234 THE ANTIOEN-ANTIBODT REACTIONS 



of hsemolytic antibody, according to the initial concentrations of the various reagents 

 (Heidelberger, Weil and Treffers 1941). The regularity of the molecular ratio in 

 a given system, and the fact that C'l is in excess of hsemolysin, provides quantitative 

 confirmation of the view that the action of complement is not enzymic. However, 

 Haurowitz and Yenson (1943), in confirming the quantitative findings of Heidel- 

 berger and his colleagues, estimated that, while the complement required for haemo- 

 lysis, if spread in a unimoleculaf film on the surface of the red cell, would cover 

 6 per cent, of it, the amount of saponin or oleic acid required for non-specific 

 lysis of a red cell was over 1,000 times the amount required for a complete uni- 

 molecular film. In their view, the high activity of a small amount of complement 

 suggested an enzymic nature. It should be noted that heat inactivation of 

 complement does not destroy all its combining powers (Heidelberger 1941, Pillemer, 

 Chu, Seifter and Ecker 1942). 



By testing the compltmentary activity of all possible recombinations of the various 

 fractions and mixtures of fractions of guinea-pig complement, Pillemer, Seifter and Ecker 

 (1942) have shown that C'l is not the only combining component of complement. C'4 

 and C'2 are also fixed, but contribute little in the absence of C'l. C'l and C'2 are fixed 

 in the absence of C'3 and C'4 (see also Ueno 1938). Little C'3 is fixed, but this acts in a 

 manner similar to a catalyst to produce the complementary effect (e.g. haemolysis) on 

 systems with which C'l, C'2 and C'4 have combined. 



With human comjjlement, they found the fixation of C'2 and C'3 varied with the nature 

 of the bacterium employed as antigen. In the absence of C'l or C'4, very little fixation 

 occurred. In the absence of C'2, C'4 was fixed in large amounts, and in the absence 

 of C'3, both C'2 and C'4 were fixed. The influence of the different components upon 

 fixation of others was, as in the guinea-pig, curiously arbitrary. As with guinea-pig com- 

 plement, all four components were necessary for the two specific comi^lement effects, 

 lysis of sensitized red cells, and death of sensitized bacteria. Neither purified C'l and 

 C'2, nor fractions containing C'3 or C'4, were active alone, though, if fixed to the cell, 

 they were active on the subsequent addition of the remaining components. Bactericidal 

 and haemolytic power of single components and of various mixtures ran exactly in parallel, 

 indicating a close similarity of action on the respective cell surfaces (Dozois, Seifter and 

 Ecker 1943, 1944 ; Seifter, Dozois and Ecker 1944). 



The nature of the fixation is still in doubt. Heidelberger, Weil and Treffers 

 (1941) suggest that complement-combining components differ from normal globulin 

 in being able to form loose, dissociable unions with antibody molecules, and that 

 these unions are stabilized when the components are surrounded by antibody 

 molecules in specific antigen-antibody aggregates. In cases where excess of the 

 component combines in presence of minimal amounts of antibody, a similar loose 

 combination with antigen is postulated. 



Complement is inhibited by various anticoagulants, but they appear to act by reason 

 of their acidic or basic nature or of their large molecular size, and their action does not 

 indicate an immediate connection between coagulation and complement (Wadsworth, 

 Maltaner and Maltaner 1937, Ecker and Pillemer 1941). For example, C'2 and C'4 are 

 unstable in acid solutions, which at the same time depress the ionization of calcium neces- 

 sary for coagulation (Pillemer and Ecker 1941a) ; again, the action of lipoid solvents in 

 reducing complementary activity may be attributed to their independent action on hydrated 

 complement, for the dehydrated complement may be extracted by alcohol and ether 

 without loss of activity on reconstitution with water (Ecker, Pillemer and Grabill 1938). 

 The tanning agent, sodium hexametaphosphate, inactivates complement apparently by 

 an alteration of the groups on which the activity depends, as a result of combination 

 with the basic groups of the proteins concerned (Gordon and Atkin 1941). 



