which, according to Marc, the dissolved molecules are abstracted from the motherliquid and deposited on the surface of the crystal by a process identical with adsorption.^ Slow growth is favorable to the formation of large crystals, because time is then allowed for the regular orientation of the surface-molecules thus deposited. Organic growth apparently also depends on the apposition of newly formed molecules to the similarly constituted molecules already laid down in the soKd state as structure; and this consideration may explain why, in living organisms, where definiteness of form and of .structural characters is essential, the rate of growth is slow. Probably no essential distinction is to be drawn between adsorption and chemical combination; in adsorption the surface molecules of the adsorbent are alone concerned because

' Cf. Bayliss, "Adsorption as a Preliminary to Chemical Reaction," Proceedings of the Royal Society, B, LXXXIV (191 1), 81. of the solidity of the adsorbing phase and the remoteness of the internal molecules from the sphere of reaction. Since surface-relations are all-important in heterogeneous catalysis, all conditions modifying the composition, electrical polarization, or other characters of the active surfaces influence the catalytic activity of the system. Hence surface-active substances as a class have a marked effect on such catalyses, and their influence on the chemical activity of Hving matter is undoubtedly in large part referable to this effect. A brief review of the action of these substances on the catalytic and other properties of heterogeneous systems will indicate the nature of the factors.

One of the most complete recent studies of the anticatalytic action of surface-active substances on enzyme action is that of Warburg and WieseP on the zymasecontaining ''pressjuice" of yeast. All substances of the anaesthetic class retard the alcohoHc fermentation caused by this enzyme, although the non-hving enzyme requires higher concentrations than the hving cell for the same proportional degree of retardation. In homologous series the concentration necessary for a given retardation decreases in the usual manner with increase of molecular weight. The following orders of relative action were found for different compounds. Alcohols: methyl < ethyl < propyl <isobutyl; urethanes: ethyl < propyl <isobutyl; nitriles: aceto<propio<isovalero; ketones : acetone < methyl-propyl < methyl-phenyl. It

was also noted that all of these compounds in sufficient concentration caused precipitation in the press-juice, and that the orders of relative precipitating effectiveness and anticatalytic action were the same; this order is also that of relative narcotic action. This parallelism between precipitating action and narcotic action recalls Claud Bernard's hypothesis that a partial coagulation of protoplasmic constituents is the essential condition of narcosis/ With the living cell, however, much lower concentrations are required to stop fermentation than with the enzyme solution, so that the parallel between the inactivation of the structureless enzyme solution and the inhibition of fermentation in the Hving cell is not complete. This difference may indicate the importance of the vital organization as such, or it may depend on the presence of special compounds (Kpoids) in the living cell. Warburg and Wiesel found, however, that dried yeast cells (extracted with ether and acetone) exhibited a well-marked fermentative action, which was arrested by narcotic compounds in somewhat high concentrations. Meyerhof found a closely similar anticatalytic action of the same compounds in solutions of yeast invertase;^ and in this case also the effect was associated with a precipitating action; similar observations on oxidasecontaining tissue-extracts have been made by Battelli and Stern.^ Vernon^ also observed a general inhibitory action of narcotics on tissue-oxidases ; the effect was

^ Claude Bernard, Leqons sur les Anesthesiques et sur rAsphyxie, Paris (1875), P- 154. imperfectly reversible and may be regarded as destructive rather than simply anti-catalytic. Narcotics also check heterogeneous catalysis in purely inorganic systems; e.g., the oxidation of oxaHc acid by charcoal (Warburg)^ and the decomposition of H2O2 by colloidal platinum (Meyerhof),'' and the same order of relative action is again seen. The foregoing association of a precipitating action with an anticatalytic action indicates an alteration of surface-conditions, but precipitation as such is not a necessary accompaniment of this action. Many organic compounds (alcohols) precipitate solutions of proteins and other colloidal compounds, but usually the concentrations required for precipitation far exceed the anticatalytic concentrations.

Other effects, also dependent on surface-action, have an intimate bearing on the present problem. Of special interest is the action of narcotic compounds on suspensions of lecithin. Changes in the viscosity, gelforming properties, and precipitabiUty of lecithin emulsions are characteristic. Many lipoid-solvents (alcohols, etc.) increase the viscosity of these emulsions to a greater degree than can be accounted for by the increase in the viscosity of the aqueous phase.^ In somewhat concentrated emulsions (10-12 per cent) the addition of ether even causes gelation, so that the test tube can be inverted without spilling; this effect is also caused by alcohols (n-propyl up to capryl), esters (ethyl formate, acetate,

propionate, and nitrate), ethyl ether, EtCl, EtBr, chloretone, paraldehyde, CCI4, hydrocarbons hke benzol, toluol, and xylol, but not by urethanes and lower alcohols.^ In a lecithincontaining system such as protoplasm this influence might act anticatalytically by slowing diffusionvelocities; any chemical process whose rate depended on the rate of diffusion would thus be retarded. Increased hindrance to the movement of ions would be shown in decreased electrical conductivity. Loewe^ has in fact found that artificial membranes impregnated with lecithin exhibit an increased electrical resistance in the presence of anaesthetics; in a system Uke hving protoplasm such an effect would retard chemical reactions dependent on electrochemical conditions. The increase in viscosity is probably to be referred in part to the formation of adsorption-films at the surface between the lecithin particles and the water.^ It is possible also that changes in the relative volumes occupied by the colloidal particles and the aqueous phase may play some part; presumably the lipoid-soluble compound concentrates in the lecithin in accordance

It should be noted that in the case of some other colloidal organic compounds the formation of gels may be prevented instead of promoted by the addition of surface-active substances; this effect was observed by Schryver^ in the gelation of Na-cholate in the presence of various surface-active organic compounds. He found a retardation in the rate of gelation, the effect running in general parallel with capillary activity and narcotic action. This phenomenon is analogous to protective action in colloidal precipitation.

The protection of suspensions of lecithin against precipitation by electrolytes is also an effect characteristic of many surface-active compounds. Hober and Gordon^ found that suspensions containing ether, chloroform, chloral, or amyl alcohol were less readily precipitated by alkaH-earth cations (Ca, Ba) than the control suspensions; i.e., were stabihzed; and they characterized this action as ''narcotization of the plasma membrane colloid lecithin." Koch and MacLean^ found that the stabilizing effect was not uniform with different anaesthetics; some compounds so act, but others are indifferent, while still others further precipitation, especially the lower alcohols and paraldehyde, which are highly water-soluble. My own observations on the precipitation of lecithin by CaCla and HCl confirm this result, but they show that at appropriate concentrations the great majority of anaesthetics have a stabiHzing effect. The compounds examined included alcohols