534 2. ANALOGS OF ENZYME REACTION COMPONENTS 



Finally, the arguments of Gubler (1961), that pyrithiamine must induce 

 disturbances other than in o;-keto acid metabolism, I feel are not entirely 

 valid. He states that since the a-keto acid oxidase activities are appreciably 

 lower in thiamine-deprived rat livers than in the livers of analog-treated 

 rats, some other disturbances in physiological function must contribute to 

 the deficiency symptoms and death. However, it is unlikely that the changes 

 in liver metabolism have much to do with either the symptoms or death, 

 and he actually found that the a-keto acid oxidase activities in the brain 

 are lower in pyrithiamine-treated rats than in thiamine-deprived rats. The 

 other argument, that pyrithiamine causes a polyneuritis that is difficult 

 or impossible to reverse by administration of thiamine whereas a-keto acid 

 oxidase activities can be readily restored in tissue extracts by adding thia- 

 mine-PP, may be significant, but these data could be just as easily explained 

 by an inhibition of thiamine kinase (preventing the synthesis of thiamine- 

 PP in the animal) or a very slow rate of exchange between thiamine and 

 enzyme-bound pyrithiamine-PP in the intact nervous tissue. 



ANALOGS OF RIBOFLAVIN AND FAD 



Riboflavin functions in metabolism as riboflavin-5'-phosphate (flavin mo- 

 nonucleotide, FMN) and flavin-adenine dinucleotide (FAD) in various oxi- 

 dizing enzymes and electron transport. The flavin coenzymes are usually 

 very tightly bound to their respective apoenzymes and are not dissociated 

 during extraction of the enzyme preparations. Indeed, in some cases, such 

 as succinate oxidase, the flavin component can be liberated only by pro- 

 teolytic digestion, with fragments of peptides attached, and the activity 

 cannot be restored by addition of any flavin compound. In most cases it 

 is thus difficult for analogs to replace or compete with the flavin coenzyme, 

 particularly in preparations from animal tissues, although in microorganisms 

 the flavoenzymes are generally more readily dissociable. The binding of 

 FAD seems to involve the isoalloxazine ring (perhaps the imino group at 

 position 3), possibly the ribityl portion, the phosphates, and the adenine 

 ring. Chelation to apoenzyme-bound metal ions, such as iron, is likely be- 

 cause most flavoenzymes contain such metal ions, but there is still some 

 doubt as to whether the metal ions function primarily in binding or in elec- 

 tron transfer. 



Animals and a few bacteria depend on exogenous riboflavin but it is 

 synthesized in plants and most microorganisms. The pathway of riboflavin 

 biosynthesis is not well understood and has been studied mainly in a few 

 microorganisms used for the commercial production of riboflavin; the reac- 

 tions by which riboflavin is transformed into active coenzymes are better 

 documented. An abbreviated scheme of biosynthesis and breakdown is re- 

 presented here to facilitate discussion of the actions of analogs. 



