STRUCTTTRE OF THE VITAMINS A 695 



responsible for thf^ positive iodoform test. The principal argument 

 against Formula II is the fact that lycopene, unlike (8-carotene, fails to 

 function as a provitamin A.^* No active provitamin A is known which 

 does not contain one intact unsubstituted j8-ionone ring. The evidence 

 would appear to l)e pre))()nd(Mantly opposed to the Kai'rei' open-chain struc- 

 ture (II). 



The third formula i)r()pos<Ml l)y Morton and collaborators'^^ is based 

 upon the claim that the "('•..„ aldehyde""- oljtained when vitamin Ao is 

 oxidized with aluminum /erZ-but oxide in benzene using diethyl ketone as a 

 hydrogen acceptor (the so-called Oppenauer reaction) is, in fact, an alde- 

 hyde of vitamin Aj. Moi'eover. the aldehyde corresponding to Formula II 

 is very different from retinene-i, which is closely related to vitamin A2 alco- 

 hol. Finally, Morton and Creed"-^ have shown that /3-carotene serves as a 

 provitamin A for both vitamin Ao and vitamin Ai. This would seem to 

 preclude the open-chain formula or the C'22 formula and to support a C20 

 structure. However, one bit of evidence speaks against the Morton 

 formula. On ozonolysis, no a.a'-dimethylsuccinic acid could be demon- 

 strated. Such a degradation product would surely nave been expected if a 

 double bond had existed between carbons 3 and 4. In spite of this dis- 

 crepancy. Formula III would appear to be the most probable one of the 

 three proposed. 



Although the conflicting data leave the question as to the structure of 

 vitamin Ao still undecided, a definite answer should soon be forthcoming, 

 on the basis of the pure crystalline product recently isolated by Shantz.^*^ 



Gillam and his associates'^ consider that vitamin A2 has biological activity 

 for rats, while Karrer, Geiger, and Bretscher,'^^ on the other hand, claim 

 that it possesses no such activity. However, more recently Jensen, 

 Shantz, Embree, Cawley, and Harris, '^^ using a liver oil concentrate ob- 

 tained from the northern pike (Esox lucius), which contained only impure 

 vitamin X> and was uncontaminated with vitamin Ai, concluded that 

 vitamin Ao possesses a minimum biopotency of 47,500 U.S.P. units per 

 gram. Shantz and Brinkman''*^ have revised this figure still further to 

 1 ,300,000 units per gram, using as a comparison the vitamin A acetate ref- 

 erence oil. This would indicate that vitamin Ao is only about 40% as ef- 

 fective as crystalline vitamin Ai, which has a biopotency originally reported 

 to be 4,300,000 U.S.P. units per gram,-*-' but more recently revised to a value 

 of 3,333,000 I'.S.P. units per gram on the basis of the new U.S.P. standard. 



'^2 E. Hawarth, I. M. HeiUmjii, AY. K. Jone.s, A. L. Morrison, and J. B. Polva, J. 

 Chem. Soc, 19S.9, 128-132. 



>" R. A. Morton, and R. H. Creed, Biochem. J., 33, 318-324 (1939). 



^** E. M. Shantz, Science, 108, 417-419 (1948). 



'« J. L. Jensen, E. M. Shantz, N. D. Embree. J. D. Cawlev, and P. L. Harris, ./. Biol. 

 Chem., 149, 473-477 (1943). 



'^* E. M. Shantz, and J. H. Brinkman, Report of 114th Meeting of the American 

 Chemical Society, Washington, D. C, Aug. 30 to Sept. 3, 1948, Abstract No. 24, p. 17c; 

 J. Biol Chem., 183, 467-471 (1950). 



