Current Research in Toxicology (Jun 2020)
Toxicological testing of syringaresinol and enterolignans
Abstract
Lignans are secondary plant constituents with dibenzylbutane skeletons found in cereals, oilseeds, and nuts. Two members of this class, syringaresinol (Syr) and secoisolariciresinol (Seco), occur at relatively high levels in cereals and processed food products as well as in coniferous trees. In vitro studies have shown that Seco and its metabolites enterodiol (END) and enterolactone (ENL), which are formed by intestinal microbes, exhibit strong antioxidant activity because of their phenolic character. The biological activity and discussion of dietary supplementation with these substances led to questions about the potential adverse health effects of these compounds, which are explored here. Syr and the metabolites END and ENL were investigated by combining structural information generated in silico with practical testing in vitro. An in silico structure-activity analysis was performed using ToxTree and NexusPrediction to suggest plausible mechanisms of toxicity and estimate toxicological endpoints of these compounds. Structural alerts were generated based on the presence of phenolic units with coordinating substituents that could potentially form quinoid structures, promote reactive oxygen species (ROS) formation, bind to cellular structures, or damage chromosomes. To assess the in silico results, the cytotoxicity and genotoxic potential of the studied compounds were tested in vitro using the resazurin reduction and comet assays, respectively. Incubating HepG2 and HT29 cells for 1 h or 24 h with 0–100 μM Syr, END, or ENL induced no cytotoxic effects. Additionally, even the highest tested concentrations of END and ENL showed no modulation of background and total DNA damage. The initial in silico screen thus generated structural alerts linked to toxicological endpoints, but experimental assessments of the studied compounds revealed no detectable toxicity, demonstrating the need for individual mechanistic experimental verification of in silico predictions. This approach makes it possible to connect known biological activity, such as reported antioxidative effects, to underlying mechanisms such as proton abstraction or donation. This in turn can yield insights – for example, that a compound's tendency to act as a pro- or anti-oxidant (and hence to exert adverse or beneficial health effects) may depend on its concentration and the cellular state.
