Protoplasmic Action and Nervous Action

according to the partition-ratios. Any tendency to concentrate at surfaces (e.g., the general cell surface or the surfaces of other hpoid-containing protoplasmic structures) would be favorable to such solution; i.e., would render it more rapid and greater in degree than it would be otherwise, since the partition-equilibrium would then be between the solution in contact with the surface of the lipoid particle and the solution in the interior of the particle.

Meyer's experiments on the influence of temperature on the critical anaesthetizing concentrations of certain compounds^ gave further indications that the solubility of these compounds in the cell-lipoids is the essential factor in the physiological effect. He chose six compounds whose oil-water partition-ratios vary considerably with temperature, and determined the minimal concentrations required to anaesthetize tadpoles at the two temperatures 3° and 30°. These concentrations are given in the following table :

In the first three compounds (A) the relative HpoidsolubiHty decreases with rise of temperature, and the ^ Meyer, Arch. f. exper. Path. u. PharmakoL, XL VI (1901), 3'38. Recently Mary E. Collett has investigated the effects of variation of temperature on the narcotic action of various compounds on marine organisms; cf. Proceedings of the Society for Experimental Biology and Medicine, XX (1923), 259. critical narcotizing concentration increases; in the last three (B) the conditions are reversed. Tadpoles completely anaesthetized at 30° in m/250 chloral hydrate revive and become active on cooling the solution; on warming they again become inactive. Such an effect is difficult to explain except on the basis of the greater lipoid-solubility of the anaesthetic at the higher temperature, since adsorption is in general increased by lowering the temperature. Meyer therefore concludes that the anaesthetic produces its effect by dissolving in the celllipoids. Such experiments seem to indicate clearly that the solvent action of the protoplasmic lipoids is a main factor in anaesthesia; and since anaesthesia has close affinities with normal variations of irritabiUty, such as sleep and fatigue, they also point to the conclusion that under normal conditions the lipoids are important in the cell largely because of their peculiar properties as solvents.

The relation of the solvent properties of lipoids to permeability has already been considered. The Upoids thus represent the organic solvents of the cell, and apparently the properties of the protoplasmic system vary according to the nature and concentration of the substances which they hold in solution. This is a conclusion of much general interest, apart from its special relation to anaesthesia, since variations in the proportions of water-soluble to lipoid-soluble substances no doubt occur constantly in living protoplasm. The question of why such variations alter irritability, spontaneous activity, and metaboHsm will be considered more fully later in connection with stimulation.

In many cases it has been shown that the organic anaesthetics accumulate in cells in greater concentration than in the medium, and this fact is usually interpreted as favoring the partition theory of narcosis. Pohl^ found that the blood corpuscles and brain of deeply narcotized dogs contained from three to four times as much chloroform as the serum; Hedin^ investigated cryoscopically the distribution of alcohols, aldehydes, ketones, esters, and ether between corpuscles and plasma, and found that most compounds, especially the more highly lipoid-soluble, collected in higher concentration in the cells than in the plasma. More recently Warburg and Wiesel^ have made similar observations on the blood corpuscles of birds, using alcohols, ketones, urethanes, thymol, formaldehyde, and HCN. All of these compounds, in appropriate concentrations, were found to decrease oxygen consumption, and the tendency to concentrate in the cells ran closely parallel with this effect. Thus the lower alcohols (up to butyl alcohol) were the least effective in reducing oxygen consumption, and showed correspondingly a somewhat lower solubility in the protoplasm than in the medium; methyl ure thane, diethyl urea, and acetone behaved similarly; amyl alcohol and isobutyl urethane were more effective, and were about equally distributed between cells and medium; while the most effective substances — phenyl-methyl ketone, thymol, phenyl urethane — ;all showed decided concentration in the cells. When solutions were used that decreased the oxygen consumption by 50 per cent, phenyl-methyl ketone was found to be about twice, phenyl urethane three times, and thymol nine times as

concentrated in the cells as in the medium. Formaldehyde and HCN also underwent concentration in the cells. The series, methyl alcohol < butyl alcohol <amyl alcohol < phenyl -methyl ketone < phenyl ure thane < thymol represents the order both of increasing physiological action and of increasing concentration in the cells. On the whole the foregoing studies of the distribution of narcotics between cell and medium appear to favor the partition theory of the action of these compounds. It should be pointed out, however, that the order of relative adsorption is in general the same as that of Hpoidsolubihty. The probabiKty, as already pointed out, is that both solution and adsorption are factors in the total effect.

The relation between the narcotic or other physiological action of organic compounds and their influence on surface-conditions has recently received much investigation. The majority of compounds of the lipoidalterant group have a marked influence in lowering the surface-tension at the boundary-surfaces between their aqueous solutions and air or other adjoining phase; at the same time, in accordance with the Gibbs-Thomson rule, they undergo increase of concentration (or adsorption) at such surfaces. In homologous series both the influence on surface-tension and the degree of adsorption increase progressively with increase in molecular weight; and the view that the physiological action is determined by this surface-action, i.e., by the condensation of the compounds at the protoplasmic phase-boundaries, rather

than by solution in the Hpoids has been supported by many recent investigators, especially Czapek, Traube, and Warburg/ The general fact that the physiological action of homologous compounds increases progressively with increase in molecular weight has long been noted. Richardson,^ in 1869, in a study of the pharmacological action of alcohols, called attention to this rule, which applies also to the effects on lower organisms; thus, according to Regnard,^ the first six alcohols have equal effects in suppressing the growth of yeast in the following concentrations :

In a series of papers beginning in 1904, Traube has directed special attention to the parallelism between surface-activity and physiological action.'* For example, 2 Richardson, "Physiological Researches on Alcohols," Medical Times and Gazette, VIII (1869) (cited from Czapek, loc. cit.). 3Regnard, Compt. rend. Soc. Biol., X (1889), 124. Warburg and Wiesel obtained similar results with the series of ure thanes {loc. cit.). the surface-tensions (against air) of 0.25 m aqueous solutions of alcohols decrease in regular order as follows:

The order of relative adsorption by charcoal and other adsorbents is similar; if the degree of adsorption runs parallel with the lowering effect on surface-tension, and if equal adsorption corresponds to equal physiological action, we should expect that solutions of the same surface-tension (isocapillary solutions) would have the same physiological effect. Czapek and Traube have in fact demonstrated a close parallelism between the influence of a large number of compounds on airwater surfacetension and their physiological action. Traube has called attention to the fact that in a homologous series of compounds the degree of activity, both physical and physiological, increases very generally about three times with each increase in molecular weight. According to this rule the isocapillary concentrations of the successive compounds of the series should diminish in geometrical progression, with one-third as exponent, as the molecular weight increases. He gives the following determinations for the series of alkyl acetates :

In a considerable number of cases the degree of physiological action has been shown to follow a similar rule; the observations of Fiihner on haemolysis (cited below) are a good example. If physiological activity is in fact a function of capillary activity, solutions of equal surface-tension should exhibit equal narcotic action or otherwise produce equal effects in protoplasm; and Czapek has brought forward evidence that this is very frequently the case.' Using a large ninnber of surface-active organic compounds, he determined the surface-tensions of those solutions which had equal effect in liberating tannin from plant cells (chiefly the leaves of Echeveria) ; this effect depends on a permeability-increasing action analogous to that accompanying cytolysis. In general he finds that solutions of a concentration just sufficient to cause exosmosis of tannin have very nearly the same surface-tension against air; viz., about two-thirds that of pure water; according to his hypothesis, the surface-tension of the protoplasm becomes zero in such solutions and an effective surface of separation ceases to exist. Kisch^ also finds that iso capillary solutions of alcohols have equal effects in liberating invertase from yeast and molds and in inhibiting the growth of yeast cells; and H. Zuckerkandl^ has observed a similar relation in the protoplasmic streaming of plant cells. According to Traube and others, haemolysis follows the same rule.^

The following table summarizes some of the results of Czapek and Kisch with alcohols. Critical surf ace -tensions (water = i ) of solutions causing Czapek's ascription of the effects which he observes to a definite or critical lowering of the surface-tension of the plasma membrane is, however, of doubtful validity, since there is no necessary parallelism between the influence of a given substance on the surface-tension at a water-air interface and its influence on the tensions at other interfaces. This has recently been pointed out by Lorant;^ for example, in comparing the tensions exhibited by various liquids in contact with air and with water, respectively, Lorant finds the following:

Lorant also made observations on the surfacetensions between various organic fluids (e.g., ether) and salt solutions. Usually the influence of neutral salts on surface-tension was in the direction of an increase. Of the different anions CI has the greatest effect, and I and CNS the least; with ethyl ether and nitrome thane, the chlorides, sulphates, and bromides increased the interfacial tension, while the iodides and thiocyanates decreased it. Similar conditions were found with CHCI3 and CCI4, but in this case iodide also somewhat increased the surface-tension.

It would appear that the physical relations (of adhesion, mutual solubility, etc.) between water and the organic compound, as well as between the latter and the non-aqueous protoplasmic phase or structure (e.g., membrane) are of importance in the physiological effect. According to Traube the narcotic action of organic compounds is determined by what he calls their "Haftdruck" (''adhesiontension"), i.e., special attraction to or affinity for water ;^ the tendency of any compound to pass out of aqueous solution and concentrate in the surface layer between water and the other phase — i.e., to undergo adsorption — is in general the greater the less its affinity for water. This is one manner of interpreting the relation noted by Richet and others between waterinsolubility and narcotic action; but since solubility in water and solubiUty in organic solvents — e.g., in the esters of higher fatty acids which form the organic solvents of protoplasm — have similarly reciprocal relations, this consideration does not enable us to decide whether a solution-effect or an adsorption-effect is the essential factor in the physiological action. The recent investigations of Langmuir and Harkins on adsorption^ indicate, how-

ever, that there is no fundamental difference between these two processes; the affinity for water seems dependent usually on the terminal or polar group of the organic compound (COOH, NH2, OH, etc.), and an adsorbed compound may be one in which part of the molecule has an affinity for (equivalent to solubility in) water, while the other part has not, but is attracted more strongly by the other phase. In such cases the position at an interface may be the chief position of equilibrium, and the predominant effect may be adsorption, with limited solution in either phase. Such a view implies that the transition from adsorption to partition is a continuous one, and explains why highly surface-active compounds usually have high lipoidwater partition-coefficients. Such compounds will enter into solution in the non-aqueous phase, provided this is also a solvent. They may, however, condense at the surface of material in which they do not dissolve, and in so doing influence chemical action at such surfaces. Traube calls attention to the fact that the catalytic action of finely divided non-solvent materials like carbon and platinum may be thus influenced; and he places narcotics in the class of ''anti-catalysers";^ i.e., they are regarded as decreasing the catalytic and hence the chemical activity of living matter by some form of surface-action, e.g., by occupying the interfacial positions (where chemical activity appears to be greatest) in the heterogeneous protoplasmic system and displacing the chemically reactive compounds.^ This view, while partial, may well be correct in certain cases, although it

probably does not cover the entire range of phenomena included under narcosis. The fact that narcotic compounds arrest spontaneous activity, and in general act as depressants of vital processes and of irritability, shows that they interfere with the energy-yielding chemical reactions of protoplasm. The essential problem relates to the means by which this effect is produced, whether it is primary or secondary; i.e., there are the alternative possibilities: (i) that the primary action may be a modification of the structural conditions on which the chemical reactions depend; and (2) that the reactions themselves may be influenced directly; e.g., by some form of anticatalytic action.

Warburg and his associates have made an extensive study of the influence of narcotizing compounds on the oxygen consumption of Kving cells, and the results of this work show many striking parallels with those already described. The cells used in the various determinations included sea-urchin eggs, erythrocytes (chiefly of birds), yeast, lymphocytes (from thymus), spermatozoa (of fishes), liver cells (of frog and mouse), and bacteria (Vibrio, Staphylococcus, Bacillus Typhi) .^ In all cases the rate of oxygen consumption was decreased, reversibly, in the presence of a sufficient concentration of anaesthetic. The facts point in general to some kind of physical rather than specifically chemical interference with the oxidation reactions. Thus the effect produced by a particular compound is largely independent of the chemically

' For a summary of this earlier work of Warburg, see Milnch. med. Wochenschr., LVIII (191 1), 289; also Warburg and Wiesel, loc. cit. characterizing group; e.g., two nitriles may be of very unequal effectiveness — require different concentrations to produce the same degree of depression, although possessing the same polar group — and the same is true of other compounds. In any series the depressant action on oxidation is greater with the higher members of the group; and the relative effectiveness of the different compounds is closely similar to that observed in experiments on narcosis. For example, the several alcohols were found to lower the oxygen consumption of birds' erythrocytes by about 50 per cent in solutions of the following concentrations;^ Overton's determinations of the minimal anaesthetizing concentrations of the same compounds for tadpoles'" are cited for comparison:

As an example of experiments with bacteria {Vibrio Metschnikovii) the following series may be cited; to diminish oxygen consumption by about half the following concentrations of ure thanes were required: The relative effects of these compounds on the anaerobic growth of yeast were similar; the following solutions produced about the same degree of inhibition : These results^ are similar to those of Regnard with alcohols, cited above. Usui, working under Warburg's direction,^ found also a decrease in the oxygen consumption of vertebrate tissues (liver, central nervous system) under the influence of narcotic compounds (alcohols, ketones, urethanes, methyl urea, and phenyl urea) ; but in order to produce marked depression of oxidations much higher concentrations were required than in normal reversible narcosis, and the effect was imperfectly reversible. This result is interesting as indicating that anaesthesia is not necessarily associated with a decrease of intracellular oxidations, as Verworn and others have supposed; in fact, Warburg, Winterstein, Loeb and Wasteneys, and others have shown in a number of instances that anaesthesia when perfectly reversible does not necessarily involve a decrease in oxygen consumption.^ Diminished oxidation is to be regarded rather as a secondary consequence than as a cause of narcosis. Apparently the chemical

effect is secondary to some physical modification produced in the protoplasm by the narcotizing compound.^ Certain definite changes in the physical properties of protoplasm, analogous in many respects with those produced by salts, have been observed in various cases to accompany the action of narcotizing compounds; these changes indicate that underlying narcosis there are definite modifications of the structural conditions in protoplasm; and presumably it is to such modifications that the changes in physiological properties and activity are to be referred. As we have seen, the distinctively vital processes are controlled by structural conditions; structural change impHes physiological change.

' For a more detailed discussion of the relation of narcosis to oxidation processes see my review, "The Theory of Anaesthesia" {Biological Bulletin, XXX (1916), 311, also American Yearbook of Anesthesia, I, i). According to Warburg (cf . his recent article on the physical chemistry of cell-respiration, Biochem. Zeitschrift, CXIX [1921], 134) the protoplasmic oxidations occur at the surface of the solid cell structures, which adsorb the water-soluble oxidizable compounds; narcotics influence oxidations by changing the physical and chemical character of the surfaces.

He expresses his general conclusions on the conditions of protoplasmic oxidations as follows: "Two chief means are employed by the cell to diminish the chemical resistance at the regions of oxidation; namely, adsorption and the catalytic action of heavy metals .... Cell respiration is a capillary process occurring at the iron-containing surfaces of the solid cell-constituents. By adsorption at these surfaces the inert organic compounds become capable of reacting with O2 just as do aminoacids at the surface of charcoal. This view does not explain respiration in the physical sense, but classes it with general phenomena of the

inorganic world Narcotics check the cell-oxidations by occupying the surfaces and thereby displacing the oxidizable compounds. The same action is exhibited by different narcotics when the same fraction of the active surface is occupied by the narcotic" (pp. 152, 153). Changes of permeability, of viscosity, and of resistance to the action of cytolytic or other injurious conditions are the most evident physical effects produced by lipoid-alterant compounds in living protoplasm. During narcosis there appears very generally to be a decrease of permeabihty, an increase in the resistance to structural breakdown or cytolysis, and an increase of protoplasmic viscosity. From the general nature of these changes it would seem that the structural substratum of the Uving matter assumes temporarily a denser or physically more stable condition. In any event it is clear that the modification is in such a direction as to interfere with stimulation, a process, which (as we shall see later) involves structural changes in the irritable system. Hence what may be described as a stabilization, decrease of susceptibiHty to structural change, is indicated as in all probabiHty the essential physical condition underlying narcosis; but any such general term gives little indication of the detailed nature of the physical modification produced in the living protoplasm. The manner in which narcosis modifies stimulation-processes will be considered in more detail later; in the present section we shall merely describe briefly those changes in the physical state of protoplasm which appear to have some bearing on the question of the nature of the conditions determining narcosis.

The changes observed in the larva of Arenicola are simple and instructive; normal larvae transferred from sea water to pure isotonic NaCl solution undergo stimulation and marked increase of permeabihty as shown by loss of pigment; on the other hand, larvae which are first placed in a solution of a magnesium salt, or in sea water containing a suitable anaesthetic (ether, chloretone, alcohol) in the narcotizing concentration, and are then transferred to NaCl solution (preferably containing anaesthetic), show no such effect; there is no immediate loss of pigment and Httle or no stimulation. The breakdown of cilia in the NaCl solution is also prevented.' A protective (antitoxic) or stabilizing action is associated with the narcotizing action in this organism; and essentially similar conditions have been found in the eggs of the sea-urchin Arhacia.^ Analogous observations have been made on other cells by a number of investigators. Arrhenius and Bubanovic found that the breakdown of blood corpuscles in hypotonic media was hindered by anaesthetics;^ similar observations have more recently been made by Linzenmeier and Runnstrom; haemolysis and agglutination by foreign proteins may also be diminished by anaesthetics."* These '^ stabilization" effects are observed in the concentrations corresponding to the anaesthetizing range; higher concentra-

3 Publications of Nobel Institute (1913), No. 32, cited from Hober's Physikalische Chemie der Zelle u. der Geivebe, p. 466. 4 See Linzenmeier, Arch. ges. Physiol., CLXXXI (1920), 169, and CLXXXVI (1921), 272; for similar observations on bacteria cf. Vorschiitz, ibid., CLXXXVI (1921), 290; Runnstrom, Biochem. Zeitschrift, CXXIII (192 1), I. See also the observations of Traube {Biochem. Zeitschrift, X [1908J, 371) and Clowes {Proceedings of the Society of Experimental Biology and Medicine, XI [1913], 8). These changes of properties resulting from the modification of surface-films have an interesting relation to those produced by addition of proteins to suspensions of bacteria and blood corpuscles, and recently investigated by Northrop and de Kruif {Jour. Gen. Physiol., IV [1922], 655), Eggerth and Bellows {ibid., p. 669), and Coulter {ibid., p. 403).

tions produce irreversible structural change or cytolysis. Such facts indicate, in general, that in the narcotizing solutions the plasma membranes become more resistant to alteration. Hober's observations on the action of anaesthetics in hindering the production of injurycurrents in muscle by potassium salts illustrate the same condition; the local negativity (an index of increase of permeability) develops much more slowly in the presence of ether, urethane, and other narcotizing compounds than in the pure solution/

A decrease of permeability to water-soluble diffusing substances and ions, and also in some cases to water, is an effect of a related kind. The entrance of watersoluble dyes into plant cells {Spirogyra) is retarded during anaesthesia;^ neutral salts may also produce this effect both in animal and plant cells.^ Decreased permeability to ions is indicated by decreased electrical conductivity; this has been demonstrated by Osterhout in Laminaria, by McClendon in sea-urchin eggs, and by Joel in blood corpuscles. "* In Laminaria a reversible decrease of conductivity is found only in moderate concentrations of ether and other anaesthetics, corresponding to the anaesthetizing concentrations; stronger solutions cause marked and irreversible increase of

conductivity, indicating a destructive or cytolytic effect. In fertilized sea-urchin eggs, anaesthetics (chloral, hydrate, alcohols, urethane) decrease the permeability to water, as shown by the decreased rate of shrinkage in hypertonic sea water containing the anaesthetizing compound.' The penetration of acids into the pigmentcontaining mantle cells of nudibranchs is also retarded by anaesthetics.^ Changes of protoplasmic viscosity, as indicated by changes in the readiness with which cell structures are mechanically displaced (by centrifuging), have also been observed, but the character of the change appears to vary in different forms of protoplasm. In plant cells, according to the observations of Heilbronn^ on seedHngs and F. Weber^ on Spirogyra, ether in the anaesthetizing concentrations increases the viscosity of the protoplasm; in lower concentrations, on the other hand, it decreases viscosity. This result agrees with the observations of Ewart^ and others who find that weak solutions of anaesthetics accelerate protoplasmic streaming while stronger solutions retard or arrest it. In sea-urchin eggs L. Heilbrunn^ has recently found that various anaesthetics in concentrations sufficient to prevent cell-

s Ewart, On the Physiology and Physics oj Protoplasmic Streaming in Plants, Oxford (1903). Cf. also the observations by Demoor and others cited in Czapek's Biochemie der Pflanzen, Jena (1913), p. 161. division cause decrease of viscosity, i.e., facilitate the displacement of granules by the centrifuge. In some cases, however, Heilbrunn found effects of the opposite kind; and he distinguishes two kinds of anaesthesia, in which protoplasmic viscosity is respectively increased and decreased. The significance of such changes in relation to the functional activity of the cell is not clear. They show, however, that the structural conditions within the protoplasmic system are modified reversibly by the lipoid-solvent group of compounds and that the concentrations required for this effect correspond to those which produce narcosis.

Apparently the most general inference to be drawn from the foregoing facts is that one constant accompaniment of narcosis is a modification, in the direction of greater stability or impermeability, of the physical state of the plasma membrane of the irritable cells; and there is good reason to believe that the physiological effect of narcotic compounds depends on this effect; this conclusion will receive further support when the subject of stimulation is discussed. The chief locus of action of anaesthetics thus appears to be the same as that of salts, which is evidently superficial, as already pointed out. Antagonisms between salt action and anaesthetic action are in fact readily demonstrable in many cases. ^ Salts like NaCl in pure solution tend to disintegrate the plasma membranes, and the addition of an anaesthetizing compound to the solution frequently retards or prevents