458 V. CHEMISTRY OF PHOSPHATIDES AND CEREBROSIDES 



again filtered, and is washed with acetone. Sphingomyelin is extracted 

 from it by the use of a 9 : 1 petroleum ether-methanol mixture. The extract 

 is filtered, concentrated to a small amount, of thin syrupy consistency, and 

 is precipitated with a large excess of acetone (1000-1500 ml.). Since this 

 precipitate still gives a slightly positive Molisch reaction, it is again dis- 

 solved in a 9:1 petroleum ether-methanol mixture and is put through an 

 AI2O3 chromatographic column for selective adsorption of cerebrosides.'^^ 

 Hydrolecithin-free sphingomyelin is then recovered from the concentrated 

 solution by precipitation with acetone and recrystallization from hot ethyl 

 acetate. 



(6) Method 2. The hydrolecithin may be effectively removed from the 

 crude sphingomyelin by extraction with 97% acetone (97 volumes of ace- 

 tone to 3 volumes of water). The crude powder obtained as described in 

 Method 1 is mixed with sand and is extracted for 3 weeks with 97% ace- 

 tone in a Soxhlet apparatus. The sphingomyelin, which remains in the res- 

 idue, is dissolved in 1000 ml. of a 9:1 petroleum ether-methanol mixture; 

 this is filtered and concentrated to a small volume. Sphingomyelin is pre- 

 cipitated with an excess of acetone and is recrystallized from ethyl acetate. 



This method has the disadvantage of giving only a small yield of sphingo- 

 myelin, since most of the desired phospholipid is extracted by the 97% ace- 

 tone, along with the hydrolecithin. 



(5) Tests for Purity of Sphingomyelin 



Because of the similarity of the properties of sphingomyelins to those of 

 the lecithins on the one hand, and of the cerebrosides on the other hand, the 

 preparation of reasonably pure samples presents extreme difficulties. The 

 melting point or the optical rotation is not a suflEiciently precise physical 

 constant to be of much value in identification. The absence of cerebrosides 

 from the samples can be proved in a reliable manner by the application of 

 the Molisch test for carbohydrates, since this is an extremely sensitive test. 

 The other test which has been employed is the N:P ratio, which should 

 give a value of 2:1. However, a considerable quantity of monophospha- 

 tides might be present without significantly altering this proportion. Of 

 considerably greater importance in establishing the purity of a sphingomye- 

 lin preparation is proof of the absence of glycerol. This is possible by the 

 method of microdetermination for this compound which has recently been 

 developed by Blix.^-' When this test was applied to presumably pure 

 sphingomyelin preparations, it was found that it was positive, ^"'^^^ which 

 indicated that considerable amounts of glycerol-containing phosphatides 

 (probably saturated lecithins) were present. 



3" S. J. Thannhauser and P. Setz, /. Biol. Chem., 116, 527-531 (193G). 

 3" G. Blix, Mikrochim. Acta, 1, 75-77 (1937). 



