Genotoxic activities of the food contaminant 5-hydroxymethylfurfural using different in vitro bioassays

Abstract

5-Hydroxymethylfurfural (5-HMF) is known as an indicator of quality deterioration in a wide range of foods. 5-HMF is formed as an intermediate in the Maillard reaction and has been identified in a wide variety of heat-processed foods. In recent years, the presence of 5-HMF in foods has raised toxicological concerns: data have shown cytotoxic, genotoxic and tumoral effects but further studies suggest that 5-HMF does not pose a serious health risk. However the subject is still a matter of debate. We investigated the genotoxicity of the food-borne contaminant 5-HMF using the Ames test, the micronucleus (MN) and the single-cell gel electrophoresis (SCGE) assays in the human metabolically active HepG2 cell line. Cytotoxic effect of 5-HMF was first assessed using Alamar Blue™ as a sensitive sub-lethal assay. 5-HMF did not induce any genic mutation in bacteria whatever the concentration in the Ames test. Furthermore, it does not induce clastogenic or aneugenic effects in the HepG2 cells. In contrast, 5-HMF induced HepG2 DNA damage at concentrations from 7.87 to 25 mM in the comet assay suggesting a weak genotoxic effect of 5-HMF in the HepG2 cells probably repaired.

Introduction

5-Hydroxymethylfurfural (5-HMF) (Fig. 1), a common major product of the Maillard reaction, is formed in many foods from reaction of reducing sugars and amino acids (Antal and Mok, 1990, Cuzzoni et al., 1988). Janzowski et al. (2000) reviewed the different available data regarding this compound. In an analytical screening of nearly 500 food samples, relatively high 5-HMF concentrations (exceeding 1 g/kg) have been reported for specific products such as dried fruits, caramel products, and juice made from dried plums. Very high peak concentrations up to 3.5 g/kg were found in dried pears, up to 9.5 g/kg in caramel products (Bachmann et al., 1997), up to 6.2 g/kg in instant coffee powder and up to 13.9 g/kg in respective substitutes (Schultheiss et al., 1999). Formation of 5-HMF in foods varies with processing and storage conditions and is especially dependent on temperature and pH. This has been confirmed in model experiments with fruit juices and concentrates as well as in heat-treated milk (Gomis et al., 1991, Kern, 1964, Morales et al., 1992). 5-HMF has also been identified in chewing tobacco, cigarette smoke and wood smoke (Baldwin et al., 1994, Black, 1966, Chou and Hee, 1994). A contribution of 5-HMF to flavour soy sauce has even been described (Kiahara, 1995).

It is not clear whether food-borne exposure to 5-HMF represents a potential health risk for humans. 5-HMF at high concentrations is cytotoxic, causing irritation to eyes, upper respiratory tract, skin and mucous membranes. An oral LD₅₀ of 3.1 g/kg body weight has been determined in rats (Ulbricht et al., 1984). Neither data from epidemiologic studies or case reports on potential association of 5-HMF with cancer risk in humans nor chronic carcinogenicity studies are available. However, certain indications of 5-HMF for tumorigenic activities have been mentioned in rats and mice: 5-HMF can act as an initiator and also as a promotor, as shown for the induction of colonic aberrant crypt foci (Zhang et al., 1993). Induction of skin papillomas has been described by Surh and Tannenbaum (1994) after topical application of 10–25 μmol 5-HMF to mice. In contrast, in another study, the increase of skin tumour rates associated with 5-HMF-treatment was not statistically significant (Miyakawa et al., 1991).

Controversial results have been published on mutagenicity/genotoxicity of 5-HMF in vitro, particularly when tested with the Ames test. This test has often been performed on 2 strains only (TA 98–TA 100) instead of the 5 strains as recommended by the OECD guideline no. 471. The results varied also whether a metabolic activating system was used or not (Nishi et al., 1989). Furthermore, several authors (Surh and Tannenbaum, 1994, Lee et al., 1995, Sommer et al., 2003) have shown that the sulfotransferase system is required for detection of 5-HMF mutagenicity. These studies were performed under different conditions, then, a final conclusion concerning genotoxic/mutagenic activity of 5-HMF is still difficult. Considering the wide distribution and sometimes high concentrations of 5-HMF in certain foods, and also in fructose-containing solutions for intravenous injection, additional informations based on detailed dose–response experiments are required.

In this study, the Ames test was first performed according to the OECD guideline no. 471 to shed light on the mutagenicity of this food-borne compound using a prokaryotic model. To check the involvement of sulfotransferase in the toxic effect of 5-HMF, we chose a human cell line; HepG2; as eukaryotic model. Indeed, human cells expressing human xenobiotic metabolising enzymes (phases I, II and III) are more likely to be predictive of human susceptibility to the biological effects of chemicals, as RNA transcripts of cytochrome P450 (CYPs) and phase II enzymes, specially sulfotransferase system are present in HepG2 cells (Darroudi and Natarajan, 1993, Westerink and Schoonen, 2007a, Westerink and Schoonen, 2007b), this cell line is a reliable tool for toxicity assays. Indeed, the levels of CYP 1A1, 1A2, 2B6, 3A4 and SULT 1A1, 1A2, 1E1, 2A2 and 2A1 were mostly similar to levels found in primary human hepatocytes and this HepG2 cell line can be used to detect compounds giving false negative results in other in vitro assays (Knasmuller et al., 1998, Uhl et al., 2000, Lu et al., 2002).

Two genotoxic assays were performed on HepG2 cells: the micronucleus (MN) and the single-cell gel electrophoresis (SCGE) assays (Valentin-Severin et al., 2003) to study the mechanisms by which 5-HMF acts. The SCGE assay, also known as comet assay (McKelvey-Martin et al., 1993) was chosen to screen whether 5-HMF could affect HepG2 DNA. The MN assay was chosen as an indicator of both clastogenic and aneugenic effects (Kirsch-Volders et al., 2003), as this test is currently being incorporated into OECD guidelines for in vitro genotoxicity assays (OECD draft, Dec. 2007). In order to avoid false positive data in the genotoxicity assays, the Alamar Blue™ assay was selected as the complementary cytotoxicity test. It is a sub-lethal assay which is very sensitive and is useful for assessing cytotoxicity of agents within various chemical classes (Slaughter et al., 1999, O’Brien et al., 2000, Hamid et al., 2004).