MSCELLANEOUS NITROGEN AND SULFUR COMPOUNDS 289 



be used to separate them from other carbohydrates. For isolation of unhydrolyzed glyco- 

 sides it is necessary to inactivate the hydrolytic enzymes with boiling water or alcohol 

 and to neutralize plant acids with calcium carbonate. There is little interest in isolating 

 the hydrolysis products except where they are needed to establish the structure of the orig- 

 inal glycoside. The sugars may be obtained by methods outlined in Chapter 2 and the car- 

 bonyl compounds by distillation. The nitriles which occur along with mustard oils may be 

 separated by adding ammonia sufficient to form a nonvolatile thiourea derivative with the 

 i'so-thiocyanate and then distilling off the volatile, unreacted nitrile. During the distillation 

 if water is present, some of the nitrile may be hydrolyzed to the corresponding acid. 



There is no general method for rapid characterization of all compounds having a ni- 

 trile group. The presence of cyanogenic glycosides in plants is indicated by the evolution 

 of HCN when the tissues are broken. This gas is easily detected by putting the crushed 

 plant part in a sealed tube with a piece of filter paper which has been dipped in alkaline 

 picric acid solution. The yellow dye turns red when HCN comes in contact with it. Simi- 

 lar color reactions will be found in the general references. A few plants contain cyanogenic 

 glycosides but not the enzymes to hydrolyze them. In these cases HCN is evolved only if 

 the required enzymes are added. Other nitriles which do not evolve hydrogen cyanide are 

 customarily identified by hydrolyzing them to the corresponding acids and identifying these 

 acids by methods described in Chapters 3 and 4. 



Little is known about the biosynthesis of the nitrile group in plants. Tracer experi- 

 ments of Conn and Bove (35) have shown that the complete carbon skeleton of the aglycone 

 of dhurrin is derived from tyrosine. Presumably the nitrogen comes from the amino group 

 of tyrosine also. Similarly, Butler and Butler (36) have shown that the carbon skeletons 

 of linamarin and lotaustralin come respectively from valine and isoleucine. This evidence 

 dispels older ideas that the glycosides might be built up by a reversal of the hydrolysis re- 

 action. Since tyrosine probably can also form p-hydroxybenzyl cyanide by way of a mus- 

 tard oil glucoside, there may exist pathways for the formation of two different types of 

 nitriles from the same precursor in different plants. Ahmad and Spenser (37) have shown 

 that the oximes of a-keto acids are readily converted non-enzymatically to the correspond- 

 ing nitriles: 



NOH 



II 



RCH2CCOOH - HCH2CN + H2O + CO2 



They suggest that such a reaction may account for the biosynthesis of certain nitriles. 

 Mentzer and Favre-Bonvin (38) have used this reaction as a basis for a proposed mecha- 

 nism for prunasin biosynthesis and have shown that results of tracer experiments are in 

 accord with it. It is suggestive that no plant apparently has both mustard oil glucoside 

 and cyanogenic glycoside. Possibly they represent alternative approaches to the same 

 function. 



INDOLE DERIVATIVES 



Plants contain a large number of compounds based on the indole ring system: 



4 



•CK) 



