42 MICROSOMAL PARTICLES 



microsomes, but left ~ 80 per cent of their RNA and ~ 15 per cent of their 

 protein still in sedimentable form. The pellets obtained from DOC-treated 

 microsomes consisted of small, dense particles ~ 10 to 15 rw* in diameter with 

 a small admixture of vesicles. We interpreted the results as indicating that 

 DOC "solubilizes" the membrane and content of the microsomal vesicles while 

 affecting their attached particles to a lesser extent, and we inferred from these 

 experiments that the small, dense particles consist of ribonucleoprotein and con- 

 tain most of the microsomal RNA (~ 80 per cent), while the membrane and 

 content account for most of the protein and almost all the phospholipides of 

 the microsomes [51]. 



We soon found out that, contrary to assumptions then current, the final 

 supernatant still contained numerous structured elements of membranous or 

 particulate nature, and accordingly we made an attempt to separate these ele- 

 ments by further centrifugation of the microsomal supernatant, the fraction 

 usually considered the liquid phase of the cytoplasm or the "cell sap." We iso- 

 lated two successive postmicrosomal fractions ° in the hope that one of them 

 might consist mainly of free particles, but we found the corresponding pellets 

 to be mixtures of smooth-surfaced vesicles (probably derived from the smooth- 

 surfaced part of the reticulum), free particles, and amorphous material. The 

 RNA content of these postmicrosomal fractions was low, and its concentration 

 therein considerably lower than in the microsomes [51]. 



In the case of the liver we succeeded, therefore, in identifying the microsomes 

 as fragments of the endoplasmic reticulum, derived primarily from its rough- 

 surfaced part. We obtained evidence indicating that they are closed vesicles, 

 and that most of their RNA is present as ribonucleoprotein in their attached 

 particles. We did not succeed in isolating the free particles of the cytoplasmic 

 matrix, nor did we obtain a "clean" preparation of smooth-surfaced vesicles. The 

 information on the gross chemistry of microsomal membranes was indirect and 

 relied on subtractions and many assumptions. 



PANCREATIC MICROSOMAL AND POSTMICROSOMAL FRACTIONS 



As the next object for testing our hypothesis, we chose the pancreas of the 

 guinea pig with the following considerations in mind. The exocrine cells, 

 which form the bulk of the cell population of the gland, have an endoplasmic 

 reticulum remarkable in its large volume and in the small extent of its smooth- 

 surfaced part (figs. 1 and 2). In addition, their cytoplasmic matrix contains free 

 particles in great numbers (fig. 2). With the new material, therefore, we stood 

 a better chance to obtain a more homogeneous microsomal fraction and to 

 separate free particles in a postmicrosomal fraction. The results met our expecta- 



9 The first postmicrosomal fraction (PMj) was obtained by centrifuging the microsomal 

 supernatant for 2 hours at 105,000^; the second postmicrosomal fraction (PM 2 ), by 

 centrifuging the supernatant of PM : for 16 hours at 105,000 g. 



