Too much of a good thing?

To the editor

The boosted antioxidant activity of transgenic tomatoes containing up to 78-fold enhanced levels of the peel flavonol rutin, reported in the May issue (Nat. Biotechnol. 19 , 470–474, 2001 ), was touted to confer health promoting effects against coronary heart disease and certain forms of cancer. However, as the celebrated physician Paracelsus noted: “All substances are poisons...The right dose differentiates a poison from a remedy.” In this context, while flavonols and related polyphenols can be beneficial, they can also have potentially harmful effects.

For example, for several years quercetin, the aglycone of rutin found in red wine, has been recognized as a mutagen. Indeed, studies with mice have indicated that quercetin may have tumorigenic activity¹. The protective beneficial properties of flavonols and other dietary polyphenols are believed to result from their ability to serve as antioxidants. By donating electrons to free radicals, which might otherwise induce biomolecular damage, polyphenols are converted to phenoxyl radicals (Ph(OH)_nO^˙), which are poorly reactive. However, the very properties that confer polyphenols with the ability to quench free radicals are also responsible for the ease with which they undergo iron- and copper-catalyzed autooxidation, resulting in formation of mutagenic hydroxyl radicals (^˙OH):

Model studies have shown that simple combinations of quercetin and copper ions, which may occur naturally in chromosomes, can induce hydroxyl-radical-mediated oxidative damage to DNA, thereby providing a mechanism for mutation².

It should be no surprise that molecules possessing antioxidant properties can also behave as pro-oxidants; a prerequisite of both phenomena is the ability to undergo facile one-electron oxidation. Similar reactions are known for ascorbic acid, glutathione, and NAD(P)H. The fine line that divides anti- and pro-oxidant behavior is exemplified by trans-resveratrol, a polyphenolic believed to be responsible for some of the health benefits of red wine. This compound can damage DNA by the above mechanisms. However in the presence of the cellular thiol glutathione, which suppresses ^˙OH generation through its stabilization of Cu⁺ ions, the polyphenol displays antioxidant behavior. This involves its sparing of glutathione from Cu²⁺-mediated oxidation³.

A potentially important metal-ion independent mechanism through which dietary polyphenols may exert toxicity involves their initial oxidation to phenoxyl radicals by peroxidases (e.g., myeloperoxidase and lactoperoxidase). Generally speaking, such radicals are not sufficiently reactive to propagate lipid peroxidation and DNA damage, but they can oxidize glutathione to its thiyl radical (GS^˙). The glutathionyl radical then combines with the parent thiolate anion (GS⁻), forming the glutathione disulfide radical anion (GSSG^˙−), which is an extremely powerful one-electron reductant, activating molecular oxygen to potentially harmful species (O₂^˙− and H₂O₂).

Indeed, several flavonols have been shown to activate large amounts of oxygen by this mechanism in the presence of only catalytic amounts of a peroxidase. Because the flavonol also acts as a catalyst (its phenoxyl radical is repaired by glutathione), relatively small quantities have the potential to induce severe toxicity⁴.

Not wishing to diminish the importance of novel dietary antioxidants, or the potential benefits that may be derived from a diet artificially enriched in flavonols, it is hoped the preceding comments highlight the need for studies into the fundamental, free-radical chemistry of novel antioxidants to keep pace with progress in food technology.