[8 



I 10 SUBCELLULAR PARTICLES 



going into great detail, we may note that the G fraction, i.e., the fraction con- 

 taining recognizable Golgi material, accounted for approximately 6 per cent of 

 the total nitrogen of the homogenate. Lipid phosphorus was concentrated in this 

 fraction nearly five-fold compared to the homogenate, in agreement with histo- 

 chemical studies that have indicated a concentration of lipid material in the Golgi 

 structures of intact cells (see discussions by Bourne (2) and Baker ( i )). Alkaline 

 phosphatase activity ( measured by the release of inorganic phosphorus from 

 ;8-glycerophosphate at pH 9) was studied because of histochemical evidence that 

 this activity was associated with the Golgi regions of the epithelial cells of the 

 small intestine (9, 7, 18). Again we noted a five-fold concentration of activity in 

 the G fraction, and also a three-fold concentration of the P, or small particle, 

 fraction. Additional data (23), not presented here, indicated that the Golgi frac- 

 tion was very active in hydrolyzing several other phosphate esters, such as adeno- 

 sine triphosphate (ATP) and adenylic acid; and in general the evidence sug- 

 gested that high phosphatase activity was one of the outstanding characteristics 

 of the isolated Golgi material. 



Diphosphopy ridine nucleotide-cytochrome c reductase ( DPNH-cy to. c reductase ) 

 activity, which we shall see later is a component of the microsomes of epididymis 

 as well as those of liver (12), was also present in the Golgi fraction. Perhaps the 

 rnost surprising observation was related to the ribonucleic acid (RNA); nearly 

 20 per cent of the entire amount in the homogenate was recovered in the Golgi 

 fraction, at a relative concentration of about 3.5-fold. There had been no previous 

 evidence, either from stained preparations or ultraviolet microscopy, that RNA 

 was associated with the Golgi system in intact cells. 



It was recognized at the time that the Golgi fractions may well have been 

 contaminated by submicroscopic material of differing mtracellular origin (23). 

 In connection with this question, and also because it was desirable to study the 

 morphology of the isolated Golgi material with the electron microscope when 

 reliable techniques for doing so became available, we took up the problem again 

 after a period of several years. It soon became evident that the high RNA con- 

 tent of the Golgi fraction was probably the result of contamination with small 

 ribonucleoprotein particles (19) that had been released from their original at- 

 tachment to the microsomal membranes through the action of the high salt 

 concentration employed in the gradient fractionation. For example, figure 4 

 presents the results of an experiment in which the layer (O) just above the Golgi 

 fraction was examined separately and found to be extremely rich in RNA. Elec- 

 tron microscopy showed that this fraction consisted chiefly of small, electron- 

 dense particles, probably subunits of the original Palade granules. 



It seemed desirable, then, to modify the fractionation procedure so as to 

 achieve a better spatial separation of the large, light Golgi material from the 

 much smaller, but more dense (20), ribonucleoprotein particles. The technique 



