1. Introduction
Food and drinking water are major sources of human exposure to both mutagens and endocrine-disrupting chemicals (EDCs) globally [
1,
2,
3,
4,
5,
6,
7]. This is alarming in view of the fact that food and water are prerequisites of human life.
The sources of chemical mutagens in food vary remarkably, depending on the foodstuff and processing methodology. However, emphasis has traditionally been placed on reducing the levels of possible mutagenic residues in meat, grain, vegetables
etc. prior to processing, neglecting the possibility of a less clear-cut risk: the formation of these mutagenic compounds in food as a result of processing. Yet, processed food items are reported to contain chemical substances known to have mutagenic, genotoxic and carcinogenic effects, and thus acting as a key global contributor to human cancer risk [
8,
9,
10,
11]. Polycyclic aromatic hydrocarbons (PAHs) and heterocyclic amines have been reported in processed food (mainly meat and fish products) at various concentrations all over the world [
12,
13,
14,
15]. The formation of these chemical mutagens during food processing has been demonstrated to depend on a number of factors such as cooking time, method of cooking and type of heat source [
3,
8,
12]. For example, the Ames test shows a correlation between meat-processing temperature and the number of revertants generated per gram of meat [
3,
4]. Chung
et al. [
12] also reported that charcoal-grilled pork contained higher levels of PAHs (10.2 µg/kg) compared with other methods of processing. Likewise, high concentrations of PAHs have been found in smoked-cured fish in Ghana [
13]
The contamination by mutagenic PAHs of thermally treated high-protein foods such as charcoal-grilled meat products is mainly due to the direct pyrolysis of food fats and the deposition of PAHs from smoke produced through incomplete combustion of the thermal agents [
16]. Unfortunately, this method of food processing is the method of choice in most developing nations, including Nigeria. Although knowledge of proper processing techniques would help reduce the risk of generating mutagenic compounds in food, a recent study showed that only 4.76% of 63 subjects involved in food processing in Nigeria had a formal training in a food safety/hygiene-related discipline [
17]. Similar percentages have also been reported in Kenya and Ghana [
18,
19].
Regarding EDCs, the bulk of information available is on compounds possessing estrogen-like activity. Phytoestrogens and food contact materials are the main sources of human exposure to xenoestrogens in food [
20,
21,
22]. While the health effects of phytoestrogens remain controversial, synthetic xenoestrogens have been associated with certain cancer types, reproductive disorders, developmental abnormalities and other adverse physiological effects in both humans and wildlife [
23,
24,
25]. In this light, it is quite worrisome that drinking water sources as well as bottled mineral and flavored waters have been reported to contain estrogenic substances [
5,
6,
7,
26]. The estrogenic activity in bottled mineral and still water is mainly attributed to the prevailing use of several phthalates and other plasticizers including bisphenol A in packaging materials [
5,
6]. These chemicals are increasingly raising concern, because they may leach into consumer products in normal use [
27,
28,
29]. There are over 50 chemical compounds authorized for use in food contact materials which are known to have endocrine-disrupting potential [
30]. Interestingly, when food contact materials are assessed for their health risk, they are not routinely tested for their endocrine-disrupting potential [
31]. However, the Endocrine Society has expressed its concern about the widespread exposure of humans to these chemicals, as they are capable of affecting multiple endpoints within a living system [
32].
Chemical mutagens and EDCs in food and water samples have usually been determined by various methods of analytical chemistry. However, these methods suffer from a number of limitations in their ability to elucidate the entire range of chemical mutagens and EDCs in a single experiment, including an unknown number of yet-to-be identified compounds. In vitro assays offer the advantage of detecting all substances that contribute to the functional property (mutagenicity or estrogenic activity) being assessed in food, water and environmental samples. Therefore, in the present study, we sought to determine the genotoxic and estrogenic properties of food and water samples by in vitro assays. We focused on Nigerian products, because the customary food processing methods there are potentially risky in this regard (see above) and because, to the best of our knowledge, such information does not yet exist in the body of scientific literature.
4. Discussion and Conclusions
Processed food items and bottled water are consumed in increasing quantities all over the world. Therefore, it is of utmost importance to ensure that in addition to their microbial safety, the products do not contain chemicals which might pose a toxicological risk to consumer health. A conceivable potential risk in this regard is the formation of genotoxic compounds during the processing of foodstuffs and leaching of food contact materials into food and water. Regular screening studies are necessary to verify that the methods used by food vendors are appropriate and sound also from this point of view. The present investigation aimed at exploring the current situation in Nigeria.
To the best of our knowledge, this is the first study on mutagenicity of food products from Africa. To compare our data with those of previous studies is challenging because of the paucity of published data on mixture effects combined with the wide variation in food types in different parts of the world. However, it is possible to compare foodstuffs based on the number of revertants their extracts generate in the Ames test and its derivatives, and we will utilize this approach.
Food processing methods as well as the sales of processed food items in Nigeria are poorly—if ever—regulated. Furthermore, it has been reported that the majority of Nigerians involved in food processing do not have formal training on food safety issues or related techniques [
17,
41]. This may bear on the present finding that the majority of food items (75%) investigated were mutagenic in the standard plate incorporation assay for at least one of the three batches when studied in
Salmonella TA 100 strain. On the other hand, in
Salmonella TA 98 strain only 25% of food extracts were found to yield a mutagenic response, possibly due to a weaker sensitivity of this strain compared with TA 100 or to the type of mutagens present.
The conventional Ames test outcome cannot, however, be taken at its face value in the case of food extracts as these may be sources of localized release of proteins, peptides or histidine itself onto the bacterial plates [
2]. To prevent this potential misinterpretation of ostensible mutagenicity, “treat-and-wash” as well as methylcellulose overlay assays were performed for all samples eliciting a positive outcome in the conventional Ames test. The results of these complementary assays were consistent for some samples (bean cake, suya, hamburger, fried chicken and chin-chin), further reinforcing our initial findings with the Ames test. Extracts of bean cake and suya stood out from among the positive samples. All batches of bean cake exhibited mutagenic activity in the treat-and-wash assay with the
Salmonella TA 100 strain, both in the presence and absence of S9 mix. One of these lots (batch 1) generated a conspicuously high number of revertants, almost five-fold its control (DMSO), with metabolic activation. A similar situation was observed with the MC overlay assay, as all bean cake samples were mutagenic in
Salmonella TA 100 strain in the presence of S9 mix. Similarly, all batches of suya were consistently mutagenic in the treat-and-wash assay, with the
Salmonella TA 100 strain, either in the presence or absence of S9 mix.
Bean cake is commonly consumed in different parts of Nigeria, irrespective of ethnicity, religion or social status. A probable explanation for the mutagenicity test results observed with extracts of bean cake is in the method of its processing. Bean cake is processed by deep-frying for several minutes. Deep-frying has previously been reported to result in the formation of mutagenic and genotoxic compounds in the final product [
42]. Food vendors in Nigeria are also known to repeatedly reuse their frying oil, which is often already of questionable quality, for several days or weeks. This may have contributed to the high number of revertants obtained with extracts of bean cake and fried chicken. Double heat-treatment of cooking oil has been shown to cause an increase in the genotoxic activity of food products [
43,
44]. During frying, cooking oil undergoes deterioration through various chemical and physical processes such as oxidation, polymerization, hydrolysis and cyclization, leading to the formation of both volatile and non-volatile undesirable by-products [
43]. These derivatives are partially absorbed by the fried food, which thus becomes carcinogenic [
45]. For example, the PAH compounds benzo[
a]pyrene and benzo[
a]anthracene are all well-known human carcinogens which have been detected in different types of cooking oil [
45].
The positive mutagenicity test results obtained with suya were not unexpected. Suya is 100% beef, and it is a special type of delicacy, mainly consumed in Nigeria, irrespective of social status. All suya products are processed the same way: by charcoal-grilling. After processing, the products are left to be heated on the charcoal for several hours, until they are purchased. This processing method typically explains the reason for the mutagenicity test results obtained with extracts of suya in our study. The contamination of thermally treated high-protein foods, such as charcoal-grilled meat products, by PAHs and heterocyclic aromatic amines is well established [
12,
46,
47,
48]. The building up of PAHs in this case is due to their generation by direct pyrolysis of food fats and the direct deposition of PAHs from smoke produced through incomplete combustion of the thermal agent [
16]. Heterocyclic aromatic amines, in turn, are formed through the condensation of creatine/creatinine and the strecker degradation radicals (pyridines and pyrazines) generated from the reaction of sugars and amino acids during the Maillard reaction [
49]. The present findings are worrisome, because meat-cooking habits have been linked with several forms of cancer [
50,
51,
52,
53]. In Argentina, for example, cooking meat at a high temperature and close to the cooking source has been linked with increased incidence of colorectal cancer [
54]. This is also the case in Hawaii and the Netherlands [
55,
56]. More recently, a number of PAHs have been reported in different types of smoked meat in Serbia, Latvia and Sweden [
14,
57,
58]. However, no nexus has been established in relation to increased incidence of cancer in these countries. In Nigeria, there is a paucity of data on the incidence of different cancer types, but the two major forms, breast and prostate cancers, may be increasing [
59]. Both of them have been associated with meat-cooking habits [
60].
Hamburger products have previously been reported as a major source of chemical food mutagens to consumers [
38,
61]. The results obtained in our study further reinforce this view, as two different lots of hamburgers examined were found to be mutagenic in all three assays in
Salmonella TA 100 strain, with one of the lots (batch 2) being both directly and indirectly mutagenic in all three assays. Stavric
et al. [
61] previously reported hamburger products purchased in Ontario, Canada, to be mutagenic in a similar assay, but only with
Salmonella TA 98 strain. The number of revertants generated in that study ranged from 63 to 1042 rev/g (average: 199 rev/g). This is in contrast to our study, in which hamburger products were only mutagenic with the
Salmonella TA 100 strain, and not TA 98. In a recent study in Finland [
2], the number of revertants generated with extracts of hamburger products were slightly lower than those obtained in this study, both with
Salmonella TA 100 and 98 strains. Although these findings might seem to implicate the current cooking methods of hamburgers in Nigeria, the present outcome may not be entirely attributable to the processing methods. This is due to the fact that high levels of potassium bromate, a well-known mutagen and human carcinogen, have been detected in bread in different parts of Nigeria [
62,
63,
64]. In one of these cases, Alli
et al. [
62] found that even the lowest level of potassium bromate in their bread samples was over 150 times higher than the maximal permissible limit.
Overall, the mutagenicity test outcome of our study is in keeping with previously published data on food mutagenicity elsewhere [
3,
4,
61], but somewhat at odds with a recent study published in Finland, where only 40% of the processed food items investigated showed mutagenic properties in the conventional Ames test [
2]. In further contrast with the current findings, for most food varieties in the study by Omoruyi and Pohjanvirta [
2], only a single batch proved positive. This may reflect more refined food processing techniques in Finland as compared with Nigeria.
In Nigeria, it is estimated that about 25% and 53% of people living in urban and rural areas, respectively, lack access to pure, portable water [
65]. This is related to recent outbreaks of several water-borne diseases in major states of the country, specifically cholera [
66,
67]. It has prompted entrepreneurs to continuously establish water plants, in which pure water samples are mainly packaged in plastic sachets.
Our study demonstrates that pure water sachets may contain estrogen-like chemicals. Five of the 16 samples investigated were discovered to be estrogenic in our
in vitro test system, with EEQs ranging from 0.79–44.0 ng/L. Both the frequency of positive samples and their concentrations were actually lower than we feared, considering that the proprietors of pure water sachet factories in Nigeria are principally entrepreneurs with little or no knowledge of water quality (microbiological, physicochemical or toxicological). There are two recent studies carried out in Europe in which estrogenic activity of water samples was assessed by a comparable
in vitro yeast assay to that of ours. Pinto and Reali [
68] analyzed mineral waters packed in polyethylene terephthalate (PET) bottles in Italy. The levels they detected varied from 0.03 through 23.1 ng/L (mode 9.5 ng/L) EEQs. Somewhat surprisingly, tap water made of either surface water or ground water contained approximately 15 ng/L EEQs. In another study, Wagner and Oehlmann [
6] determined estrogenic activities in 20 major brands of bottled water in Germany. Twelve of these samples proved positive with the levels ranging from 2.64 to 75.2 ng EEQ/L (average 18.0 ng/L). Interestingly, in their material, the highest estrogenic activities were recorded for waters packaged in either non-reusable PET or reusable glass bottles, and even water packed in Tetra Pak™ bricks contained levels that were similar to those found in our study (14–44 ng/L). Thus, substances exhibiting estrogen-like activity are common in water samples in both industrialized and developing countries.
It is widely believed that the decline in male reproductive functions, increased incidence of different cancer types amongst young men and women and neurobehavioural diseases observed in the population of different countries may, at least partly, be attributable to exposure to estrogenic compounds, particularly during the intrauterine phase or during critical periods of postnatal development [
23,
24,
25,
69]. Studies in recent years have shown, for example, that the commonly used plasticizer, di(2-ethylexyl)phthalate (DEHP), alters gene expression in rats and that, at appropriate concentrations, it alters the development of the central nervous system in the fetus [
70]. Similarly, certain compounds, such as benzophenone used as food contact material, are reported to almost completely block the 17β-hydroxysteroid dehydrogenase type 3 enzymes that are required for testosterone synthesis [
71].
The presence in or leaching into water samples of endocrine-disrupting chemicals is influenced by a number of factors such as storage conditions, exposure to sunlight and ambient temperature [
72,
73]. Unfortunately, the environmental conditions in Nigeria (abundant sunlight and high temperature) tend to favor the migration of endocrine-disrupting chemicals from the packaging materials into water, as, for example, during transport of water containers. Therefore, sachets of water stored or transported in less appropriate conditions than our samples would be at risk of containing higher concentrations of estrogenic substances.
The bulk of dietary xenoestrogen exposure for adults has been proposed to emanate from dairy products, and total daily intake of estrogens has been estimated to be 80–100 ng [
74]. Assuming an average daily water consumption of 3 L at Nigerian latitude [
75], in the worst-case scenario based on our sample material, the intake from pure water sachets would double the estimated exposure. Hence, every effort should be taken to reduce the estrogen levels in these waters in the future.
In conclusion, the results obtained in our study show that both commercially processed food items and sachet-packed pure water sold in Nigeria, are sources of mutagen and estrogen-like chemicals, respectively. Although their concentrations are not alarming in the light of food and water analyses from other countries, measures should be taken to reduce them further and monitor their levels regularly. Since the number of samples examined here was relatively low, further survey studies are also warranted.