OPSONiyS AND BAGTERIOTROPINS 237 



alkali and hypotonic solutions and certain sodium and potassium salts. In each case, 

 by adjusting the conditions of treatment it was possible to destroy the haemolytic activity 

 without loss of opsonic activity. They concluded that the opsonic system differs markedly 

 from the complement system. Maltaner (1935) suggested that ammonia-treated serum 

 opsonized because the fourth component was supplied by the leucocytes employed in the 

 opsonic mixture. Gordon (1937) confirmed Maltaner 's observations, but maintained that 

 the opsonin was not closely related to complement, since specifically sensitized red cells 

 that have absorbed from ammonia-treated serum all the components of complement 

 excepting fourth component are not lysed in the presence of leucocytes, a source of fourth 

 component. It is possible, however, on the basis of these experiments to conclude that 

 all but the fourth component is concerned in opsonization, especially since PUlemer, 

 Seifter and Ecker's demonstration (1942) that no haemolysis takes place unless C'4 is 

 fixed previously, or simultaneously with C'l. On the other hand, the experiments of 

 Ecker, PiUemer and Kuehn (1942) clearly demonstrate the lack of any constant 

 association, in the sera of a number of mammals, of opsonization with one or other 

 of the four recognized comjDonents of complement. 



Ecker, Weisberger and PiUemer (1942) and Ecker, PiUemer and Kuehn (1942) measured 

 the opsonic action of large numbers of sera from various mammals on staphylococci, 

 both alone and in conjunction with staphylococcal antibody (bacteriotropin). In general, 

 mixtures of heat-labile opsonin and specific bacteriotropin had less opsonic activity than 

 either acting separately. There was no enhancement of opsonic activity except with 

 weak concentrations of opsonin and antibody. Though there was no relation between 

 opsonic power and haemolytic complement titre in any of the species studied, yet the 

 processes inactivating complement also reduced opsonic power. AU the normal opsonins 

 were heat-labile ; those of the guinea-pig, monkey and sheep were inactivated by ammonia ; 

 those of the monkey and sheep by absorption with yeast. The opsonic powers were 

 variously distributed between mid-piece and end-piece fractions, mid-piece predominating 

 markedly in man and guinea-pig, and moderately in monkey and cat, Uttle in sheep and 

 rabbit. In the dog, neither fraction was opsonic, but a mixture of the two was. 



The observation of Gordon and Thompson (1937) that protamine could opsonize 

 bacteria led Gordon and Atkin (1939) to test globin. Globin occurring naturaUy in the 

 body as hsemoglobui is a distinctly basic protein, forms salts with acid proteins like casein, 

 and proved to be a good artificial opsonin. This observation not only offers a possible 

 analogy for a natural opsonin, but clearly shows that body proteins other than those 

 known to be components of complement can act as opsonins. 



The quantitative measurement of opsonic or bacteriotropic action is a technical 

 problem of great difficulty. The system is a very complex one, including living 

 phagocytic cells the uniform distribution of which is exceedingly difficult to ensure. 

 It is impossible, even when the tubes are rotated by one of the various mechanisms 

 that have been devised, to obtain results of the same order of accuracy as in the 

 precipitin, agglutination or haemolytic reactions (see, for example, Hanks 1940). 



In the method employed by Wright and his colleagues, serum, leucocytes 

 and bacterial suspensions are mixed in capillary tubes and incubated at 37° C. 

 for 15-30 minutes. Their contents are then expelled on to slides. Films are 

 prepared and stained, and the bacteria contained in the first 50-100 leucocytes 

 encountered are enumerated. The relative opsonic effect of two sera is expressed 

 as the ratio between the numbers of bacteria taken up, under their influence, by 

 the same number of leucocytes. As might be supposed, the experimental error is 

 a high one (see Greenwood 1913). An alternative method, suggested by Klein 

 (1907), is to fix on some particular degree of phagocytosis as an arbitrary end-point, 

 and to dilute the serum under test until this end-point is reached ; but it seems 

 probable that the experimental error will still be large. 



