256 p. N. CAMPBELL 



The newly synthesized albumin can be released from the membranes 

 either by treatment with deoxycholate (DOC) or by breaking them up with 

 ultrasonic vibrations. Our first idea was to determine whether the protein 

 released by DOC which appeared to be serum albumin by its immuno- 

 logical properties, had in fact all the properties of serum albumin. 



We have examined extracts of the microsome pellet obtained either by 

 DOC extraction or ultrasonics. Immunoelectrophoresis on agar gel was 

 used with an antiserum prepared against a purified preparation of rat serum 

 albumin. We can find no evidence for the existence of precursors of 

 albumin with a molecular weight of less than that of the whole molecule. 

 This finding is of special relevance in the case of serum albumin since 

 Lapresle [2] and Porter [4] have both shown that degradation products 

 of serum albumin retain some of the immunological properties of the 

 native protein. 



What can be learned from the nature of the binding between the 

 membranes and serum albumin ? As already mentioned, if the microsome 

 pellet is treated with ultrasonics there is a certain release of serum albumin. 

 If we centrifuge down the residue and treat with deoxycholate we get a 

 further release of albumin. The comparable amounts are approximately 

 two-thirds by ultrasonics and a further one-third by DOC. We are at 

 present carrying out studies with the electron microscope to determine the 

 nature of the action of ultrasonics. These studies are far from complete 

 but we are accepting as a working hypothesis that the ultrasonic vibrations 

 disrupt the membranes of the vesicles and permit the soluble contents to 

 be released. The albumin which is only released by DOC we assume to 

 be particularly closely associated with the lipid substances that go to make 

 up the fabric of the membranes. 



In order to determine whether there was any metabolic difl^erence 

 between the albumin in the two fractions we injected a rat with a mixture 

 of p^C]-labelled amino acids and killed it after 3 min. We homogenized 

 the liver, isolated the microsome fraction and washed it free of the cell 

 sap (105 000 g supernatant). We then determined the specific activity of 

 the albumin in the two fractions. We found that the DOC albumin was 

 40% more active than the ultrasonic albumin whereas at this short time 

 after injection the cell sap albumin was not significantly radioactive. We 

 have also done similar estimations after the incubation of isolated liver 

 microsomes with P^C]-leucine. In this case the DOC albumin was about 

 four times as radioactive as the ultrasonic albumin. 



We cannot at present interpret these results with certainty but they 



are consistent with the idea that some stage of the synthesis of albumin 



involving incorporation of amino acids takes place in the lipid membrane. 



Next we must turn to the ribonucleoprotein (RNP) particles. Dr. 



Siekevitz has described in his paper experiments with gumea-pig pancreas 



