VII. ESTIMATION 367 



others^'- • *' have described methods in recent years. In general the procedure 

 inchides an acid extraction, preHminary CHCI3 extraction, photolysis in 

 alkaline solution (pH 13 to 14 or 0.5 N NaOH), reacidifioation to pH 4.5, 

 and extraction of the lumiflavin into CHCI3 . Illumination with a 200-watt 

 bulb appears as satisfactory as ultraviolet light, and destruction does not 

 occur with over-illumination.^ Permanganate oxidation after photolysis'* 

 may be desira])le. There is disagreement as to whether one simple extraction 

 is sufficient*-' ** or whether continuous extraction must be used.^ 



Although some of the technical operations appear cumbersome for rou- 

 tine use, this is probably only a reflection of the lack of effort to improve 

 and simplify the procedures. More intensive study and comparison with 

 other methods is needed. 



c. Microfiuorometric Methods 



Burch et al} developed methods for serum, red blood cells, and whole 

 blood which require only 25 cubic mm. of material. Riboflavin in the white 

 cells may be obtained with only 0.1 ml. of blood. The amounts of free ribo- 

 flavin, riboflavin phosphate, and flavin adenine dinucleotide (FAD) may 

 also be determined. Under the conditions used, the relative fluorescence 

 of FAD is only 14 % of riboflavin. By determinations before and after hy- 

 drolysis (5 % trichloroacetic acid at 38° for 20 hours) the amount of FAD is 

 obtained. Riboflavin mononucleotide is practically absent from serum and 

 may be ignored. Hydrosulfite reduction is used to obtain blank readings. 



This method has proved extremely useful, and as yet no serious criticisms 

 or improvements have been recorded. The small volumes required make 

 it especially useful for field surveys in nutrition. Serum riboflavin values 

 have been shown to be related to riboflavin intake in experimental animals,** 

 especially the FAD content, which fluctuates less and falls later than the 

 free riboflavin level. Whole blood levels are less useful,**- *^ since the ribo- 

 flavin content is greatlj^ influenced by the number of red cells. The diagnos- 

 tic value in human beings remains to be proved but appears promising. 



d. Estimation of Riboflavin Nucleotides 



Bessey et aZ.*^ developed a method for the fluorometric determination of 

 riboflavin, riboflavin mononucleotide, and flavin adenine dinucleotide in 

 tissues. The differential fluorescence of FAD as compared to riboflavin and 



32 A. Fujita and K. Matsuura, J. Biochem. (Japan) 37, 445 (1950). 



" K. Yagi, Igaku to Scibutfiugaku 18, 264 (1951) [C./l. 45, 9591 (1951)]. 



^* K. Suvarnakich, Tlie Measurement of Riboflavin Deficiency, Thesis. Harvard 



School of Public Health, 1951. 

 " A. E. Axelrod, T. D. Spies, and C. A. Elvehjcm, Proc. Soc. Exptl. Biol. Med. 46, 



146 (1941). 

 '« O. A. Bessey, O. H. Lovvry, and R. 11. Love, J. Biol. Chem. 180, 755 (1949). 



