Apiaceous vegetable constituents inhibit human cytochrome P-450 1A2 (hCYP1A2) activity and hCYP1A2-mediated mutagenicity of aflatoxin B₁

Abstract

In humans, apiaceous vegetables (carrots, parsnips, celery, parsley, etc.) inhibit cytochrome P-450 1A2, a biotransformation enzyme known to activate several procarcinogens, including aflatoxin B₁ (AFB). We evaluated eight phytochemicals from apiaceous vegetables for effects on human cytochrome P-450 1A2 (hCYP1A2) activity using a methoxyresorufin O-demethylase (MROD) assay and a trp-recombination assay. Saccharomyces cerevisiae was used for heterologous CYP1A2 expression and this yeast strain is also diploid and auxotrophic for tryptophan due to mutations in the trp5 alleles. When these two alleles undergo AFB-induced mitotic recombination, gene conversion occurs, allowing yeast to grow in the absence of tryptophan. The apiaceous constituents psoralen, 5-methoxypsoralen (5-MOP), 8-methoxypsoralen (8-MOP), and apigenin were potent inhibitors of hCYP1A2-mediated MROD activity in yeast microsomes, whereas quercetin was a modest hCYP1A2 inhibitor. Naringenin, caffeic acid, and chlorogenic acid did not inhibit hCYP1A2-mediated MROD activity. The 2-h pretreatment of intact yeast cells with psoralen, 5-MOP, and 8-MOP significantly improved cell survival after subsequent 4-h AFB treatment and reduced hCYP1A2-mediated mutagenicity of AFB. Apigenin also significantly decreased mutagenicity. These results suggest that in vivo CYP1A2 inhibition by apiaceous vegetables may be due to the phytochemicals present and imply that apiaceous vegetable intake may be chemopreventive by inhibiting CYP1A2-mediated carcinogen activation.

Introduction

Cytochrome P-450 1A2 (CYP1A2) mediates the formation of highly reactive, carcinogenic intermediates from a variety of procarcinogenic parent compounds including certain heterocyclic amines and polycyclic aromatic hydrocarbons (Boobis et al., 1994, Malfatti and Felton, 2001, Shimada et al., 2001). CYP1A2 also activates aflatoxin B₁ (AFB) (Gallagher et al., 1996), a metabolite of the mold Aspergillus flavus that contaminates peanuts, corn, and other commodities. AFB is a primary factor in hepatocellular carcinoma development and is one of the most potent known human hepatocarcinogens (Eaton et al., 2001, Hussein and Brasel, 2001). Although both CYP3A4 (Yamazaki et al., 1995) and CYP1A2 can activate AFB to a carcinogenic epoxide, in vitro kinetic studies suggest that CYP1A2 may be the predominant means of activation at the low concentrations of AFB that would be found in the liver after dietary exposure (Gallagher et al., 1996). The highly reactive exo-AFB-epoxide formed by CYP1A2 and CYP3A4 can covalently bind to the N-7 position of guanine in DNA and lead to somatic gene alterations (Salbe and Bjeldanes, 1989). Because AFB has to be activated in order to exert its carcinogenic effect, inhibition of CYP1A2 may reduce the risk of AFB-mediated carcinogenesis. For example, oltipraz, an antischistosomal drug approved by the US Food and Drug Administration, has been shown to protect rats from AFB-induced carcinogenesis (Kensler et al., 1987). Although oltipraz can induce phase II conjugating enzymes, it has been proposed that the protective effect is more likely due to decreased AFB activation; in vitro, oltipraz decreases CYP1A2 metabolism of AFB through competitive inhibition (K_i = 10 μM) and possibly irreversible inhibition (Langouet et al., 1995). A chemoprevention study in an area with high AFB exposure supports this concept. Wang et al. (1999) found that administration of oltipraz (500 mg/week for 1 month) resulted in a 51% decrease (P = 0.030) in median levels of urinary aflatoxin M₁, a marker metabolite of CYP1A2. Although oltipraz was recently found to significantly induce CYP1A2 mRNA expression in human hepatocytes, the induction was without concomitant increases in CYP1A2 enzyme activity (Gross-Steinmeyer et al., 2005) suggesting the inductive effect of oltipraz does not exceed its inhibitory effect on catalytic activity with regard to CYP1A2. Interestingly, oltipraz is a synthetic derivative of 1,2-dithiole-3-thione, a natural constituent in cruciferous vegetables (Kensler et al., 1987).

Dietary consumption of apiaceous vegetables (Apiaceae, which includes carrots, celery, parsnips, parsley, dill, cilantro, etc.) has been shown to inhibit CYP1A2 in humans and it was suggested that the inhibition was possibly related to the phytochemical content of apiaceous vegetables (Lampe et al., 2000). Apiaceous vegetables are rich in the furanocoumarins psoralen, 5-methoxypsoralen (5-MOP), and 8-methoxypsoralen (8-MOP) (Lombaert et al., 2001, Beier, 1990, Duke, 2004). They also contain the flavonoids (as glycosides and as aglycones) apigenin, quercetin, and naringenin and the phenolic acids caffeic acid and chlorogenic acid (Duke, 2004). Using various CYP1A2 substrates, 8-MOP, apigenin, quercetin, and naringenin have been reported to inhibit CYP1A2 in human liver microsomes (Ono et al., 1996, Zhang et al., 2001, Dierks et al., 2001, von Moltke et al., 2004, Lee et al., 1998, Pastrakuljic et al., 1997, Fuhr et al., 1993) and 8-MOP and 5-MOP were reported to inhibit caffeine metabolism in humans (Mays et al., 1987, Bendriss et al., 1996, Tantcheva-Poor et al., 2001). However, it is possible in these cases that other enzymes present competed, at least in part, for the same substrates used to probe for CYP1A2 activity. There are data showing significant CYP1A2 inhibition by chlorogenic acid and caffeic acid in liver microsomes from rats (Baer-Dubowska et al., 1998) and hamsters (Teel and Huynh, 1998). There are also data on the chemopreventive effects of quercetin, chlorogenic acid, and caffeic acid indicating reduction in AFB activation and DNA adduct formation in mouse liver microsomes (Bhattacharya and Firozi, 1988) and in strain TA98 of Salmonella typhimurium in the presence of rat liver microsomes (San and Chan, 1987). But since these results were obtained using animal-based systems, species differences with regard to cytochrome P-450 activity (Bogaards et al., 2000) may limit the applicability of the data to humans.

Our objective was to determine if psoralen, 5-MOP, 8-MOP, apigenin, quercetin, naringenin, caffeic acid, and chlorogenic acid inhibit human CYP1A2 (hCYP1A2) in a recombinant yeast strain expressing hCYP1A2 and whether these components inhibit the mutagenicity of AFB (see Fig. 1 for chemical structures of phytochemicals and AFB). Because this system is devoid of other enzymes that compete for the hCYP1A2 substrate probes used in our investigation, it provides a useful approach for evaluating effects on hCYP1A2 directly. First, we used microsomes from yeast expressing hCYP1A2 to screen the apiaceous constituents for the most potent hCYP1A2 inhibitory effects and to determine mechanism of inhibition. We then used intact yeast expressing hCYP1A2 to test if the screened compounds inhibited AFB-induced mutagenicity as measured by trp5 recombination in living cells.