651 



N. E. Tolbert 



fixation product and a specific phosphatase for its hydrolysis 

 exists, it seems probable that phosphoglycolate is the immediate 

 precursor of glycolate. Past theories for glycolate biosynthesis 

 should be reconsidered to include the phosphate ester. 



Ci plus d condensation : Synthesis directly from CO2 would 

 produce uniformly labeled glycolate as is actually observed. 

 Proposals for such a pathway have not yet been supported by con- 

 vincing data (17, 18). 



Carbon cycle : It has long been postulated by many of us that 

 glycolate arose from a sugar phosphate of the photosynthe tic car- 

 bon cycle, but space limitation does not permit an extensive re- 

 view of this literature. Bassham, Benson, Calvin, et al_. (19) 

 demonstrated that carbons 1 and 2 of RuDP , hexose phosphate and 

 sedoheptulose phosphate were uniformly labeled and could thus 

 give rise to a uniformly labeled C2 moiety. Ribose-l-C^"* added 

 in vivo and ribose-5-phosphate-l-C-'-'* added in vitro gave rise to 

 glycine-2-C^'* (20, 21, 22). Such results are consistent with 

 cleavage between C2 and C3 of a keto sugar phosphate so that car- 

 bon 2 of the sugar became the carboxyl of glycolate. These re- 

 actions can be modeled after TPP linked transketolase , except 

 that a free C2 piece would be liberated rather than transferred. 

 Several points about this theory await clarification. The yields 

 of glycine from ribose-5-phosphate have been small, and no one 

 has detected the cleavage enzyme. Small yields with crude en- 

 zymes are suspect, particularly since RuDP nonenzymatically de- 

 composes into a multitude of products including C2 pieces. In 

 Bradbeer and Racker's abstract (23) on the production of free 

 glycolate from f ructose-6-phosphate in the light by chloroplasts , 

 the yields were also very small. Since phosphoglycolate appears 

 to be the precursor of glycolate, the sugar should be a diphos- 

 phate, which is not known to be cleaved by transketolase. Thus 

 phosphoglycolate formation might arise from a cleavage of xylu- 

 lose, fructose or sedoheptulose diphosphates. RuDP should not 

 be a precursor, because it does not have the necessary transcon- 

 figuration of the hydroxyl groups between carbons 3 and 4 as is 

 generally required by aldolases, transaldolases and transketo- 

 lases. Phosphoglycolaldehyde might be a precursor for phospho- 

 glycolate. Although aldolase has been shown to catalyze the 

 formation of xylulose diphosphate from glycolaldehyde phosphate 

 plus dihydroxyacetone phosphate (24), the reverse of this reaction 

 would produce the C2 -phosphate from carbons 4 and 5 of the pen- 

 tose. During photosynthesis these two carbons are also equally 

 labeled with C^^^ 



Recently we have examined two facets of this problem. A 



