Multi-Toxic Endpoints of the Foodborne Mycotoxins in Nematode Caenorhabditis elegans

Aflatoxins B1 (AFB1), deoxynivalenol (DON), fumonisin B1 (FB1), T-2 toxin (T-2), and zearalenone (ZEA) are the major foodborne mycotoxins of public health concerns. In the present study, the multiple toxic endpoints of these naturally-occurring mycotoxins were evaluated in Caenorhabditis elegans model for their lethality, toxic effects on growth and reproduction, as well as influence on lifespan. We found that the lethality endpoint was more sensitive for T-2 toxicity with the EC50 at 1.38 mg/L, the growth endpoint was relatively sensitive for AFB1 toxic effects, and the reproduction endpoint was more sensitive for toxicities of AFB1, FB1, and ZEA. Moreover, the lifespan endpoint was sensitive to toxic effects of all five tested mycotoxins. Data obtained from this study may serve as an important contribution to knowledge on assessment of mycotoxin toxic effects, especially for assessing developmental and reproductive toxic effects, using the C. elegans model.


Introduction
Mycotoxins are toxic secondary metabolites produced by fungi growing on agricultural commodities in the field or during storage [1,2]. These naturally-occurring mycotoxins display diverse chemical structures accounting for their differing biological properties and effects [3].
Aflatoxins represent a group of closely related difuranocoumarin compounds mainly produced by Aspergillus flavus, A. parasiticus, and four naturally-occurring aflatoxins (B 1 , B 2 , G 1 , and G 2 ) were identified. Aflatoxins have been found in a variety of agricultural commodities, but the most pronounced contamination has been encountered in maize, peanuts, cotton seed, and tree nuts with the levels ranged from 0.11 to 4030 µg/kg [9]. Aflatoxin B 1 (AFB 1 ) is the most prevalent and toxic, and is also known as being one of the most potent genotoxic agents and hepatocarcinogens [10,11]. Developmental and reproductive toxic effects and immunotoxic effects of AFB 1 have recently been recognized in the research field. Nevertheless, there are fewer studies in the literature for studying toxic effects of foodborne mycotoxins in C. elegans. Evidence that could link C. elegans with mycotoxin toxicities in humans included the presence of CYP450 orthologue, which can metabolize AFB 1 similar to in humans, as well as various orthologues of glutathione transferase, one of the most well-known phase II metabolic and detoxification mechanisms of AFB 1 . Furthermore, there is a high degree of conservation between C. elegans and mammalian species in processes controlling development, neurobiology, and stress responses, which allow us to explore molecular mechanisms of reproductive, developmental, and transgenerational effects of mycotoxins.
In this study, we investigated multiple toxic endpoints of common foodborne mycotoxins, AFB 1 , DON, FB 1 , T-2, and ZEA with structures shown in Figure 1, in C. elegans model, including lethality, toxic effects on growth and reproduction as well as influence on lifespan. Data presented clearly demonstrated that C. elegans model can predict toxic effects of mycotoxins, and can use for mechanistic studies of mycotoxins-induced adverse health effects.

Lethality
The wild-type N2 strain was treated with various concentrations of tested mycotoxins for 24 h and their LC50 values were calculated and shown in Table 1. Lethality in control worms was less than 10% in all cases. The potency for lethality, as represented by LC50 values, in the wild-type N2 worms was T-2 > AFB1 > ZEA > FB1 > DON. Toxicity ranking based on LC50 is T-2 (1 mg) > AFB1 (20 mg) > ZEA (76 mg) > FB1 (235 mg) > DON (657 mg).

Lethality
The wild-type N2 strain was treated with various concentrations of tested mycotoxins for 24 h and their LC 50 values were calculated and shown in Table 1. Lethality in control worms was less than 10% in all cases. The potency for lethality, as represented by LC 50 values, in the wild-type N2 worms was T-2 > AFB 1 > ZEA > FB 1 > DON. Toxicity ranking based on LC 50 is T-2 (1 mg) > AFB 1 (20 mg) > ZEA (76 mg) > FB 1 (235 mg) > DON (657 mg).

Toxic Effects on Growth
As shown in Figure 2, five mycotoxins affected the growth of worm as indicated by body length, in dose-effect (p < 0.01) and time-effect manner (p < 0.05). Following 72 h exposure, AFB 1 and T-2 at the concentration of 8 mg/L caused the greater growth-inhibitory effects, reaching 52.8% and 41.61% size reduction than untreated controls (p < 0.001). The median effective concentrations (EC 50 ) of AFB 1 and the T-2 was 7.31 mg/L (95%CI: 5.19-12.9 mg/L) and 16.91 mg/L (95%CI: 9.31-59.81 mg/L), which was 300 times lower than that of ZEA and FB 1 . Similar to what found in lethality testing, DON did not result in significant growth inhibitions at the concentrations between 50 mg/L and 800 mg/L ( Table 2). The morphological changes caused by exposure to these five mycotoxins at 72 h were shown in Figure 3.
Toxins 2015, 7 5 did not result in significant growth inhibitions at the concentrations between 50 mg/L and 800 mg/L ( Table 2). The morphological changes caused by exposure to these five mycotoxins at 72 h were shown in Figure 3.

Toxic Effects of Reproduction
Toxic effects on the N2 nematode reproduction, represented by number of offspring (brood size), as a function of mycotoxins concentration were plotted and shown in Figure 4. The average number of offspring for the untreated controls was 133 ± 22, comparable to previous studies [39]. The brood size was significantly reduced in at all tested concentrations for AFB1 (p < 0.001), DON (p < 0.05), FB1 (p < 0.001), T-2 (p < 0.001), and ZEA (p < 0.001), respectively, as compared to that in the untreated controls. Reproductive effects were commonly detectable at much lower concentrations of FB1, which suggested that C. elegans is a much sensitive model for testing reproductive toxic effects of FB1 as compared to other lethality and growth endpoints. The EC50 value with 95% CI was estimated from the concentration-effect curve of each treated mycotoxin and listed in Table 2. The most sensitive mycotoxin is AFB1 with EC50 of 1.69 mg /L (95% CI, 1.38-2.04 mg/L).

Toxic Effects of Reproduction
Toxic effects on the N2 nematode reproduction, represented by number of offspring (brood size), as a function of mycotoxins concentration were plotted and shown in Figure 4. The average number of offspring for the untreated controls was 133˘22, comparable to previous studies [39]. The brood size was significantly reduced in at all tested concentrations for AFB 1 (p < 0.001), DON (p < 0.05), FB 1 (p < 0.001), T-2 (p < 0.001), and ZEA (p < 0.001), respectively, as compared to that in the untreated controls. Reproductive effects were commonly detectable at much lower concentrations of FB 1 , which suggested that C. elegans is a much sensitive model for testing reproductive toxic effects of FB 1 as compared to other lethality and growth endpoints. The EC 50 value with 95% CI was estimated from the concentration-effect curve of each treated mycotoxin and listed in Table 2. The most sensitive mycotoxin is AFB 1 with EC 50 of 1.69 mg /L (95% CI, 1.38-2.04 mg/L).
Toxins 2015, 7 7 Figure 4. Toxic effects on brood size of N2 C. elegans following 72 h exposure to mycotoxins.

Influence on Lifespan
The lifespan of N2 nematode treated with 10% LC50 of five tested mycotoxins was independently recorded in order to test and compare the sensitivity of the assay. Mycotoxin treatment decreased lifespan and increased mortality rate, as illustrated by survival curve (Figure 5A), log cumulative

Influence on Lifespan
The lifespan of N2 nematode treated with 10% LC 50 of five tested mycotoxins was independently recorded in order to test and compare the sensitivity of the assay. Mycotoxin treatment decreased lifespan and increased mortality rate, as illustrated by survival curve (Figure 5A), log cumulative hazard plots ( Figure 5B), and data in Table 3. The survival curves of five mycotoxins were shifted to the left compared to untreated controls. The shape of the cumulative hazard plots, which reflected the rate of aging [40], and the y-intercept of the log cumulative hazard plots of five mycotoxins were significantly larger than that of untreated control (p < 0.0001) as assessed by OASIS. The mean lifespan exposed to AFB 1 , DON, FB 1 , T-2, and ZEA significantly decreased from 17

Discussion
C. elegans has become a popular toxicity test organism, as well reviewed in details [32,34,41]. Much of the early work explored metal toxicity and used lethality as the major endpoint [42]. A wider variety of toxicants have been tested with C. elegans in recent years and more sophisticated sub-lethal endpoints have been developed, including parameters for growth and reproduction [30]. These types of endpoints were directly applied for evaluating environmental toxicants and used as an alternative method for mammalian testing [43]. There were fewer studies in the literature devoted to assess foodborne mycotoxins toxicity using C. elegans. Leung et al. [44] found that AFB 1 induced toxic effects on growth and reproduction in C. elegans at the concentrations of 3, 30, and 100 µM, respectively. The progeny production and development rates of the nematode were significantly reduced when treated with DON at concentrations of 500 and 1000 mg/L [45]. Our present study showed that LC 50 values of tested mycotoxins were at very high concentrations with the exception of T-2 (1.38 mg/L) and AFB 1 (20.47 mg/L). These findings were consistent with results using other model systems [46,47]. Compared to LD 50 or LC 50 values obtained from other model systems such as rats, fish, and human cells in the literature [1,6,48,49] similar acute toxic response was found between rats and C. elegans for T-2. Similar or less acute toxic response for AFB 1 was found in C. elegans (20 mg/L) as compared to values in rats (2.7-17.9 mg/L). C. elegans model is more sensitive for ZEA and less sensitive for DON as compared to LD 50 in rodent model. It is hard to make conclusion for FB 1 because no LD 50 is available in rodents.
Following 72 h exposure, five mycotoxins had significant inhibitory effects on growth and reproduction endpoints of the nematode. Similar to what found for LC 50 , AFB 1 and T-2 had greater inhibitive effects than other tested mycotoxins on growth and reproduction. DON had minimal effects as compared to other four mycotoxins. The LC 50 and EC 50 (growth and reproduction) values (Table 2) were compared to evaluate the sensitivity of these toxic endpoints. As anticipated, large differences between lethality values and effective concentrations of growth or reproduction were found. In the case of AFB 1 , the LC 50 /EC 50 ratio was 2.8 for growth and 12.11 for reproduction, which indicated that the EC 50 of growth and reproduction values are more sensitive than the LC 50 value. Thus, endpoints of growth and reproduction would be much more sensitive indicators of AFB 1 toxicity than endpoints of lethality. Same cases for reproductive toxicity in FB 1 and ZEA were found. On contrary, the lethality endpoint was more sensitive for T-2 than other mycotoxins and none of these three endpoints was sensitive for DON. Findings in our study were consistent with the reports in other species [50].
Traditionally, the lethality assay was a standard toxicity assay of C. elegans model, and the advantage of the assay was the relative ease in scoring worms' mortality and analysis. However, some mycotoxins so far tested, e.g., DON, were not very sensitive to the lethality endpoint, because mycotoxins are secondary metabolites of fungi and their toxic effects cumulated over a period of time [6] in addition to their different target organs and mode of actions. The less sensitive to DON was also observed in the earthworm [50].
Lifespan, rather than physiological indicators, is resulted from complex interactions between genetic, environmental, and stochastic factors and can provide critical insights into the entire life cycle affected by xenobiotics, including mycotoxins. As shown in our data, all five tested mycotoxins could result in shortening lifespan and increase mortality rate in C. elegans. The median lifespan time was significantly decreased following treatment with mycotoxins. These results suggested that influence on lifespan may be a specific endpoint for testing toxic effects of environmental toxicants like mycotoxins. In comparison with most of the other species currently used, lifespan assessment with C. elegans has been simplified and is easy to detect using a microscope, and to analyze with the established software [35]. In summary, we evaluated multiple endpoints for testing mycotoxin toxicities. Lethality endpoint was more sensitive for T-2 toxicity. The toxicity ranking for LC 50 is T-2 (1 mg/L) > AFB1 (20 mg/L) > ZEA (76 mg/L) > FB 1 (235 mg/L) > DON (657 mg/L). Reproduction endpoint was more sensitive for toxicities of AFB 1 , FB 1 , and ZEA. The ranking for reproduction: AFB 1 (2 mg/L) = T-2 (2 mg/L) > ZEA (26 mg/L) = FB 1 (26 mg/L) > DON (487 mg/L). The growth endpoint was also sensitive for AFB 1 toxicity. The ranking for growth: AFB 1 (7 mg/L) > T-2 (17 mg/L) > ZEA (314 mg/L) ě FB 1 (362 mg/L) > DON (533 mg/L). Moreover, the lifespan endpoint was sensitive to test toxic effects of all five mycotoxins. Data obtained from this study may serve as an important contribution to knowledge on evaluation of toxic effects of mycotoxins using C. elegans model, especially for assessing developmental and reproductive toxic effects of mycotoxins exposure in humans and animals.

Materials
Mycotoxins selected for this study, including aflatoxin B 1 , deoxynivalenol, fumonisin B 1 , T-2 toxin, and zearalenone, were purchased from Sigma-Aldrich Chemical Co. (St. Louis, MO, USA). Purity of each toxin (95%-99%) was tested with the appropriate analytical tools (HPLC, LC/MS, and GC/MS). Stock solutions were made with dimethylsulfoxide (DMSO) and kept under argon. Worms used in the present study, wild-type Bristol (N2), and E. coli strain OP50 were purchased from the Caenorhabditis Genetics Center (Minneapolis, MN, USA). All the worms used in this research were hermaphrodites. Worms growth medium (NGM) was made as previously described by Brenner [29]. All other chemicals and reagents were purchased commercially at the highest degree of purity available.

Mycotoxins Exposures
Five mycotoxins in stock solutions were diluted to different concentrations of test solutions. Three-to four-day old worms were dispensed into each well of a 12-well plate. Each well contained a mixture of 990 µL complete K-medium, 10 µL test solution, and OP50. The 1% DMSO was found not affecting nematode growth or reproduction (data not shown). The exposure concentrations were selected based on preliminary lethality assays or solubility testing in complete K-medium with 1% DMSO, e.g., AFB 1 had solubility limits of~50 mg/L K-medium.

Lethalality Assay
All worms were cultured at 20˝C in Petri dishes. Lethality tests were performed on the three-day old wild-type worm for 24 h exposure to different concentrations of mycotoxins using methods described by Donkin and Williams [51]. Briefly, each test consisted of five concentrations plus a control, in which 10˘1 worms (30 worms for each concentration) were transferred to 12-well tissue culture plates containing 1 mL of the test solution in each of five wells. Mycotoxins solutions were prepared in K-medium (0.051 M NaCl and 0.032 M KCl) [52], because worms suffer osmotic stress in deionized water. At the end of the exposure period, worms were counted and scored as live or dead under a microscope; they were judged to be dead if they did not respond to touch using a small, metal wire. All experiments were repeated for three times and the LC 50 values were derived through a Probits analysis.

Measurement of Growth Endpoint
Growth was assessed by measuring change in body length over a 72 h exposure period. The synchronized L-2 worms were used to develop at 20˝C either in control or five mycotoxins at different concentrations in K-medium with food. After exposure 72 h, 20 worms were mounted into a glass pad containing 10% formalin solution. Body length analysis (head to tail) was performed using an Olympus SZX9 microscope (Olympus America Inc. Center Valley, PA, USA) and Infinify analyze software (V5.0.2, Lumenera Corporation, Ottawa, ON, Canada, 2009). Three independent experiments were performed and, for each experiment, at least 20 control and treated worms were analyzed.

Measurement of Reproductive Endpoint
Reproduction was tested using the 72 h assays described by Dhawan et al. [42]. The test solutions consisted of different concentrations of AFB 1 (0-8 mg/L), or DON (0-800 mg/L), FB 1 (0-800 mg/L), T-2 (0-8 mg/L), ZEA (0-80 mg/L), respectively. One adult worm from an age-synchronized culture was placed in each 1 mL of test solution. Three wells were used for each concentration and exposed under the same conditions as described for the growth test. Three days later, the number of offspring at all stages beyond the eggs was determined [39]. For each test concentration and control, the average number of progeny from three wells was obtained for each test replicate, and the testing was repeated three times.

Life-Span Experiment
Lifespan analysis was conducted at 20˝C as described previously [53]. Synchronized young adult worms were placed on NGM agar plates and treated with 10% LC 50 of five tested mycotoxins and 0.1 mg/mL of 5-fluorodeoxyuridine (5-FUDR, Sigma, St. Louis, MO, USA) which was used to block progeny development [54]. Control experiments indicated that 5-FUDR did not affect worms' lifespan. Worms were transferred to fresh treatment plates every other day, and scored every day by gentle prodding with a platinum wire to test for live or dead worms. Those had ceased pharyngeal pumping and failed to move, even after repeated prodding, were scored as dead and removed from the plates. Worms that had crawled off onto the sides of the plate and died away from the agar were censored. A minimum of 100 worms was counted and scored per condition per experiment. Lifespan was defined as the time elapsed from when worms were put on treatment plates to when they were scored as dead. Three independent life-span studies were performed.

Statistical Analysis
The concentration-response relationships for lethality, growth, reproduction, and lifespan were generated from three independent replicate tests. The median lethal concentration (LC 50 ) and median effective concentration (EC 50 , concentration producing a 50% reduction in body size or offsprings compared to control) with 95% confidence intervals (CI), were calculated using logistic regression. Response variables that were not normally distributed were transformed by logarithmic to improve normality. Generalized Linear Models (GLMS) was used to evaluate the significant difference among treatments and between all treatment levels and the control. The SAS 9.4 (SAS Institute, Cary, NC, USA) was used for data analysis and a p-value of 0.05 or less was considered to be statistically significant.
To determine the effects of experimental treatments on survival, a comprehensive comparison of survival datasets between an experimental group and a control group was analyzed using OASIS (online application of survival analysis, http://sbi.postech.ac.kr/oasis) [49]. The average survival time was obtained by using log-rank test, whereas those of a specific time point can be obtained by using Fisher's exact test [13]. If two data sets at 90% mortality show no statistically significant, the weighted log-rank test was used instead of Log-rank test, which developed by Fleming and Harrington [48] and was sensitive to early differences.

Conclusions
We evaluated multiple toxic endpoints for five common foodborne mycotoxins. Lethality endpoint seemed more sensitive for T-2 toxicity. Reproduction endpoint was more sensitive for toxicities of AFB 1 , FB 1 , and ZEA. Growth endpoint was also sensitive for AFB 1 toxicity. Moreover, lifespan endpoint was sensitive to test toxic effects of all five mycotoxins. Data obtained from this study suggests C. elegans model can serve as a good model organism for evaluation of toxic effects of mycotoxins, especially for assessing developmental and reproductive toxic effects.