218 GEORGE H. HOGBEOOM AND WALTER C. SCHNEIDER 



chondria, that the fraction actually consisted of an insoluble residue and 

 that any soluble material present originally in the particles had been ex- 

 tracted during the isolation procedure. (A similar, and still rather contro- 

 versial, situation exists with respect to cell nuclei isolated in aqueous 

 media (cf. footnotes 60, 88).) During the past three years, however, a 

 considerable amount of experimental evidence has been accumulated in 

 support of the conclusion that the mitochondrial membrane not only is a 

 relatively impermeable structure but remains so after the particles have 

 been isolated in isotonic sucrose solutions. Thus approximately 60% of 

 the total nitrogen of suspensions of isolated mitochondria is released into 

 solution when the particles are disrupted either by exposure to sonic oscil- 

 lations^^ ■^'^ or by other means. ^^ Included among these soluble compounds 

 are several enzymes that appear to be localized exclusively in mitochondria. 

 Furthermore, recent investigations have indicated that the membranes of 

 isolated mitochondria are even relatively impermeable to certain polar 

 compounds of low molecular weight. These studies will be discussed in 

 more detail in a later section. 



The existence of a mitochondrial membrane thus seems to be definitely 

 established both by the above biochemical investigations dealing with the 

 properties of mitochondria and by numerous cytological studies. The elec- 

 tron micrographs obtained by Palade,'^ for example, clearly demonstrate 

 the presence of a rather thick limiting membrane. The hypothesis advanced 

 by Harmon" and by Green^^ that the behavior of mitochondria is com- 

 patible with a gel-like structure and does not require the presence of a 

 semipermeable membrane is therefore no longer tenable. 



In several investigations, it has been possible to obtain evidence that 

 certain enzymes associated with mitochondria are not present in the 

 fraction as a result of redistribution and adsorption. Thus when intact 

 mitochondria were mixed with solutions of mitochondrial deoxyribonu- 

 clease and ribonuclease, and the preparations were then centrifuged, 30% 

 of the deoxyribonuclease and none of the ribonuclease was removed from 

 the solutions by adsorption on the mitochondria.^- An equivalent amount 

 of microsomes added to the nuclease solution adsorbed 71 % of the de- 

 oxyribonuclease and 54% of the ribonuclease.^'- Despite the fact that mito- 

 chondria were capable of adsorbing some of the deoxyribonuclease, these 

 findings indicated that, if the observed distribution of deoxyribonuclease 

 and ribonuclease in cell fractions had been an artifact due to redistribution 

 and adsorption, both of the enzymes should have been recovered mainly 



89 G. H. Hogeboom and W. C. Schneider, Nature 166, 302 (1950). 



9« G. H. Hogeboom and W. C. Schneider, J. Biol. Chem. 194, 513 (1952). 



9' G. H. Hogeboom and W. C. Schneider, Science 113, 355 (1951). 



»2 W. C. Schneider and G. H. Hogeboom, J. Biol. Chem. 198, 155 (1952). 



