Abstract
Liquid smoke products are widely used as a food additive to create a desired smoke flavour. These products may contain hazardous chemicals generated during the wood-burning process. However, the toxic effects of these types of hazardous chemicals constituting in the commercially available products are largely unknown. Therefore, a test battery of cell-based in vitro methods, covering different modes of actions of high relevance to human health, was applied to study liquid smoke products. Ten liquid smoke flavourings were tested as non-extracted and extracted. To assess the potential drivers of toxicity, we used two different solvents. The battery of in vitro methods covered estrogenicity, androgenicity, oxidative stress, aryl hydrocarbon receptor activity and genotoxicity. The non-extracted samples were tested at concentrations 0.002 to 1 μL liquid smoke flavouring/mL culture medium, while extracted samples were tested from 0.003 to 200 μL/mL. Genotoxicity was observed for nearly all non-extracted and all hexane-extracted samples, in which the former had higher potency. No genotoxicity was observed for ethyl acetate-extracted samples. Oxidative stress was activated by almost all extracted and non-extracted samples, while approximately half of the samples had aryl hydrocarbon receptor and estrogen receptor activities. This study used effect-based methods to evaluate the complex mixtures of liquid smoke flavourings. The increased bioactivities seen upon extractions indicate that non-polar chemicals are driving the genotoxicity, while polar substances are increasing oxidative stress and cytotoxic responses. The differences in responses indicate that non-extracted products contain chemicals that are able to antagonize toxic effects, and upon extraction, the protective substances are lost.