Sterigmatocystin in foodstuffs and feed: aspects to consider

ABSTRACT Sterigmatocystin (STC) is a possible human carcinogen (2B) according to International Agency for Research on Cancer classification and has been associated with immunotoxic and immunomodulatory activity, together with mutagenic effects. It might be found in numerous substrates, from foods and feeds to chronically damp building materials and indoor dust. Although European Food Safety Authority concluded that the exposure to STC to be of low concern for public health, reinforces the need of data concerning exposure of European citizens. Climate change can represent an increased risk of exposure to STC since it is a crucial factor for agro-ecosystem powering fungal colonisation and mycotoxin production This aspect can represent an increased risk for European countries with temperate climates and it was already reported by the scientific community.


Background
Food and feed can be contaminated with a mixture of mycotoxins, toxic secondary metabolites produced by fungal species. Fungal invasion and consequent mycotoxin production occurs both in the field and/or during the storage of crops (de Nijs et al. 2016). The level of contamination is dependent on many factors such as fungal interactions, type of crops and environmental factors (Battilani et al. 2012; Van der Fels-Klerx 2016). In food, other aspects can influence the profile and levels of the mycotoxins present, namely, industrial processes involved and, after being acquired, household preparation process (Nijs et al. 2016). In addition, we should consider that mycotoxins are extremely difficult to eliminate from food even after the cooking process because they are quite stable molecules. All this explains why mycotoxins can be present in the food or environment long after death and disintegration of the toxic fungus (Peraica et al. 1999;Halstensen 2008;Alborch et al. 2011;Viegas et al. 2015).

Aspergillus section Versicolores
The identification of aspergilli down to the species level was traditionally based on the morphological features (Raper and Fennel 1965). More recently, Houbraken et al. (2014) and Hubka et al. (2014) proposed the system comprising four subgenera (Aspergillus, Circumdati, Fumigati, Nidulantes) with 20 sections based on phylogenetic approach. The official fungal DNA barcode applying ITS locus (Schoch et al. 2012) has seemed to be insufficient for accurate identification of aspergilli and their sexual morphs, so, additional marker analyses were developed: calmodulin gene(CaM), beta-tubuline (BenA) and RNA polymerase II second largest subunit (RPB2) sequencing, along with extrolite spectrum data as well Frisvad 2015). Nowadays, 17 species are assigned to the Aspergillus section Versicolores:: A. amoenus, A. austroafricanus, A. creber, A. cvjetkovicii, A. fructus, A. griseoaurantiacus, A. hongkongensis, A. jensenii, A. pepii, A. protuberus, A. puulaauensis, A. subversicolor, A. sydowii, A. tabacinus, A. tennesseensis, A. venenatus and A. versicolor (Jakšić Despot et al. 2017). According to the works by Chen et al. (2016) and Hubka et al. (2016), the section Versicolores was merged to the section Nidulantes as the A. versicolor clade, to maintain monophyly of Aspergillus. The newest combined phylogeny of aspergilli at the section Versicolores based on GenBank sequences was published by Jakšić Despot et al. (2017).
Along to the typical aspergilluma biseriate head on a long stipe, several A. versicolor and A. sydowii isolates produce diminutive conidial heads resembling penicillate conidiophores. The recommended sexual name Emericella should be used for species in the section Versicolores. Aspergillus versicolor (Vuil.) Tirab. was historically the most commonly reported representative of the entire section. Its herbarium strain CBS 538.65 bears ITS barcode EF652442, BenA = EF652266, CaM = EF652354, RPB2 = EF652178, followed by A. sydowii (Bainier and Sartory) Thom and Church, herb.: IMI 211,384, ITS barcode: EF652450, BenA = EF652302, CaM = EF652390, RPB2 = EF652214 . So, that is why A. versicolor strains were reported growing worldwide, prospering in many habitats, including foodstuffs and feedstuffs as well as indoor environments, and being implicated in various human and animal health hazards, from mycoses to mycotoxicoses due to production of sterigmatocystin (STC). Of note, it has been already reported that their high prevalence in some specific occupational environments is a consequence of this evidence, such as swineries (Sabino et al. 2012;Viegas et al. 2013).

STC production ability
Mycotoxin STC synthesis is restricted to species in four sections in Aspergillus (Ochraceorosei, Versicolores, Nidulantes and Flavi) (Rank et al. 2011). Most of the aspergillus species from the section Versicolores are able to produce STC, namely, A. amoenus, A. creber, A. cvjetkovicii, A. fructus, A. griseoaurantiacus, A. hongkongensis, A. jensenii, A. pepii, A. protuberus, A. puulaauensis, A. subversicolor, A. tennesseensis, A. venenatus and A. versicolor (Jurjievic et al. 2013;Visagie et al. 2014;Jakšić Despot et al. 2017). According to Frisvad (2015), the metabolic profile is unique for each fungal species entity with a high degree of chemoconsistency among different isolates of the particular species. But, so far, at the section Versicolores, there are chemical markers characterised just for A. versicolor and A. sydowii (Samson et al. 2010). Liquid Chromatography Mass Spectrometry (LC-MS) LC/MS-based methods proved to be accurate to identify chromatograms of fungal extracts in general. At the moment, some other extrolites of A. versicolor were found, like polyketides (Lee et al. 2010), stephacidin A and notoamide B (Greshock et al. 2008) or kipukasins (Jiao et al. 2007), respectively. STC and dihydrosterigmatocystin are the penultimate precursors of aflatoxinspolyketide-derived furanocoumarins. It has been demonstrated that 25 identified genes clustered within a 70-kb DNA region in the chromosome are involved in their biosynthesis (Townsend 1997). The homologous genes and their corresponding enzymes acting in each bioconversion step in the biochemical pathway common to aflatoxins and STC were described later on as well (Yu et al. 2004). For example, in A. nidulans, the last in the row crucial gene seems to be stcP encoding O-methyltransferase B required for the conversion of dimethylsterigmatocystin to STC (Kelkar et al. 1996).

STC toxicity
STC is a possible human carcinogen (2B) according to IARC classification (McConnell and Garner 1994) and showed immunotoxic and immunomodulatory activity (Liu et al. 2014), together with mutagenic effects (Gao et al. 2015). It might be found in numerous  Viegas et al. (2017) substrates, from foods and feeds to chronically damp building materials and indoor dust. In 2013, the European Food Safety Authority (EFSA) was asked by the European Commission to deliver a scientific opinion on STC in food and feed. The Panel on Contaminants in the Food Chain (CONTAM) from EFSA was responsible for this opinion. However, due to the absence of exposure data for the European population, the margin of exposure approach for substances that are genotoxic and carcinogenic could not be applied for STC, and therefore, the risk of STC for human health was not characterised.
Despite this, it was possible to collect all the available information related to STC toxicokinetics, toxicity, mode of action and dose-response assessment by comparing with aflatoxin B1. The following information was available in the EFSA Journal and, more precisely, in the scientific opinion on the risk for public and animal health related to the presence of STC in food and feed (EFSA Panel on Contaminants in the Food Chain (CONTAM) 2013).
This report concludes the exposure to STC to be of low concern for public health based on the relative carcinogenic potency of STC and AFB1 and exposure data. However, the need of data concerning exposure of European citizens was also mentioned (EFSA Panel on Contaminants in the Food Chain (CONTAM) 2013).

STC toxicokinetics
There is limited information available related to STC toxicokinetics. However, the accessible data suggests that absorption of STC is limited following oral exposure.
In the same way, data on the biotransformation of STC is also insufficient. Few studies published to date indicate that phase I metabolism of STC comprises cytochrome P450 (CYP450)-mediated formation of a reactive epoxide as well as monohydroxylation and dihydroxylation reactions. In a more detailed manner, STC is metabolised in the liver and lung by various CYP450 enzymes into different hydroxymetabolites and its reactive exo-epoxide that readily forms DNA adducts (EFSA Panel on Contaminants in the Food Chain (CONTAM) 2013; Walkow et al. 1985).
As phase II metabolites, a glucuronide of STC and of monohydroxy-STC has been observed and reported, together with a sulphate conjugate of monohydroxy-STC and a glutathione adduct of a monooxygenated STC. Excretion of both conjugated parent STC and its hydroxylated metabolites occurs via bile and urine. Nevertheless, the structure of most of these metabolites is not completely known and more research is necessary to allow the availability of more detailed information (Walkow et al. 1985; EFSA Panel on Contaminants in the Food Chain (CONTAM) 2013).

Toxicity of STC
Due to the structural similarities, aflatoxins and STC share relevant toxic effects, including genotoxicity and carcinogenicity (Miller and Trenholm 1994; EFSA Panel on Contaminants in the Food Chain (CONTAM) 2013). However, in contrast to aflatoxins, only limited information on occurrence and toxicity of STC is available.
Liver and kidneys are the target organs of acute toxicity. However, the acute oral toxicity is relatively low (range between 120 and 166 mg/kg body weight). STC is hepatotoxic in rat, mouse, monkey and guinea pig. The incidence of hepatocellular necrosis and haemorrhages increases with dose and duration of exposure. In the kidney, hyaline degeneration, tubular necrosis and haemorrhages were described in rats and/or monkeys exposed to STC (Purchase and Van Der Watt 1969; EFSA Panel on Contaminants in the Food Chain (CONTAM) 2013).
Results from in vivo and in vitro studies suggest that STC may have also immunomodulatory activity, but strong conclusions cannot be drawn (Huang et al. (2002), Xing et al. (2005), and Zhang et al. (2012)

cited in EFSA Panel on Contaminants in the Food Chain (CONTAM) 2013)).
STC is mutagenic in both bacterial and mammalian cell assays after metabolic activation. Subsequently, STC induces chromosomal damage both in vitro and in vivo in experimental animals (Curry et al. 1984;Ueda et al. 1984;Mori et al. 1986;Crofton-Sleigh et al. 1993;Abdel-Wahhab et al. 2005).
Various studies aimed to compare the genotoxicity of STC and AFB1. However, the uncertainty regarding their actual concentration in the test system, the efficiency of the activation/detoxification metabolic routes and the repair rate of induced lesions does not allow a direct comparison of the relative mutagenic potency of these mycotoxins (EFSA Panel on Contaminants in the Food Chain (CONTAM) 2013).
In previous studies, tumourigenicity of STC was observed after oral, intraperitoneal, subcutaneous and/ or dermal administration in the animal species tested (rat, mouse, Mongolian gerbils, monkey and fish). After oral exposure, premalignant and malignant lesions such as hepatocellular carcinomas (HCC), haemangiosarcomas in the liver, angiosarcomas in the brown fat, lung adenomas and incidental findings in other organs were reported (EFSA Panel on Contaminants in the Food Chain (CONTAM) 2013).
Base on the available information, the CONTAM Panel of EFSA concluded that STC is genotoxic and carcinogenic (EFSA Panel on Contaminants in the Food Chain (CONTAM) 2013).
Additionally, a study developed by Miller et al. (2010), which considered exposure of STC via the indoor environment, concluded that following inhalation (intratracheal installation) of STC, a non-specific but severe inflammatory response of the lung tissue was observed. Similarly, severe cytotoxic and inflammatory damage of lung tissue as well as breaking down of self-cleaning mechanism of airways in rats in vivo were observed after intratracheal instillation of STC containing complex extrolites of an A. versicolor strain of indoor origin in the studies by Piecková et al. (2011Piecková et al. ( , 2015.

STC mode of action
The mode of action of STC can be described as follows. Phase I metabolism results in metabolic activation that promotes the formation of N7-guanyl DNA adducts. These adducts are likely to be responsible for the STE mutagenic effects (Essigmann et al. 1979(Essigmann et al. , 1980. A dose-dependent formation of DNA adducts of STC was found in the concentration range between 1 and 3 mg STC per liver (Reddy et al. 1985; cited in EFSA Panel on Contaminants in the Food Chain (CONTAM) 2013).
STC induces cytotoxicity, inhibition of cell cycle and mitosis, as well as an increased formation of reactive oxygen species (ROS) and lipid peroxidation in vivo (Kawai et al. 1984;Ueno et al. 1995;Xie et al. 2000;Sivakumar et al. 2001;Bünger et al. 2004).
The conclusion made by EFSA Panel on Contaminants in the Food Chain (CONTAM) (2013) was that the genotoxicity of STC is based on the formation of DNA adducts that, if unrepaired, increase the likelihood of mutation fixation. Moreover, when comparing with AFB1, most in vitro studies with purified DNA indicate that the level of induced N7-guanyl adducts is higher after AFB1 than STC exposure, supporting the view that AFB1 is a more potent liver carcinogen than STC. Various in vitro and in vivo investigations have demonstrated that STC exerts cytotoxicity, inhibition of cell cycle and mitosis, as well as an increased ROS formation and lipid peroxidation in vivo. However, most of the in vitro assays have been conducted with rather high STC concentrations, not representing the real human exposure scenario that should be a chronic exposure. Therefore, the observed effects of those studies have to be interpreted with caution not allowing to make conclusions regarding the potential adverse effects of (low dose) dietary exposure to STC (EFSA Panel on Contaminants in the Food Chain (CONTAM) 2013).
More recently, a study developed by Wang et al. (2015) tried to confirm that STC exposure is a risk factor for oesophageal cancer and that STC may induce DNA damage and G2 phase arrest in immortalised human oesophageal epithelial cells (Het-1A). Indeed, the study developed allowed to conclude that STC can induce different cell cycle arrest in primary human oesophageal epithelial cells and immortalised human oesophageal epithelial cells in vitro (Wang et al. 2015).
In 2017, Jiang and co-authors aimed to investigate whether checkpoint adaptation occurs in GES-1 Cellosaurus cell line (GES-1) cells treated with STC. The results suggested that STC induces an initial G2 arrest that is subsequently followed by G2 phase checkpoint adaptation, which may potentially promote genomic instability and result in tumorigenesis (Jiang et al. 2017).
Additionally, a study developed by Huang et al. (2014) in human pulmonary cells in vitro observed that STC induced DNA damage and affected key proteins involved in cell cycle regulation to trigger genomic instability, which may be a potential mechanism underlying the developmental basis of lung carcinogenesis.

Dose-response modelling
Despite the evidence on genotoxicity and carcinogenicity, only a limited tumourigenicity database was available for dose-response assessment since most of the studies published have several limitations that do not allow to be used for dose-response modelling (EFSA Panel on Contaminants in the Food Chain (CONTAM) 2013).
Being aware of this limitation, the CONTAM Panel of EFSA compared the carcinogenic potency of STC and AFB1 in the BMD10 values. After the comparison of the BMD10 of STC for the occurrence of haemangiosarcomas and that of AFB1 for the occurrence of HCC, the CONTAM Panel concluded the carcinogenic potency of STC is approximately three orders of magnitude lower than that of AFB1 (EFSA Panel on Contaminants in the Food Chain (CONTAM) 2013).

Future perspectives regarding food and feed contamination
The report by the CONTAM Panel from EFSA mentioned the need for more occurrence data on STC in food and feed across European countries to allow an accurate assessment of dietary exposure. Furthermore, the prediction of climate change and how it can influence fungal contamination and mycotoxin production should be considered. Therefore, besides not knowing in detail what is the actual exposure to this mycotoxin in Europe, the new scenario of climate change brings new challenges due to a probable new exposure trends, particularly in countries with temperate climate.