200 FLAVONOIDS AND RELATED COMPOUNDS 



substances covered in this chapter such as xanthones, stilbenes, etc., although from their 

 structures they probably have similar precursors. Also unknown are the more specific 

 pathways by which one type of flavonoid is formed rather than another. Speculation on 

 these matters is presented in the reviews of Geissman and Hinreiner (78), Bogorad (79), 

 and Grisebach and Ollis (80). The diagram here attempts to present what seem to be likely 

 pathways. 



Tracer experiments by several workers have established that the B ring of flavonoids 

 comes from shikimic acid: 



COOH 



The A ring is formed by head-to tail grouping of three acetate molecules. The aliphatic 

 three-carbon chain is probably added to ring B before ring A is formed to produce a Cg - 

 C3 compound. This broad picture of flavonoid biosynthesis has been shown to hold for 

 quercetin (81, 82), cyanidin (83), phlorizin (84), and catechins (85). It presumably applies 

 also to other flavonoids. More generally, it is presumed that all aromatic rings having 

 ortho hydroxyl groups arise from shikimic acid and all aromatic rings with lueta hydroxyl 

 groups arise from acetate. The tracer experiments have ruled out formation of ring A 

 from inositol or phloroglucinol as has sometimes been suggested. They have also shown 

 that such Cg-Cs compounds as phenylalanine, cinnamic acid and ferulic acid (85) are ef- 

 ficient precursors of the C6(B)-C3 portion of flavonoids. The accumulation of p-coumaric 

 acid esters in flower buds oi Antirrlilnuni niajus and their disappearance as the colored 

 flowers develop also points to p-coumaric acid as a pigment precursor (87). 



It is frequently observed that in a given species all of the different flavonoids have 

 the same ring hydroxy lation pattern, differing in methylation, glycosylation and the struc- 

 ture of the C3 portion (88). Such an observation suggests that there is a common C15 in- 

 termediate that is converted to the different flavonoids after the ring hydroxylation pattern 

 has been established. In fact there is a good likelihood that the hydroxylation of ring B 

 is established in a C6-C3 intermediate before ring A is added to it. There is controversy 

 as to whether the different classes of flavonoids are formed by completely divergent path- 

 ways from a single precursor or whether interconversion of the flavonoid classes can 

 occur without mediation of a common precursor. The question has been raised particu- 

 larly with regard to the leucoanthocyanidins and anthocyanidins. It seems quite possible 

 that some plants carry out this conversion whereas in other plants the two classes arise 

 as end products of parallel pathways (89,90,91). Seshadri (92) has argued for the key im- 

 portance of flavanonols in the biosynthesis of other flavonoids. Grisebach and Patschke 

 (93, 94) have shown that chalcones are converted to flavanones, aurones, anthocyanidins, 

 flavonols, and isoflavones. 



It is widely accepted that isoflavones are produced from flavones by migration of the 

 phenyl group from C-2 to C-3. This view has been supported by the experimental work 

 of Grisebach and Doerr (95) who fed carboxyl-labelled phenylalanine to Trifolium pratense 

 and isolated an isoflavone with label at C-4. On the other hand Seshadri (92) has suggested 

 that a branched C6-C3 compound may undergo initial condensation with ring A. Possibly 

 different plants use different biosynthetic routes. 



Nothing has been said to indicate the point where methylation or glycosylation occurs. 

 It is usually assumed that these steps are the last in the sequence, and the genetic experi- 



