Protoplasmic Action and Nervous Action

^ Schryver, Proceedings of the Royal Society, B, LXXXVII (1914), 3 Koch and McLean, Journal of Pharmacology and Experimental Therapeutics, II (19 10), 249. from n-propyl up, esters, urethanes, hydrocarbons, chloroform, nitromethane, ethyl ether, chloretone, phenyl urea, and others.^ In general these effects in lecithin suspensions are referable to several factors, of which the chief probably are solution of the compound in the colloidal particles, formation of adsorption-films which change the electrical polarization or other physical constants of the particles, and increase of viscosity. The total effect in most cases is increase in the physical stabihty exhibited by the system in the presence of conditions tending to alter the state of aggregation.

The protective influence exerted by proteins and other surface-active colloids in the precipitation of metallic hydrosols by electrolytes is apparently a closely related phenomenon. Its conditions have been studied carefully by Zsigmondy,^ who finds wide variations in the effectiveness of different compounds; e.g.., gelatine is highly effective as compared with peptone. He assigns to each protective colloid a characteristic ^'gold number" : this number defines the quantity of the colloid required to prevent the precipitation of a standard gold suspension by a definite concentration of NaCl. In this case the protective effect undoubtedly depends on the formation of adsorption-films, which prevent coalescence or flocking of particles. Adsorption-films of soap, protein, or similar substances play an analogous part in the formation of emulsions, as already pointed out, the suspended droplets being thus prevented from fusing. In a similar manner these substances prevent sedimentation in finely

^ Cf. my The Theory of Ancesthesia, p. 361. * Zsigmondy, Colloids and UUramicroscopy. divided suspensions of water-insoluble materials (chalk, Ca-phosphate, etc.). In his textbook^ Hober has described various other instances of protective action and has discussed briefly the general biological significance of this phenomenon. In h\dng organisms, where water-insoluble materials constitute an indispensable element in the formation of all permanent structure, protective action is of great importance. Many otherwise insoluble materials are thus kept in a permanent state of fine dispersion or pseudosolution. It is well known that uric acid and cholesterol are present in serum in concentrations far higher than correspond to their solubiHty in water; presumably they are held in suspension as ultra-microscopic particles by the protective action of the serum proteins. Pauli and Samec^ have shown that gelatine, serum albumin, and other proteins may keep large quantities of insoluble calcium salts (sulphate, phosphate, carbonate) in apparent solution. The silver salts in photographic plates and the Ca-phosphate in milk are other instances of insoluble materials kept in fine dispersion by protective colloids. The suggestion has been made that the formation of pathological concretions in higher animals (gallstones, uric acid deposits) is an expression of deficiency in protective colloids. It is probable that in the formation of normal structures, such as bone, by separation of insoluble salts as a finely divided and structurally regular deposit, the protective action of the protoplasmic colloids is a necessary factor; presumably in the absence of this factor the particles would be flocked

by the salts present and the uniform and gradual building up of regular structure would be impossible. All of the above-described effects depend on the formation of interfacial films which alter the physical properties of the surfaces. Apparently also the chemical or adsorptive and hence catalytic properties of the surfaces are secondarily altered by the deposition of such foreign materials; and the foregoing evidence indicates that the an ti catalytic action of surface-active compounds depends on a surface-concentration of this kind.

Anticatalytic action is, however, also well known in homogeneous solutions and even in gases, so that this form of explanation does not always apply. Numerous instances of anticatalysis in homogeneous solutions are cited by Traube in his Theorie der Narkose;^ Bigelow's observations^ on the oxidation of Na2S03 by oxygen furnish many striking illustrations; he found that this reaction was depressed by a large number of organic substances; for example, a trace of mannite (about .0014 per cent) decreased the reaction velocity by 50 per cent; benzyl alcohol and aldehyde, isobutyl alcohol and ethyl alcohol were even more effective in checking the reaction. It is interesting to note that the alcohols showed increasing effectiveness with increasing molecular weight, and that the effect decreased with the number of hydroxyls in the compound (e.g., monohydroxylic alcohols > dihydroxyhc > trihydroxyHc, etc). Aromatic compounds were more effective than chain compounds.^

Anticatalysis in homogeneous solution has the appearance of being a different phenomenon from the forms of anticatalysis considered above, and its conditions are obscure. Possibly the nearest analogies are with photocatalysis; e.g., a catalytic substance may play a role analogous to that of a photochemical sensitizer; it is clear that interference with the action of the latter would arrest the photocatalytic action. Just as the chemical effect of Hght is influenced by th'e presence of substances with special Hght-absorptive properties, as in sensitized photographic plates, so also catalysis by chemical compounds may be influenced by other compounds having special chemical relations with the catalyzer. Such considerations are perhaps vague, but since many of the conditions controlHng reaction-velocities are imperfectly understood, it seems best not to attempt further definiteness at present.

Of the conditions, other than temperature and the presence of catalyzers, influencing reaction-velocities, light and other forms of radiation and electricity are the most important. Apparently all chemical reactions are influenced by radiation of appropriate wave-length, and the acceleration caused under these conditions is called "photocatalysis." It differs, however, from the chemical forms of catalysis in that energy is added to the reacting system from without; in this respect the conditions may be compared to those present when an electric current is passed into a solution from an electrode ; at the surface of transition between solution and electrode chemical reactions are induced (electrolysis), the effect depending directly on the quantity of electricity (i.e., number of electrons) transferred between the molecules of the electrode and those of the solution. Similarly the chemical effect of light is referred to faciHtation of the transfer of electrons between molecules. In virtue of its physical character as electromagnetic oscillation, light alters the range of movement of the electrons; and when the periodicities of electron motion and ethervibration correspond, this range may be increased to a degree sufflcient to enable adjacent molecules to interact. The electrons affected are apparently the valence or combination electrons of the molecules concerned. The chemical action of light is thus ultimately to be

related to its general influence on electrons, an influence shown in the photoelectric effect. The phenomena of induced reactions, which are almost certainly of great importance in cell-metabolism, are probably also to be afhliated in a general way with photocatalysis and electrolysis (which may perhaps be called " electrocatalysis ") 5 but it is impossible here to do more than direct attention to these possibiH- ties, the investigation of which is the subject of physical chemistry rather than of biology. It is sufficiently evident that all conditions influencing chemical reaction-velocities are of fundamental biological interest.

In considering the case of heterogeneous catalysis (or chemical contact effects) and the influence of anticatalyzers, the effect of the latter on contact-potentials should be noted. These potentials are affected by many organic substances, especially surface-active compounds;^ the strength of the current in a battery and the rate of the associated chemical effects may thus be decreased. For example, in the formation of precipitation-structures from zinc and Fe under the influence of local circuits, the presence of alcohols, esters, and other surface-active compounds of the anaesthetic groups has a well-marked retarding influence.^ The concentrations required for pronounced retardation are similar to those effective in the above-described forms of anticatalytic action, and the effect may be described as anticatalytic, although its conditions are probably complex, the influence on

' Cf. the papers of Gouy, Abl, Grumbach, Loeb and Beutner cited below. viscosity and on adsorption entering in addition to that on contact-potentials. In a recent review of the facts and theories of contactcatalysis Bancroft^ cites various instances where electrolysis and electrode-potentials are altered by foreign substances. Thus the P.D. at which O2 is freed at a platinum surface is found to be influenced by the electrolytes present. With platinum electrodes oxidations occur more readily at platinized than at smooth surfaces, apparently because of the ''catalytic" action of the finely divided platinum. The presence of cyanide and other compounds reduces the rate of oxidation occurring at an electrode under a given P.D.; for example, a neutral solution of Na2S203 is oxidized to tetrathionate at a platinized anode with a P.D. of 0.44 volts and a current-density of 3X10"'* amperes per square centimeter. If a trace of Hg(CN)2 is added, the anode P.D. for the same current rises to 0.48 volts. Various salts have marked influence on the electrochemical processes at smooth anodes.^ Gouy^ made an extensive study of the effects of various compounds on the surface-tension maxima in the capillary electrometer. Usually this maximum corresponds to a minimal P.D. between the Hg and the H2 SO4; but the P.D. and the surface-tension are both changed by the substance added, so that the position of the maximum is shifted, and this influence was found to be greatest with highly surface-active substances. Similar observations were made by Abl

* Cf. Foerster, Elektrochemie wassriger Losungen (19 15) for further details. on the electromotive force of cadmium-amalgam cells/ In the case of contact-potentials between water and other dielectrics, Grumbach^ found in general that organic compounds which lower the surface-tension of water decrease these potentials. With CH3OH, C2 H5OH, and C4H9OH the effect increases in the order of increasing molecular weight; the curves relating potential change and concentration resemble the corresponding curves of surface-tension and adsorption. The observations of Loeb and Beutner^ on the contact-potentials between organic membranes (apple skin) and electrolyte solutions containing alcohols also show a decrease of P.D. with the addition of alcohol. With the first three alcohols the effect increased was in the order of Ci<C2<C3 (p. 302). Similar results were obtained with solutions of lecithin in guaiacol. The concentrations required to produce a decided influence on the potentials were, however, much higher than the physiologically effective or narcotizing concentrations.

Traube nevertheless regards the influence of surfaceactive substances on the contact-potentials between the living cell and the medium as an important factor in the physiological effect, and MacaUum has expressed a similar view.'* It is doubtful, however, if this can be generally true, since the bioelectric potentials are not necessarily decreased in anaesthesia. For example, in pure sugar solution, which produces in muscle effects 4 A. B. Macallum, Surface Tension and Vital Phenomena, University of Toronto Studies (1912), No. 8, pp. 68 ff.

similar to anaesthesia, the bioelectric P.D. is increased.' The local negativity produced by anaesthetics under certain conditions, as in AUcock's experiments,^ is probably to be referred to the permeability-increasing effect of the compounds in higher concentrations. The connection betv/een adsorption and contact catalytic action has been usually regarded as a special case of the mass-action law. Since any increase in the concentration of a reacting substance involves a proportional increase in reactionvelocity, and since the concentration of many dissolved substances is increased in the layer of solution adjoining an interface, it follows that in any such case there must be increased reactionvelocity in that region. Hence if the adsorbing surface is sufficient in area, as with fine subdivision, a large proportion of the reacting material may be present in the surface layer at any time, and a considerable increase in reactionvelocity may result. Direct parallels between adsorption and catalysis have in fact been observed in certain cases.^

It is, however, questionable if this influence is in itself sufficient to account for the effects observed; in many cases the surface appears to exercise some specific chemical influence; and certain contact catalyzers, especially platinum, have an accelerative action which is far greater than can be accounted for on the ground of their adsorptive capacity alone. "* ^ Cf. Briinings, Arch. ges. Physiol., XCVIII (1903), 241. Macdonald's work also shows an increase of the demarcation potential with decrease in the external electrolyte content. Cf. p. 304.

From the physiological point of view the chief case of specific acceleration by a finely dispersed or colloidal catalyzer is enzyme action. There seems to be no doubt that in many cases the enzyme acts through the formation of intermediate compounds, which break down at high velocity yielding the decomposition products and the original enzyme, the latter then recombining with the substrate molecules to repeat the cycle. The formation of the enzyme-substrate combination presupposes intimate contact between the molecules of enzyme and substrate; hence a correspondence in size and pattern between the intercombining molecules (or parts of molecules; e.g., zymophore groups) is required, of the kind indicated by the ''lock and key" comparison of Emil Fischer. That relations of this kind play a part in adsorption is indicated by Marc's observation that a crystalline adsorbent like BaS04 exhibits preferential adsorption for compounds crystallizing in the same system.^ Apparently this is a simple instance of specific adsorption. Close contact of this kind gives the occasion for chemical transformations (hydrolysis, etc.) which otherwise would not occur, or at much lower velocity. According to Bayliss^ a simple reversible adsorption often precedes a more intimate ''chemical^' union between adsorbent and adsorbed substance. In specific catalyses, like those induced by enzymes, factors of a similar kind probably enter; the union is specific because of the similarity of molecular configuration between enzyme and substrate, and the chemical effect follows because the enzyme-substrate combination is

Catalytic effects due to simple increase in concentration are accordingly to be distinguished from those which depend on the appearance of special relations of some kind between the molecules of the substrate and of the catalyzer. In the latter case the formation of combinations differing from the substrate in reactivity, e.g., velocity of hydrolysis or oxidation, becomes possible; and when these temporary combinations break down, setting free the catalyzer and enabling the latter to repeat the combination, the total effect is the same as if chemical change in the substrate alone were accelerated. The general theory of intermediate compound formation would thus apply to any specific adsorption-catalysis. The adsorbent would correspond to the enzyme, and the theory of such catalysis need not differ in principle from that of catalysis in homogeneous solutions; e.g., H-ion catalysis. At present, however, there appears to be no completely satisfactory theory of this type of catalysis.

With regard to other conditions which may enter in cases of heterogeneous catalysis, it has been suggested that the increased reaction-velocities shown by many substances (sugars, etc.) in protoplasm are phenomena of the same kind as the differences of velocity shown by a given reaction in different solvents, as observed by Menschutkin and others.' Bredig compares a particle of a solid catalytic agent with a drop of liquid, and regards the formation of an adsorption-film as equivalent to

solution;^ in this case different reaction-products might be obtained from the same reaction-mixture under the influence of different catalytic agents, differences of adsorption having the same effect as differences of solubihty. Selective adsorption and selective catalysis would thus be referred to the same conditions, and both referred ultimately to the same conditions as selective solubility. At present there is an increasing tendency to regard both solution and adsorption as special cases of chemical combination. But although protoplasm contains solvents in which it is conceivable that certain reactions may proceed with increased velocities, it does not seem probable that such considerations can explain the high velocities of the biologically more important reactions, such as the oxidation of sugar. Many facts, especially the phenomena of irritability, show that the conditions determining the increase of reaction-velocities in protoplasm are of a special kind and that the factor of organized structure is all-important. It is not simply a case of transferring a substance from a solvent in which its reaction-velocity is low to one in which it is high.

It has been pointed out by various authors that when adsorption is highly selective, displacements of equilibrium may occur; thus an adsorbent may change the H-ion concentration of a solution. Such effects may be attributed to the selective adsorption of ions (Freundlich) ;^ but, as we have seen, the distinction between the adsorption of a substance and its chemical combination with the surface molecules of the adsorbent cannot be sharply drawn. If, in fact, the two processes are indistinguishable, the case becomes essentially one of alteration of equilibrium following the introduction of an additional reagent into a reaction-mixture. Bancroft cites instances where the same compound undergoes different reactions under otherwise similar conditions according to the nature of the contact-agent present;^ thus with colloidal nickel as catalyzer, alcohol forms acetaldehyde and hydrogen, while with colloidal silica or alumina it forms ethylene and water; and he refers this difference in the catalytic action to the differences in the selective adsorptive action of the two substances, nickel adsorbing hydrogen and alumina water. The rate and character of the reaction undergone by a given reaction-mixture may thus vary with the character of the catalyzer, according to the latter's special adsorbent properties.

It has been mentioned that simple adsorption is insufficient to account for the great activity of certain contact catalyzers. Taylor^ points out that charcoal adsorbs carbon monoxide and oxygen but does not catalyze the reaction; but in the case of ethylene and oxygen it both adsorbs and catalyzes; hence as catalyzer, it differentiates between carbon monoxide and ethylene, although adsorbing both. Metallic oxides, on the contrary, catalyze the oxidation of carbon monoxide. It would appear, therefore, that specific, presumably chemical, relationships enter even in such simple cases.

Facts like these (the so-called ''preferential catalysis") may be regarded as e\ddence of conditions which in their higher developments in living organisms appear as the highly selective specificity of many enzymes. The recent work of Langmuir and Harkins^ has shown that molecules assume definite orientations at the surfaces at which they are adsorbed. This orientation is undoubtedly a factor in the chemical effect produced; reactive groups may thus be brought into a position in which they more readily come into contact with other molecules (or the reactive groups of other molecules), and in this manner reaction is furthered. Cases of preferential catalysis may thus be explained. The catalytic effectiveness of phase-boundaries exhibits itself in many remarkable ways. Taylor and Langmuir^ cite Faraday's observation that a perfect crystal of sodium sulphate does not effloresce until its surface is scratched or broken, when the efflorescence spreads from the injured part over the rest of the crystal. Other similar examples are well known; apparently the reactivity of molecules is altered by the adjacent molecules of reaction-product: the use of ''catalystpromoters" in chemical processes illustrates the same phenomenon. Enhanced reactivity at interfaces is in fact a very general phenomenon; and in Hving matter, with its polyphasic constitution, the conditions are exceptionally favorable for this type of influence.

It is probably significant that the substances which effect the greatest variety of contact catalyses, carbon and the metals, especially platinum, belong in the class of metallic conductors/ In such conductors, according to the electron theory, there is ready transfer of electrons from atom to atom, hence their electrical conductivity, and other properties correlated with this peculiarity (optical, etc.). It might be expected that such substances would also faciHtate the transfer of electrons to or from molecules with which they are in contact, and thus furnish the conditions necessary for chemical reaction. In other words, factors characteristic of the metallic state may enter in contact-catalysis; for example, the formation of local electrical circuits between different parts of the metallic surface, or oscillation phenomena of a frequency corresponding with that of the combination-electrons of the interacting substances (resonator eff'ects). An example of the former type of influence was seen in the simple experiment described above, in which the contact of a nobler metal or carbon accelerates or "catalyzes" the formation of ferricyanide filaments from zinc. Combinations of two contact catalyzers seem often to be more effective than either one alone; thus Shenstone found that ''platinized charcoal" was extremely effective in oxidizing alcohol, ''converting spirits of wine into vinegar in a few hours," and other cases of a similar kind are described by Bancroft in a recent review.^ Presumably the two components differ in their potential-difference against the medium,

^ Contrast, e.g., the lack of catalytic power in colloidal silica (cf. Taylor, Trans. Amer. Electrochem. Soc, op. cit., p. 150). and thus local circuits arise which have chemical effects. Rideal and Taylor also describe cases where metallic couples have a greater catalytic effect than either metal singly.' The readiness with wliich the metalKc phase conducts electricity enables any local inequahties in the metal, or in the nature or concentration of materials (e.g., salts and oxidizable and reducible compounds) present in the solution, to give rise to electric currents between different parts of the surface. These local currents may secondarily influence chemical reactions, as in other cases of electrolysis at metalHc electrodes. Effects of the reverse or anticatalytic kind may result when the metal- Hc surface is altered in such a way as to lessen the local potential-differences or increase the resistance of the local circuits; strongly adsorbed organic compounds may have both of these effects, as already pointed out.

Electrochemical oxidations at platinum anodes are subject to variations (partly mentioned above) which are possibly referable to the existence of local circuits; these exercise their own influence independently of the E.M.F. appHed from without, and give rise to interference and summation phenomena of various kinds. The possibiHty that the formation of local circuits plays a part in the normal catalytic acti\'ity of platinum and other metals in liquid systems seems to have been insufficiently considered by chemists. The essential conditions of such catalysis are perhaps best illustrated by the decomposition of hydrogen peroxide, which is effected by a large number of finely divided metals of which platinum is especially active. The striking

parallels between this action and the similar action of organic fluids and cell-extracts have long been known. The decomposition of H2O2 by living tissues is now regarded as due to a special enzyme, catalase. The relation of this catalytic action of tissue-extracts to enzyme action and fermentation was early recognized by Schonbein, who characterized the splitting of H2O2 as a model or prototype of all fermentative processes (Urbild alter Gdhnmgen)^ A study of the conditions under which it occurs in the presence of metals may thus throw some light on the general nature of catalytic effects in heterogeneous systems and especially in Hving protoplasm.

Among other metals mercury shows great activity in decomposing H2O2. The most striking feature of the catalytic decomposition of H2O2 in contact with a mercury surface is that under certain conditions the process exhibits a definite regular rhythm, closely resembhng physiological rhythms like that of the heart beat in its frequency, in its dependence on temperature (Qio = about 2) , and in the influence exerted upon it by chemical and electrical conditions. This phenomenon has lately been the subject of much investigation, specially by Bredig and his students, and its detailed conditions have been studied most closely by Antropoff.^ The chief result of this study has been to show that the catalytic rhythm is dependent on the alternate formation and dissolution of a surface-film of oxidation-product ('^peroxidate") formed by interaction between the mercury and the peroxide. Since evidence from various sides indicates

that many vital processes (stimulation, cell-division) are also dependent on changes undergone by surfacefilms — those forming the plasma membranes and other protoplasmic partitions — this feature of the H2O2 catalysis suggests that the foregoing physiological parallel may be more than a superficial one, and that the similarities depend on a fundamental identity in the conditions controlHng the course of the reactions in the two systems. It is necessary therefore to examine more closely into the nature of the conditions controUing the activity of the Hg-H202 system.

The rhythm is best shown in 10 per cent aqueous solutions of H2O2. The film which forms over the surface of pure mercury in this solution is gold-brown in color, and the conditions for the rhythmical reaction are best when the solution is slightly on the alkaline side of neutrahty. The evolution of oxygen occurs during the breaking down of the film; when the film covers the entire surface of the mercury the reaction ceases. If the solution is slightly acid, the film is stabilized and no evolution of gas occurs; at the appropriate degree of alkaHnity it is alternately formed and broken down with a regular rhythm of about ten to fifteen per minute (at 18°) . The film shows great sensitivity to the presence of foreign substances, including salts and surface-active organic compounds (such as ether and olive oil); the latter abolish the rhythm.^ It is also highly susceptible to changes in electrical polarization. Many other striking parallels with physiological rhythms are described in the article by Bredig and Wilke.^

The fact indicating most clearly the essential nature of the conditions governing the course of the reaction is that the rhythm of decomposition is associated with a parallel rhythm of electrical potential. The mercury in contact with the H^O^ solution is always found cathodal with reference to the calomel electrode, but during the active phase, while O2 is being freed, it is less so than during rest; i.e., the metal becomes anodal relatively to the resting condition. According to Antropoff's measurements, the inactive mercury is about 0.12 volt more cathodal (i.e., nobler) than the active. This difference resembles that between passive and active iron in nitric acid solution, the inactive mercury corresponding to the passive iron. A change of surface-tension accompanies the change of potential, the rounded convex surface of the mercury becoming flatter (from decreased surface-tension) as the film^ forms; this behavior is in accordance with the Lippmann- Helmholtz rule of electrocapillarity, according to which the surface-tension decreases with increase of the potential-difference across the surface. Graphic registration of the curve of potential change shows that its course runs closely parallel with the curve of oxygen evolution as measured by a manometer.

Close observation shows that the gas is evolved only during those times when part of the mercury is filmcovered and part bare; further, that the evolution of gas occurs chiefly near the boundary between the bright and the film-covered surfaces. When a regular rhythm is established, the reaction during each cycle is observed to pass rapidly over the surface of the mercury in a wavelike fashion. At the beginning of a cycle, when the whole mercury surface is fihn-covered and inactive and is flattening as a result of the lowering of surfacetension, a rupture of the film appears, usually at the margin of the drop, disclosing the bright metallic surface beneath. Instantly an effervescence starts at the edges of this fissure and then sweeps over the whole surface of the metal; a new film is then formed and the cycle is repeated. In an inactive drop, an artificial mechanical rupture will often initiate a reaction which similarly spreads rapidly over the whole surface.