December 4, 1902] 



NA TURE 



i'5 



The aim of the present investigation was to study the haemo- 

 lytic action of tetanus lysin and its reaction with antilysin in 

 the light of ordinary chemical reactions, and to compare both 

 these phenomena with similar actions brought about by sub- 

 stances of known molecular weight, constitution and purity. 



The method of estimating the hemolytic power, which was 

 employed in all the experiments, consisted in allowing the 

 substance under examination to act for a given time upon 

 an emulsion in normal saline, or other liquid, of a known 

 quantity of well-washed blood corpuscles, and then estimating 

 the amount of haemolysis produced colorimetrically by com- 

 parison with standard tubes prepared from varying quantities 

 of the same blood by complete haemolysis with distilled water. 



The investigation falls naturally into two parts, the first of 

 which deals with the hemolytic action of tetanus lysin compared 

 with that of caustic soda and ammonia. 



The haemolysis of a blood corpuscle by a base such as caustic 

 soda or by tetanus lysin is a phenomenon of considerable com- 

 plexity and appears to take place in two stages — the combina- 

 tion of the haemolytic agent with the material of the corpuscle, 

 and the haemolysis of this compound by the "lysin" which 

 remains free. The three substances under investigation differ 

 from each other in the rate at which they unite with the cor- 

 puscles and also in the stability of the compounds which are 

 produced. 



Caustic soda combines very rapidly and forms a very stable 

 compound ; the consequence of this is that when a certain 

 definite number of blood corpuscles are present, practically the 

 whole of the alkali is taken up and very little haemolysis occurs. 

 With small amounts of blood, haemolysis is complete, but as 

 the amount of blood is increased beyond the amount which 

 can be completely haemolysed, the alkali is thereby withdrawn 

 in increasing amounts from the solution, so that the extent of 

 hremolysis rapidly diminishes. Tetanolysin, on the other hand, 

 combines much more slowly with the corpuscles and forms a 

 much less stable compound, which is partially decomposed into 

 its constituents, or hydrolysed, by the water of the solution. 

 Hence, in the case of the lysin solutions, there is always some 

 free lysin to effect the haemolysis of the lysin-corpuscle combina- 

 tion, and, as a consequence, the falling off after the maximum 

 is not nearly so marked. Ammonia takes up a position inter- 

 mediate between caustic soda and lysin. 



All these haemolytic actions are affected by the presence of 

 certain foreign bodies, among which salts, albumin and serum 

 have hitherto been examined. It seems probable that salts 

 have two distinct effects. In the first place, they probably 

 render the corpuscles more susceptible to the attack of the 

 haemolytic agent, and hence tend to increase haemolysis. This 

 tendency is not counteracted in any way in the case of the 

 tetanus lysin, and hence an increase in the action is' in this 

 case observed. The compounds of the alkalis with the cor- 

 puscles, on the other hand, are affected by salts containing 

 the same ion, much in the same way as a weakly dissociated 

 salt, in which case the dissociation is decreased and the salt 

 then enters less readily into reaction. Hence the caustic-soda 

 combination is affected in this way by sodium salts and, since 

 the diminution of haemolysis thus produced outweighs the 

 increase due to the effect of the salt on the corpuscles, a nett 

 decrease of action is observed. The ammonia combination is 

 less strongly dissociated than the soda combination, and is 

 therefore still more strongly affected by the presence of 

 ammonium salts. 



The dissociation spoken of in this case is the electrolytic 

 dissociation of a salt or salt-like compound into its ions, and 

 must not be confused with the hydrolysis mentioned above. 

 Thus a salt-like sodium carbonate is at the same time partly 

 dissociated into its ions, and partly hydrolysed by the water of 

 the solution into caustic soda and carbonic acid ; sodium 

 chloride, on the other hand, is much more completely dis- 

 sociated into its ions, but is practically not hydrolysed at all. 



The effect of egg albumin and normal serum is also to diminish 

 the haemolytic power, both of the bases and of tetanus lysin, 

 but whilst the effect on the bases is very slight, that on the 

 lysin is considerable. It would seem that in each case the 

 albumin combines with the haemolytic agent, forming a com- 

 pound in which the haemolytic power is modified to a certain 

 extent. The properties of caustic soda and ammonia are only 

 slightly affected ; those of the tetanus lysin, on the other hand, 

 are more profoundly modified. This explanation is confirmed 

 by b the fact that the further addition of albumin exerts no 



NO. 1727, VOL. 67] 



appreciable effect after a certain limit has been reached. Normal 

 serum, on the other hand, has a progressively increasing effect 

 on the lysin, and in fact behaves like a mixture of a large amount 

 of albumin with a small amount of an antitoxin. 



Further information is gained as to the nature of the haemo- 

 lytic action by the determination of the velocity of the change, 

 and this reveals a still greater complexity. The reaction 

 exhibits a very decided period of induction ; when the sub- 

 stances are first mixed, the change begins to take place at a very 

 low rate, which gradually increases as the change proceeds. 

 Such a period of induction occurs in certain well-known chemical 

 reactions, although its exact significance is not perfectly under- 

 stood. In the case under consideration, the authors suggest 

 that it "depends on the circumstance that the red blood cor- 

 puscles' cellular membrane must be destroyed before haemolysis 

 can occur." The actual velocity is found to be proportional to 

 the concentration of the haemolytic agent, so that if the dose 

 be doubled, the time required to produce a given amount of 

 haemolysis is halved. This result is of great importance because 

 it shows that the haemolytic action of bases is not due to the 

 hydroxyl ions, in which case the velocity would be proportional 

 to the square root of the concentration. The same thing is 

 shown by the fact that ammonia acts more rapidly than caustic 

 soda, although it is much less strongly dissociated. 



The second part of the investigation deals with the important 

 subject of the action of the antilysin on tetanus lysin. 



When increasing quantities of antilysin are added to a fixed 

 amount of lysin, the haemolytic power of the mixture is not 

 diminished in direct proportion to the amount of antilysin 

 added, but the effect of each successive portion of antilysin is 

 less than that of the preceding one, the diminution of haemo- 

 lytic power being rapid at first and then becoming more and 

 more gradual. If the results be plotted with the amounts of 



S 



4 

 z 



05 10 15 



■ C C. of Antitoxin Solution . 



antilysin added as abscissae and the haemolytic powers of the 

 resulting mixtures as ordinates (the amount of lysin being con- 

 stant throughout), a curve of the form shown above results. 

 This curve represents what is usually know as the toxin spectrum 

 of Ehrlich. 



When we compare this phenomenon with the action of an 

 acid on an alkali, we find that it does not resemble what occurs 

 when an equivalent of hydrochloric acid is added to caustic 

 soda, for in this case the alkalinity diminishes in direct propor- 

 tion to the acid added, the last portion of acid having exactly 

 the same neutralising effect as the first. 



On the other hand, it corresponds precisely with the phenomena 

 observed when a base such as ammonia is treated with a weak 

 acid, like boric acid. In fact, if ammonia be treated as a lysin 

 and boric acid as an antilysin, and haemolytic experiments be 

 made in precisely the same way as with tetanus lysin and anti- 

 lysin, the curves of haemolytic power produced in the two cases 

 are of precisely the same kind. Now the phenomena which 

 occur when boric acid is added to ammonia and in similar cases 

 have been carefully examined by physical chemists, and they 

 are known to be due to the fact that, in a solution of this kind, 

 the ammonium borate which we should expect to be formed is 

 partially hydrolysed by the water into its components, so that 

 the liquid contains ammonium borate, water, free ammonia 

 (ammonium hydrate) and free boric acid. The case is sus- 

 ceptible of mathematical treatment according to Guldberg and 

 Waage's law, and the equivalents of the substances and the co- 

 efficient of dissociation can be calculated from the observations. 



