214 IV. CONVERSION OF FAT TO CARBOHYDRATE 



as the test animals, and Liithje,^" who employed panci'eatectomized dogs, 

 were able to account for only approximately 40% of the ingested glycerol 

 as "extra sugar" in the urine. Chambers and DeueP^ reported that a com- 

 plete conversion of glycerol to D-glucose obtains in the phlorhizinized dogs. 

 The incomplete recovery reported earlier may have resulted from an in- 

 complete absorption of the alcohol. 



2. The Transformation of Fatty Acids with Odd-Numbered Carbons 



to Carbohydrate 



It has long been recognized that propionic acid is normally converted to 

 carbohydrate. Ringer^- reported that all of the propionic acid molecule 

 (100%) was converted to glucose by the phlorhizinized dog. Further proof 

 of the convertibility of propionic acid to carbohydrate is afforded by the re- 

 peated demonstration of glycogenesis after the oral administration of pro- 

 pionate.^^ Moreover, additional confirmation of the glycogenic action of 

 propionate is the demonstration that tripropionin gives rise to more gly- 

 cogen when fed to rats than obtains after equivalent amounts of triacetin or 

 tributyrin are given. ^^ As indicated in the last section, this is to be attrib- 

 uted to the fact that both the glycerol and the fatty acid moieties of the 

 triglycerides serve as a source of the glycogen in the case of tripropionin, 

 while only the glycerol behaves in this way in the case of the neutral fats 

 composed of short-chain acids having an even number of carbons. 



The longer-chain fatty acids having an uneven number of carbons are 

 likewise convertible to glycogen; it seems most likely that this change oc- 

 curs to the extent that such odd-chain acids are converted to propionic acid. 

 Ringer^'* reported that the five-carbon fatty acid, valeric acid, is converted 

 to glucose to the extent of three carbons by the phlorhizinized dog. The 

 convertibility of valeric (C5), heptanoic (C7), and pelargonic (C9) acids to 

 glycogen was proved when they were fed to rats as their sodium salts- ^ 

 or as their ethyl esters. ^^ The ethyl ester of undecylic (dO acid has like- 

 wise been showTi to yield glycogen in the rat.^^ Finally, it has also been re- 

 ported that the level of liver glycogen deposition is higher for trivalerin or 

 triheptylin than for the triglycerides of the corresponding even-chain fatty 

 acids in which only the glycerol moiety is available for glycogenesis. The 



30 H. Luthje, Deut. Arch. klin. Med., 80, 98-104 (1904). 



31 W. H. Chambers and H. J. Deuel, Jr., J. Biol. Chen., 65, 21-29 (1925). 



32 A. I. Ringer, J. Biol. Chem., 12, 511-515 (1912). 



33 H. C. Eckstein, J. Biol. Chem., 102, 591-594 (1933). 

 3^ A. I. Ringer, /. Biol. Chem., I4, 43-52 (1913). 



35 J. S. Butts, H. Blunden, W. Goodwin, and H. J. Deuel, Jr., ./. Biol. Chem., 117, 

 131-133(1937). 



