VIII. STANDARDIZATION 373 



gives figures from whicii the average riboflavin content of the sample is 

 reuchly calculated. 



The extraction of riboflavin offers little tlifliculty because of th6 looseness 

 with which it is bound, its stability to heat and acid, and the fact that the 

 known bound forms, riboflavin phosphate and flavin adenine dinucleotide, 

 have the same activity as riboflavin for the test organism. Autoclaving the 

 finely divided sample with a tenfold excess of 0.1 N llCl at 15 lb. pressure 

 for 15 to 20 minutes has proved very effective. The cooled mixture is di- 

 luted and filtered through paper to remove traces of fatty acids, which 

 interfere with the response to riboflavin. Properly diluted aliquots of the 

 filtrate, adjusted to pH 6.8 to 7.0, are used for assay. All operations, during 

 both extraction and assay, should be carried out in darkness or subdued 

 light, since riboflavin is readily destroyed by Ught. 



Accuracy of the procedure is attested by agreement of the results obtained 

 for a great variety of materials with those obtained by chemical and rat 

 growth procedures. ^^' ^^ A micro modification of the procedure that permits 

 determination of riboflavin in amounts from to 0.004 7 has been de- 

 scribed.^^- *^ 



]\Iost other microbiological procedures for riboflavin also utihze L. casei 

 as the test organism in a slightly modified medium. ^^ An exception is the 

 procedure of Kornberg et al.,^^- ""^ which employs a strain of Leuconostoc 

 mesenteroides said to be some fifty times more sensitive to riboflavin than 

 is L. casei. These methods may prove useful for special applications. They 

 have not, however, been tested nearly as widely as the Snell-Strong proce- 

 dure recommended above. 



VIII. Standardization 



M. K. HORWITT 



With the elucidation of the chemical nature of riboflavin, the need for 

 a standard of activity for vitamin Bo or G became less urgent. The present 

 I^.S.P. Reference Standard is a recrystallized sample of riboflavin obtainable 

 from the U.S. P. Reference Standards Committee. Comparisons of purified 

 riboflavin with the older units of activity showed that one Bourquin-Sher- 

 man rat growth unit,^ the daily addition of which will produce an average 

 gain of 3 g. per rat per week, was equal to about 2.5 7 of riboflavin.-' ^ 



69 O. H. Lowry and O. A. Bessey, J. Biol. Chem. 155, 71 (1944). 



^» H. A. Kornberg, R. S. Langdon, and V. H. Cheldelin, Anal. Chem. 20, 81 (1948). 



1 A. Bourquin and H. C. Sherman, J. Am. Chem. Soc. 53, .3501 (1931). 



* O. A. Bessey, /. Nutrition 15, 11 (1938). 



