Genotoxicity and Antigenotoxicity of selected South African indigenous plants

aToxicology and Ethnoveterinary Medicine, Public Health and Zoonosis Programme, Agricultural Research Council-Onderstepoort Veterinary Research, Private Bag X05, Onderstepoort, 0110, South Africa; bDepartment of Paraclinical Sciences, University of Pretoria, Private Bag X04, Onderstepoort, 0110, South Africa; cDepartment of Pharmaceutical Sciences, University of Antwerp, B-2610 Antwerp, Belgium; dLaboratory of Toxicology, O.D. Public Health and Surveillance, Scientific Institute of Public Health, Juliette Wytsmanstreet 14, B-1050 Brussels, Belgium; cDepartment of Biomedical Sciences, University of Antwerp, B-2610 Antwerp, Belgium


Introduction
Deoxyribonucleic acid (DNA) damage in living organisms occurs spontaneously or could be induced by genotoxins and can lead to gene mutations, chromosomal aberrations and rearrangement of the chromosomes through translocation, deletion and inversion (Wang et al. 2003;Sloczynska et al. 2014).
Mutagenicity plays a crucial role in carcinogenesis and it may lead to different types of cancers and genetic diseases, which are increasing at an alarming rate in human beings and animals (Nagarathna et al. 2013).
Globally, cancer is one of the leading diseases and is expected to become the leading cause of morbidity and mortality in the next decades (Canceratlas.cancer.org 2014).
Aflatoxins, a class of mycotoxins, contaminate various foodstuffs including animal feeds and foods such as nuts, corn, cereals, oilseeds, and dehydrated foods during production, harvest, storage and food processing (Bennett and Klich 2003;Madrigal-Santillan et al. 2010). They are the most common known mutagens and linked with the incidences of genetic diseases, especially hepatocellular cancer and other liver diseases such as aflatoxicosis. Aflatoxins consist of four major groups namely, B1, B2, G1 and G2 (Zain 2011). However, aflatoxin B1 is the most potent genotoxin, highly mutagenic and carcinogenic metabolite known so far. They are recognized as human carcinogens (class 1) by the international agency for research on cancer (IARC). Aflatoxin B1 is metabolized in the liver cells by cytochrome P450 enzyme into a highly reactive aflatoxin B1-8, 9-epoxide, which binds to the guanine residues forming G to T transversion mutation. This biotransformation of aflatoxin B1 induces DNA adducts which leads to mutation, genetic and oxidative damage, thus resulting in cancer (Bhat et al. 2010;Ferrante et al. 2012;Tiemersma et al. 2001).
Various strategies have been employed in the control and prevention of contamination with aflatoxins, but most of them have major drawbacks that limit their use, starting from limited efficacy due to limitless reservoir to loss of essential nutrients and high costs. Therefore, potential strategies that will detoxify aflatoxins without altering the nutritional value of food and feed are needed. Scientists today are exploring the plant kingdom to search for antimutagens or anticarcinogens that are capable of decreasing or inhibiting the mutagenic effects of aflatoxins (Alabi et al. 2011;Sloczynska et al. 2014). Plants contain many bioactive compounds with promising activity against many diseases including genetic diseases such as cancer that could be explored for drug discovery and development (Palombo 2011;Street and Prinsloo 2013).
This study focused on the screening of South African indigenous plants for their antimutagenic or antigenotoxic potentials against aflatoxin B1 induced mutagenicity. These plant extracts were also evaluated for their mutagenicity to confirm that they were not mutagenic. The plants were selected based on results from preliminary screening in our laboratory (unpublished results). The antigenotoxicity of the plant extracts was tested using the Salmonella microsome and Vitotox assays. These two assays are genotoxicity bioassays commonly used in the screening of genotoxic substances (Sloczynska et al. 2014;Verschaeve et al. 1999).

Sample collection and processing
Twenty-two plant species collected from South African National botanical gardens (Lowveld, Walter Sisulu and Pretoria) and in the university of Pretoria botanical garden (Manie van der Schijff Botanical Garden) are listed in Table 1. The table also shows the common names, plant part used as well as the accession number for the plants. The plant material (leaves, seeds or fruits) was dried in an oven set at 45 o C.
Thereafter, the plant material was ground to a fine powder and stored in airtight containers in the dark at room temperature until use. Voucher specimens for the collected plant species were deposited in the H.G.W.J. Schweickerdt herbarium of the University of Pretoria.

Sample extraction and preparation
Ten grams of ground powder of each plant material was sequentially extracted with 100 mL of dichloromethane (Merck) followed by 90% methanol (Merck) by vigorous shaking for 2 h in a rotary shaker.
Thereafter, the crude extracts were filtered under vacuum using Whatman No.1 filter paper (Merck).
Organic solvents were concentrated using a rotary evaporator (Buchi) and then dried under a stream of cold air. Stock solutions of 100 mg/mL extracts were prepared and dissolved in dimethyl sulfoxide (DMSO; Merck) or methanol.

Ames assay
The Ames assay was performed using the pre-incubation test. Two S. typhimurium tester strains were used in the Ames test, including the frame shift mutation detecting strain TA98 and the base-pair substitution detecting strain TA100 (Moltox) as described by Maron and Ames (1983 where T is the number of revertants per plate in the presence of mutagen and the test solution and M is the number of revertants per plate in the positive control (Ong et al.1986). Absence of toxicity was confirmed by the presence of a background layer of bacterial growth in the plate.

Vitotox test
The Vitotox test was performed as described by Verschaeve et al. (1999)  Light production was measured every 5 min in each well for 4 h at 30 o C using a luminometer (Modulus Microplate Multimode Reader, Turner Biosystems). Antimutagenicity of the plant extracts against aflatoxin B1 was measured by adding 1 µg/mL of the aflatoxin B1 to each well. The signal to noise ratio (S/N) which is the light production of exposed cells divided by the light production of non-exposed (control) cells, was automatically calculated for each measurement. Genotoxicity of each sample was evaluated with the Genox/Cytox ratio. A ratio exceeding 1.5 shows genotoxicity in non-cytotoxic extracts provided that the signal is not generated in the first 20 min of measurement. However, the extract is considered toxic if S/N (for rec N2-4 and/or pr 1) rapidly decreases below 0.8. Antimutagenicity of the test sample expressed as percentage inhibition of mutagenicity was calculated as in Ames assay.

Statistical methods
Antigenotoxicity data obtained from the Ames assay was analysed using the Statistical Analysis System software package. Analyses of variance were performed using one-way ANOVA procedures and Dunnet's test to determine the significant differences between the mean (P<0.05). No statistical analysis was necessary for the Vitotox assay.

Results and Discussion
Dichloromethane and 90% methanolic extracts of the selected 22 plant species were investigated first for their potential mutagenic effects in the bacterial based Ames and Vitotox assays. This was done to rule out extracts that exhibited both genotoxic and antigenotoxic effects as they would not be good candidates in further studies. The number of revertant colonies obtained from TA98 and TA100 are in agreement with results generated in our laboratory and in accordance with those reported in literature (Maron and Ames 1983). The two strains are widely used in mutagenicity testing because they are sensitive in detecting most mutagens and carcinogens ( Results on the mutagenic effects of methanolic and dichloromethane plant extracts tested in the Ames assay using S. typhimurium strain TA 100 and TA 98 are presented in Tables 2 and 3. In the Ames test for the used TA98 and TA100, an extract is considered mutagenic when the mean number of revertant colonies produced in each plate was double or greater than two times that of the negative control (Bierkens et al. 2004;Ndhlala et al. 2010). Accordingly, most of the plant extracts tested did not have any mutagenic properties. Only methanolic extracts of M. junodis were mutagenic on TA 98 strain in a dose dependent manner, while P. hybrid produced double the number of revertant colonies as the negative control at the highest concentration tested. Few more plant extracts produced double or more than double the number of revertant colonies as the negative control on strain TA 100 without showing a dose response. These include the methanolic extracts of X. parviflora, Xylopia sp. and R. laetans. While the dichloromethane extracts of M. junodis produced more than double the colonies compared to the negative control at the lowest concentration used when tested against TA 100 tester strain. The same was observed in the Vitotox test for dichloromethane extract of U. caffra ( Figure 1B). In this instance, dichloromethane extracts of U.
caffra induced signal to noise ratio of strain rec N2-4 over the maximum signal to noise ratio of pr1 signal to above 1.5, it was also not cytotoxic as the signal to noise ratio in pr1 was not below 0.8 in a dose dependent manner. Moreover, all 44 plant extracts (methanolic and dichloromethane extracts) tested on Vitotox assay showed no evidence of genotoxicity at all tested concentrations in the absence of S9 metabolizing enzyme as none of the extracts had signal to noise ratio of more than 1.5 ( Figure 1A, 2A).
Methanolic plant extracts of H. monopetalus, Xylopia sp., L. rovulata and P. henkellii, were genotoxic in the presence of S9 in a dose dependent manner ( Figure 2B) while dichloromethane extract, P. roupelliae was genotoxic in the presence of S9 metabolizing enzymes ( Figure 1B). However, there was an increase in light production in the cytox strain, therefore these plants extracts, which showed genotoxicity are considered not genotoxic because there was an interaction between the lux gene and plants extracts. There is usually a very good correlation, about 95%, between the Ames assay and Vitotox test (Westerinck et al. 2009).
However, there may be also variations that may be observed between the two assays ascribed to the fact that different endpoints are tested (true gene mutations against SOS induction). This was also seen for a  (Schoonen et al. 2009;Westerink et al. 2009).
The Vitotox assay also allows detection of cytotoxic compounds. It uses the Cytox strain (pr1) which contains the plasmid with lux operon under transcriptional control of a constitutive promoter, thus constitutively expresses the lux operon (Chichioco-Hernandez et al. 2011;Verschaeve and others 1999). In the presence of cytotoxic compounds, there is a decrease in light production. However, the Cytox strains can also be used as the reference for non-specific enhancement of light emission (Verschaeve et al. 1999).
Therefore, the lack of a dose response in the mutagenicity test using Vitotox is due to toxicity of the highest dose tested. The S/N curve for pr1 strain, which is a useful tool in testing for toxicity alone, was below 0.8 and therefore clearly indicative of the toxicity of the highest dose used for these extracts. These plant extracts with mutagenic effects should be used with care in any form of prescription and further rigorous toxicological investigations are required before they are recommended in pharmaceuticals and drug discovery industries (Verschaeve and Van Staden 2008).
The results on cytotoxicity in the Vitotox assay showed that almost all of the methanolic and dichloromethane plant extracts were toxic at the highest concentration (0.5 mg/mL) when tested without metabolic activation. An exception was the methanolic extracts of P. falcutus, A.brachypetalus (fruit) and R.  (Figure 3, 4).
A test solution is considered antimutagenic when the frequency of genetic damage caused by the combined treatments (extracts and aflatoxin B1) is substantially lower compared to the damage induced by the mycotoxin alone. Usually, an extract is considered to have no or only weak antimutagenic properties when the percentage inhibition of mutagenicity is less than 25. When the percentage inhibition is between 25 and 40%, the extract is considered to have moderate antimutagenic properties. Finally, the extract is said to possess a strong antimutagenic activity if the percentage inhibition is greater than 40% (Abdillahi et al. 2012;Ong et al. 1986;Verschaeve and Van Staden 2008 The methanolic plant extracts tested against S. typhimurium strain TA 98 ( Figure 5B) showed strong antimutagenic properties compared to the extracts tested with strain TA 100 ( Figure 5A) whereas the antigenotoxicity of some plant extracts was influenced by the cytotoxicity of the extracts at higher concentration. Lower concentration of extract showed weak antigenotoxicity against aflatoxin B1. Of the 86% antigenotoxic extracts, 59% of the plant extracts had antigenotoxic activities of above 40% at 0.5mg/mL whereas A. brachypetalus (fruit) and L. rovulata showed moderate to weak antigenotoxicity and co-mutagenic effect against aflatoxin B1 mutagenicity ( Figure 7B). R. laetans as well as dichloromethane extracts of P. falcutus also showed interesting antigenotoxic activities in the Ames (TA100 and TA98) and Vitotox assays.
Plant extracts of Xylopia sp. were not mutagenic when tested alone. However, they showed a co-mutagenic effect with aflatoxin B1 by enhancing the mutagenic effect of the mycotoxin. Literature data on the interaction of the plant extracts investigated in this study with DNA are limited. However, the comutagenic effect of P. henkelii with 4-nitroquinoline-1-oxide (4NQO) mutagenicity has been recently reported (Makhafola et al., 2016). Extracts of P. henkelii were not comutagenic in this study which is an indication that the extracts exert their effect on direct mutagens such as 4 NQO rather than indirect mutagens. A number of previous studies suggest that other natural products including coumarins and flavonoids exerted synergistic effects on aflatoxin B1-induced mutagenicity and other direct and indirect mutagens (Goeger et al. 1999;Snijman et al. 2007). However, the comutagenic effect with AFB1 was attributed largely to an increase in the bioactivation of aflatoxin B1 to its AFB1-8,9-expoxide (Goeger et al. 1999;Snijman et al. 2007).
This study investigated plant extracts from members of different families including Anonnaceae, Asparagaceae, Asteraceae, Podocarpaceae, Proteaceae and Vitaceae. The mechanism by which some of these extracts reduced the mutagenicity of aflatoxin B1 is so far unknown. However, members of these families have been reported to contain sterols, terpenes, alkaloids, acetogenins, glycosides, amino acids and proteins as well as phenolic compounds (Mulholland et al. 2000;Parmena et al. 2012). It is well established that AFB1 requires activation by cytochrome B-450 microsomal mixed function oxidase system into AFB1-8,9-epoxide. The epoxide form adducts with DNA or undergo a detoxification process through conjugation with glutathione to form AFB1-glutathione conjugate, which are thereafter excreted. Various natural products, including those reported in species under investigation, exert their antimutagenic effect by either reducing metabolic activation of the promutagen or through interaction with its metabolic activation derivatives (Waters et al. 1990;Jeng et al. 2000). However, most compounds antimutagenic to AFB1 are intracellular blocking agents i.e. bioantimutagens and act through prevention of AFB1 from reacting with target sites, affecting DNA repair, scavenging of radicals or prevention of neoplasmic expression of initiated cells (Water et al. 1990).

Conclusion
Most plant extracts investigated in this study had antigenotoxic activities against aflatoxin B1 induced mutagenicity in either the Ames or Vitotox test or both. Although the mechanism of action of these extracts is unknown, however, it is well-known that AFB1 exerts its mutagenic effect through oxidative stress. Few plant extracts such as A. brachypetalus, H. petiole, M. caffra, P. hybrid and P. roupeliae had strong to moderate antigenotoxic activity in both tests. The activity of the latter plant extracts is of particular interest and could be confirmed in other in vitro assays such as the mammalian cells-based comet and micronucleus assays. Extracts with low toxicity could further be investigated in in vivo assays in rodents.
The bioactive plant extracts contain a complex mixture of different classes of natural products that may act in a synergistic or antagonistic manner. Further studies to characterize the active antimuatgenic compounds may therefore lead to the discovery of interesting molecules that may play an important role in liver cancer prevention.