514 Editor: M. WINITZ 
zymatic processes of oxidation of the D-isomer to the a-keto acid and transamination of the latter 
to the L-isomer. Well, we became interested in this problem of the metabolic breakdown of p- 
amino acids, not in the test tube but the whole animal, and wondered whether we could, by 
adding sufficiently large quantities of non-essential D-amino acids to the diet, so overload the 
mechanism of the animal for converting these D-amino acids to their L-isomers that it would now 
be unable to utilize the D-isomer of an essential amino acid that was ordinarily utilizable for 
purposes of growth. This we were able to demonstrate. I won’t go into this in detail, but we did 
show that we could actually pinpoint the enzymes responsible for the in-vivo conversion of a 
D-amino acid to an L-amino acid and that D-amino acid oxidase was the enzyme involved in 
the first step of this process. The important point here is that we did find that we could inhibit 
this D-amino acid oxidase activity in the whole animal to any desired extent merely by adding 
to the diet, sodium benzoate which was shown by earlier workers to be a very potent inhibitor of 
p-amino acid oxidase im vitvo. Diets containing D-amino acids that were ordinarily utilizable 
became increasingly less so the greater the benzoate concentration. We also found that when 
benzoate was added to the diet, not only was the ability of the animal to convert a D-amino acid 
to the utilizable L-isomer impaired, but that most of the D-amino acid ultimately appeared in the 
urine in very large amounts. It is of course well known that D-amino acids are generally more 
rapidly excreted into the urine than the corresponding L-amino acids, which are reabsorbed by 
the kidney tubules, recycled and more efficiently utilized. 
Returning now to the problem of the identification of D-amino acid residues in proteins—how 
can this system be employed here? Well, if the protein which is suspected of containing D-residues 
were fed to an animal—and I must again state that this is a speculative method that hasn’t yet 
been tried experimentally—ain a diet which contained a concentration of sodium benzoate sufficient 
to inhibit effectively the D-amino acid oxidase activity of that animal, then the protein would 
be hydrolyzed in the usual manner by the proteases of the gastrointestinal tract. The resulting 
enzymic digest, which might also include small resistant peptides containing D-residues, would 
then be absorbed through the gut wall into the circulatory system. These smaller peptides could 
ultimately come into contact with the aminopeptidase of the kidney where they would undergo 
hydrolysis to their component amino acids. As the D-amino acid oxidase is inhibited by the ben- 
zoate, any D-amino acids that might be present would ultimately appear in the urine and could 
subsequently be estimated 7m vitvo using some suitable technique. 
H. ROSENBERG: I am extremely grateful for this. This seems to be an excellent method, and we 
are certainly going to try it out. I may just add that perhaps a very good control to try the 
method out would probably be to put an animal on benzoate and then give it a diet of casein 
hydrolyzate plus one part in a thousand of, say, D-alanine and see whether you can pick it up 
from the urine. 
The question I wanted to ask concerns the inhibitor. There have been a number of compounds 
similar to benzoate (salicylates, etc.) which are all known to be FAD antagonists. I suppose 
they will be active in this procedure. The second thing is, what levels of benzoate do you give 
to produce inhibition. Do you know offhand? 
Winitz: The levels of benzoate that are effective in our diets—and we feed our chemically 
defined diets as crystal clear solutions in water—are 2% of a diet solution. A too-g animal will 
generally eat about 25 to 30 ml of 50% solution of diet daily, which is equivalent to about 12!/, to 
15 g of a solid food diet. 
Your first question, with regard to salicylate and various other things, touches on a very 
personal experience that I had some years ago when my older son, who was 3 years old at the 
time, somehow managed to get into the medicine cabinet, which my wife and I had thought was 
safely beyond his reach, appropriated the candy-flavored baby aspirin therefrom, and swallowed 
some 50 of those wonderfully tasting aspirin tablets. When we learned what had happened—which 
fortunately was very soon after the incident had occurred—we rushed the youngster to the hos- 
pital where he immediately had his stomach pumped in order to remove the unabsorbed aspirin. 
Well, the youngster came through the experience O.K., but it seemed incredible to me at the 
time that there really was no adequate means for counteracting salicylate toxicity; there still 
isn’t. Yet the aspirin sales in this country are in the vicinity of $160,000,000 a year, and that’s a 
lot of aspirin. 
Well, what does this have to do with Dr. RoSENBERG’s question? We looked into this benzoate 
phenomenon recently and found that when benzoate is added to a chemically defined diet which 
contains one of the essential amino acids as the D-isomer, and which will ordinarily support 
growth, then growth is markedly retarded due to the fact that the benzoate, by inhibiting the 
D-amino acid oxidase activity, effectively prevents adequate conversion of the D-isomer to its 
essential L-form. We further found that this inhibitory effect of benzoate on growth of the animal 
could be counteracted by adding sufficient glycine to the diet. Now it is well known that benzoate 
is detoxified in the whole animal by its conversion to hippuric acid, which is excreted in the 
urine as such, and that hippuric acid is a combination of benzoic acid and glycine in amide linkage. 
References p. 524 
