Minor contribution of alternariol, alternariol monomethyl ether and tenuazonic acid to the genotoxic properties of extracts from Alternaria alternata infested rice

Toxicology Letters 214 (2012) 46–52

Contents lists available at SciVerse ScienceDirect

Toxicology Letters journal homepage: www.elsevier.com/locate/toxlet

Minor contribution of alternariol, alternariol monomethyl ether and tenuazonic acid to the genotoxic properties of extracts from Alternaria alternata infested rice Christoph Schwarz a , Martin Kreutzer b , Doris Marko a,∗ a b

University of Vienna, Department of Food Chemistry and Toxicology, Waehringer Str. 38, A-1090 Vienna, Austria Institute of Applied Biosciences, Section of Food Toxicology, Universität Karlsruhe (TH), Adenauerring 20, D-76131 Karlsruhe, Germany

h i g h l i g h t s  Complex extracts of Alternaria alternata infested rice possess potent DNA strand breaking properties in vitro.  Genotoxicity exceeds by far the effects of AOH, AME and TA, the major Alternaria mycotoxins with respect to quantity.  Toxicity-guided fractionation underline that AOH, AME and TA play a minor role for genotoxicity of A. alternata extracts.

a r t i c l e

i n f o

Article history: Received 3 March 2012 Received in revised form 31 July 2012 Accepted 3 August 2012 Available online 13 August 2012 Keywords: Comet assay Altertoxin Toxicity-guided fractionation Genotoxicity Mycotoxin

a b s t r a c t Alternaria spp. are known to form a spectrum of secondary metabolites with alternariol (AOH), alternariol monomethyl ether (AME), altenuene (ALT) and tenuazonic acid (TA) as the major mycotoxins with respect to quantity. In the present study we investigated the contribution of these compounds for the DNA damaging properties of complex extracts of Alternaria spp. infested rice. Five different Alternaria strains were cultured on rice and analyzed for their production of AOH, AME, ALT and TA. The extracts of two strains with distinctly different toxin profiles were selected for further toxicological analysis. An extract from A. alternata DSM 1102 infested rice, found to contain predominantly TA, exhibited substantial DNA strand breaking properties in cultured human colon carcinoma cells in the comet assay, whereas TA as a single compound did not affect DNA integrity up to 200 ␮M. An extract of A. alternata DSM 12633 infested rice, containing in comparable proportions AOH, AME and TA, exceeded by far the DNA damaging properties of the single compounds. In contrast to AOH, AME and TA, both selected extracts induced an increase of DNA modifications sensitive to the bacterial repair enzyme formamidopyrimidine DNA glycosylase (FPG) in the comet assay, indicative for oxidative DNA damage. Toxicity-guided fractionation of the DSM 12633 extract confirmed that these effects were not caused by AOH, AME or TA. Taken together, the mycotoxins AOH, AME and TA, representing the major mycotoxins with respect to quantity in A. alternata infested food, play only a subordinate role for the genotoxic properties of complex extracts and appear not to be involved in the induction of FPG sensitive sites. © 2012 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Fungi of the genus Alternaria are important saprophytes, plant pathogens and post harvest pathogens of food and feed crops. More than 100 species have been described, though reliable classifica-

Abbreviations: ACN, acetonitrile; ALT, altenuene; AME, alternariol monomethyl ether; AOH, alternariol; DMSO, dimethyl sulfoxide; DSMZ, German Collection of Microorganisms and Cell Cultures; DSM 1102, Alternaria alternata DSM 1102; DSM 12633, Alternaria alternata DSM 12633; FPG, formamidopyrimidine [fapy]-DNA glycosylase; LoD, limit of detection; LoQ, limit of quantitation; R2 , coefficient of determination; TA, tenuazonic acid. ∗ Corresponding author. Tel.: +43 014277 70800; fax: +43 014277 70899. E-mail address: [email protected] (D. Marko). 0378-4274/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.toxlet.2012.08.002

tion and taxonomy are often difficult. Their ubiquitous occurrence and their ability to grow and produce toxins even under unfavorable conditions, such as low temperatures and low water activity (Hasan, 1995; Magan et al., 1984) make them relevant contaminants of refrigerated fruits and vegetables and stored feedstuffs (Barkai-Golan, 2001; Müller, 1991). Contamination with Alternaria spp. or its toxins has been reported in cereals (Li and Yoshizawa, 2000), corn flakes (Aresta et al., 2003), tomatoes (Andersen and Frisvad, 2004; Hasan, 1995), berries, grapes, citrus fruits (Tournas and Katsoudas, 2005), oilseed rape (Vinas et al., 1994), apples (Robiglio and Lopez, 1995), carrots (Solfrizzo et al., 2004) and other crops. Though over 120 secondary metabolites of Alternaria are known, only about 30 of them are considered toxic to humans and animals, many others acting as phytotoxins (Panigrahi, 1997).

C. Schwarz et al. / Toxicology Letters 214 (2012) 46–52

The toxins alternariol (AOH), alternariol monomethyl ether (AME), altenuene (ALT), tenuazonic acid (TA) and altertoxin-I (ATX-I) are usually regarded as the major toxins produced by Alternaria (Barkai-Golan, 2008). As for toxicity, the majority of Alternaria cultures, isolated from peanuts and various cereals were lethal to rats, when administered orally, and contained mostly TA (Meronuck et al., 1972). The lethal dose for rats was calculated as 100–200 mg/kg TA in this study. Griffin and Chu (1983) did not see any mortality or teratogenicity in the chicken embryo assay by Alternaria toxins, even at doses of 1000 ␮g, 500 ␮g or 1000 ␮g of AOH, AME or ALT per egg respectively, while for TA the 50% lethal dose was 548 ␮g per egg, neither toxin showing any teratogenic effect. Pero et al. (1973) observed an increase in mortality in mice as well as an increase in dead or resorbed fetuses per litter, but only at rather high doses of AOH or AME around 100 mg/kg. Though the acute toxicity of these major toxins is not very high, relatively little is known about possible long term effects through prolonged exposition to small amounts of Alternaria toxins. A possible connection between contamination of cereals with Alternaria spp. in the Chinese Linxian province and the high incidence of esophageal cancer in this area has been suggested (Liu et al., 1992). Extracts from Alternaria cultures, isolated from grains in this province were mutagenic in vitro (Dong et al., 1987; Zhen et al., 1991). The altertoxins I, II and III (ATX-I, II and -III) have been reported as highly mutagenic to Salmonella typhimurium TA98, TA100 and TA1537 (Stack and Prival, 1986), while AOH and AME showed no mutagenicity to TA98 or TA100 (Davis and Stack, 1994). However, Brugger et al. (2006) found AOH to be mutagenic to cultured Chinese hamster lung fibroblasts (V79) and mouse lymphoma cells at concentrations of 10 ␮M AOH and higher. Furthermore, estrogenic potential of AOH has been reported by Lehmann et al. (2006), as well as its ability to induce kinetochorenegative micronuclei in V79 and Ishikawa cells and to inhibit cell proliferation by interference with the cell cycle. AOH and AME were found to induce DNA strand breaks in human carcinoma cell lines (HT29 and A431) at micromolar concentrations, the poisoning of topoisomerases being one mode of action at least contributing to this effect (Fehr et al., 2009). In the present study we addressed the question, whether and to what extent the DNA damaging potential of complex Alternaria extracts is associated with the content of the quantitatively predominant toxins AOH, AME and TA. 2. Materials and methods 2.1. Materials Human colon carcinoma cells (HT29), the Alternaria alternata strains (DSM 1102, DSM 12633, DSM 62006 and DSM 62010) were obtained from the German Collection of Microorganisms and Cell Cultures GmbH (DSMZ, Braunschweig, Germany). Cell culture media and supplements were purchased from GIBCO Invitrogen (Karlsruhe, Germany), Sigma–Aldrich (Steinheim, Germany) and Sarstedt (Nuembrecht, Germany). AOH and AME were synthesized according to the method of Koch et al. (2005) by Dr. Merz (University of Kaiserslautern, Germany), ALT, TA and the other substances used in these studies were purchased from Sigma–Aldrich and Roth (Karlsruhe, Germany). HPLC gradient grade acetonitrile (ACN) was purchased from Acros Organics (Geel, Belgium). Long grain rice was bought at a local supermarket.

47

2.3. Extracts In order to monitor the toxin production of the strains, samples of moldy rice (ca. 1 g) were taken every 3–4 days, mechanically homogenized (UltraTurrax T10) with 2 mL aqueous KCl solution (0.15 mM KCl, pH 3 adjusted with HCl) and extracted twice with ethyl acetate (2 × 5 mL) by vigorous shaking and vortexing. Centrifugation at 4000 × g for 30 min was necessary to achieve good phase separation. The combined organic phases were evaporated to dryness using a ScanSpeed MiniVac evaporator (Labogene, Lynge, Denmark) and dissolved in methanol for HPLC analysis. For use in the comet assay and dichlorofluoresceine (DCF) assay larger samples (ca. 10 g of moldy rice) were similarly extracted with 3 × 15 mL of ethyl acetate after 21 days cultivation, the combined organic phases evaporated to dryness and dissolved in 3–10 mL fresh ethyl acetate. On the basis of their distinctive toxin profile one extract from DSM 1102 and one from DSM 12633 were selected for further investigation. 2.4. Toxin standards and standardized extracts for incubation TA was obtained from Sigma–Aldrich, Germany, as its copper salt. Prior to HPLC analysis or cell incubation the copper had to be removed by ion-exchange as described by Solfrizzo et al. (2004) using Dowex 50 WXS (100–200 mesh). All toxins were dissolved in methanol for HPLC analysis or dimethyl sulfoxide (DMSO) for cell incubation. Evaporating aliquots of 113 ␮L ethyl acetate extract from one selected DSM 1102 extract or 109 ␮L from one selected DSM 12633 extract, and dissolving the residue in 100 ␮L DMSO led to two extracts, standardized to their characteristic major toxin TA or AOH, respectively. Aliquots from these two extracts (DSM 1102 and DSM 12633, see Table 2) were used for all incubations in the comet assays and DCF assays, to allow comparison between the replicates of each experiment as well as between the different experiments. The applied extract concentrations were prepared by diluting the standardized extract with the appropriate amounts of DMSO. 2.5. HPLC analysis The HPLC system, used to analyze the crude fungal extracts, was a Shimadzu Prominence (two binary gradient pumps: LC-20AT, controller: CBM-20A, diode array detector: SPD-M20A UV/VIS, column oven: CTO-20AC, autosampler: SIL20AC and LC-Solutions real time analysis software). A Phenomenex Luna 5u C18(2) 250 mm × 4.6 mm column, equipped with Phenomenex SecurityGuard column, was used for separation. Solvent A was water, adjusted to pH 3 with formic acid, solvent B contained 90%vol ACN and 10%vol water. A gradient program with a constant flow rate of 1 mL/min was used, starting with 19% B for 1 min, increasing to 50% B in 1 min, to 55% B in 10 min, to 100% B in 10 min and holding at 100% B for 5 min. Additional 6 min equilibration time at 19% B were added between runs; the injection volume was 10 ␮L and the column thermostated to 40 ◦ C. Identification was performed by comparing retention times and UV spectra (200–600 nm) of peaks in the sample with those of the pure toxin standards, and external calibration was used for quantitation. Statistical key data of the calibrations are displayed in Table 1, the limits of detection (LoD) and limits of quantitation (LoQ) were calculated based on threefold and tenfold signal to noise (S/N) ratios, respectively. For each calibration concentration three standard solutions were prepared and analyzed, while every calibration curve comprised five concentrations, equidistantly spread over the concentration range (seven concentrations for ALT). Rather high concentration ranges were chosen for the calibrations, as most extracts contained very high levels of at least one analyzed toxin, and the analysis primarily aimed at finding Alternaria strains and culture conditions, that would allow the preparation of extracts with high toxin concentrations. Multiple injections of the every sample in different dilutions (1/5–1/100) were necessary, to include all toxins in the analysis, as their concentrations often differed by several orders of magnitude in the extracts. Furthermore, the lowest calibration concentrations were chosen to still allow an identification of the peaks by their UV spectra, as the crude fungal extracts showed a large number of small, unknown peaks in the HPLC chromatograms and the retention times would slightly vary, depending on the presence or absence of large amounts of other extract components. Due to this necessity for identification by UV spectra, the lowest calibration concentrations were higher than the calculated LoQ. Retention times were 7.4 min for ALT, 9.8 min for AOH, 10.4 min for TA and 16.8 min for AME. 2.6. Comet assay

2.2. Cultivation of fungi Four A. alternata strains (DSM 1102, DSM 12633, DSM 62006 and DSM 62010 were cultivated on moist long-grain rice (150 g rice and 85 mL distilled water in 1 L Erlenmeyer flask, autoclaved and adjusted to 38% absolute humidity with distilled water), at 24 ◦ C in the dark. Pre-cultivation of the fungal strains was done on standard V8-agar in Petri dishes, and six punch cores (sterile 8 mm biopsy punch) per flask were used for inoculation, as especially DSM 1102 showed hardly any sporulation. The flasks were shaken daily, to prevent the formation of large clumps. Four separate inoculations and cultivation experiments were performed for each strain, to analyze how reproducible the toxin formation was under the cultivation conditions used, as the toxin production by Alternaria is known to vary considerably, dependent on a variety of environmental factors.

Single cell gel electrophoresis (comet assay) was performed according to Tice et al. (2000) with HT29 cells and 1 h incubation time, as previously described Fehr et al. (2009). The treatment of duplicates with formamidopyrimidine DNA glycosylase (FPG) further allows the detection of FPG sensitive base modifications such as, e.g. oxidative damage to DNA bases. Dilutions of the standardized DSM 1102 extract were tested, containing concentrations of 0.2–200 ␮M TA in parallel to TA as a single compound at the same concentrations for comparison. Similarly dilutions of the DSM 12633 extract, containing 0.1–50 ␮M AOH were used for incubation in parallel to AOH alone at these concentrations. Cell viabilities of each Petri dish were analyzed after incubation via trypan-blue exclusion and were all between 93% and 97%, thus ruling out artifacts by cytotoxicity. The well-known redox cycler menadione was used as positive control and the vehicle DMSO as negative control. While aliquots of

48

C. Schwarz et al. / Toxicology Letters 214 (2012) 46–52

Table 1 Key data of the external calibrations for HPLC analysis of the analyzed mycotoxins. Each calibration range comprises five concentrations (seven for ALT) and for each concentration three calibration solutions were prepared and analyzed. Calibration range

(␮M)

R2

LoD (ng/mL)

LoQ (ng/mL)

Lowest calibration concentration (ng/mL)

(␮g/mL)

AOH 1–10 10–100 100–1000

0.26–2.6 2.6–25.8 25.8–258.2

0.9960 0.9992 0.9994

10

33

258

AME 1–10 10–100 100–1000

0.27–2.7 2.72–27.2 27.2–272.2

0.9973 0.9983 0.9994

22

72

272

TA 100–1000

19.7–197.2

0.9984

13

43

19700

ALT 1–1000

0.29–292.3

0.9954

27

89

292

the same extracts (Table 2) were used for incubation, each experiment consisted of three replicates, for which the cells were separately cultivated, incubated, processed and scored, in order to achieve a good degree of independence between repetitions, while ensuring comparability with the DCF assay. 2.7. Dichlorofluorescein (DCF) assay Effects on the intracellular redox level were measured via DCF assay according to Keston and Brandt (1965) using 40,000 HT29 cells and 200 ␮L cell culture medium per well in a black 96 well plate. All substances/extracts were dissolved in DMSO resulting in a final concentration of 1% DMSO in colorless DMEM incubation medium. Menadione (20 ␮M) served as positive control, DMSO as negative control. Each experiment comprised four repetitions with cells from different passages, each repetition containing six replicates on a 96 well plate. Analogously to the comet assays, all incubations were performed with aliquots of the same extracts (Table 2), to allow comparison. 2.8. Toxicity-guided fractionation Fractionation was performed by injecting 10 ␮L of a crude ethyl acetate extract of DSM 12633 infested rice into the HPLC and manually collecting fractions at the detector outlet between 0 min, 6 min, 10.3 min, 12 min, 15.8 min, 19 min, 25 min, 30 min and 35 min. The eluates were evaporated in a speed-vac to remove most of the ACN and extracted twice with the double respective volume of ethyl acetate. The organic fractions were evaporated to dryness and the residue dissolved in 50 ␮L DMSO and 5 mL serum containing DMEM for incubation, analogously to the comet assays with the complex extracts and single toxins. By comparison with respective standards, AOH was located in fraction #2, TA in fraction #3 and AME in fraction #5. Irradiation of a Petri dish with UV-B light for 2 min (Vilber BLX-312, 5 × 8 W, UV-Chamber) was used as positive control, the vehicle DMSO served as negative control.

3. Results and discussion 3.1. Fungal culture and toxin profiles Under the chosen cultivation protocol, the Alternaria strains DSM 62010 and DSM 12633 showed strong sporulation, DSM 62006

sporulated considerably less, while DSM 1102 hardly produced any spores at all. As this rendered an inoculation with defined spore suspension difficult, all strains were pre-cultivated on regular V8agar (according to DSMZ) in Petri dishes, and punch cores from the actively growing cultures were taken with a sterile biopsy punch and used for inoculation. As the extracts were intended for incubation of cell culture assays, we used autoclaved rice and cultivation in the dark, as this promised high toxin yields (Burroughs et al., 1976; Haggblom and Niehaus, 1986; Soderhall et al., 1978). All tested Alternaria strains grew similarly fast on moist, autoclaved longgrain rice in the dark, covering all grains after 3–4 days. To monitor the toxin formation by the different strains under the described culture conditions, samples of the rice cultures were taken every 3–4 days and analyzed for their levels of AOH, AME, ALT and TA via HPLC. For each strain, four replicates were prepared by separately inoculating and cultivating four Erlenmeyer flasks. These experiments were not carried out in parallel, but by inoculation at different points in time, thus reducing the risk of systematic errors, especially as uncontrolled environmental factors might considerably affect toxin formation by Alternaria (Burroughs et al., 1976; Haggblom and Niehaus, 1986; Hasan, 1995; Magan et al., 1984). A. alternata DSM 12633 was found to produce large amounts of AOH and AME together with comparable amounts of TA (690 ± 327 ␮g/g rice, 587 ± 250 ␮g/g rice and 1049 ± 363 ␮g/g rice, respectively) after 20 days. In the same time period, the strain DSM 62010 generated mostly TA (4777 ± 1409 ␮g/g rice), less AOH (52 ± 13 ␮g/g rice) and AME (206 ± 111 ␮g/g rice), whereas DSM 1102 nearly exclusively formed TA (3447 ␮g/g rice). The other A. alternata strain DSM 62006 strain exhibited the highest variance while showing rather low overall toxin formation, compared to the other strains (Fig. 1). A cultivation time of 21 days was chosen as most suitable to prepare extracts for cell incubation, as only little increase in toxin concentration was observed after that, or even slight decrease

Table 2 Toxin concentrations of the A. alternata extracts, used for incubation. Results are mean values and standard deviations of 2–3 HPLC injections with different dilutions. AOH and AME were not detected in DSM 1102 extract, ALT was not detected in either extract (all < 0.3 ␮g/mL). For incubation, aliquots were normalized to 50 mM TeA (DSM 1102) or 10 mM AOH (DSM 12633), respectively, by evaporating appropriate amounts of ethyl acetate extract and dissolving the residue in DMSO. These normalized extracts were further diluted to prepare the incubation media (all with 1% DMSO final concentration), thereby diluting all other extract compounds to the same extent. Extracts for comet assays and DCF assay A. alternata DSM 1102 extract

Toxin production of rice culture (␮g/g) Toxin concentration in crude extract (␮g/mL) Aliquots in DMSO for incubation (␮g/mL) (mM)

A. alternata DSM 12633 extract

AOH

AME

TA

–

–

– –

– –

2731 8694 9824 50

± ± ± ±

52 165 187 0.9

AOH

AME

661 ± 31 2386 ± 111 2601 ± 121 10 ± 0.5

277 999 1089 4

TA ± ± ± ±

3 11 12 0.1

794 2866 3124 16

± ± ± ±

20 72 79 0.4

7000

A

6000

mAU

AOH AME ALT TeA

49

600

TA

400 200

5000

0 4000

0

5

10

15

20

25

30

35

time [min]

B

200

mAU

3000

150

2000

0

5

3

5

7

6

8

10

15

20

25

30

35

time [min] Fig. 2. HPLC chromatograms of the (A) A. alternata DSM 1102 and (B) DSM 12633 extract with fractions indicated, using UV detection at 278 nm.

A

***

no enzyme treatment FPG treated

***

***

60 50

**

40 30

*

***

20 10

µM

ad

io

2 µM 20 µ 20 M 0 0. µM 00 2 µ 0. M 02 µM 0. 2 µM 2 µM 20 µ 20 M 0 µM

0 ne

(DSM 62010). Two extracts with distinctive different toxin profiles were selected for further toxicological examination; an A. alternata DSM 1102 extract, containing comparably large amounts of TA (8.7 ± 0.2 mg/mL) but low in AOH, AME or ALT (<0.3 ␮g/mL). In comparison, an A. alternata DSM 12633 extract featured high levels of AOH (2.4 ± 0.1 mg/mL). Aliquots of these two extracts, standardized to 50 mM TA (DSM 1102) or 10 mM AOH (DSM 12633), respectively, by evaporation and dilution were used for the in vitro experiments (see Table 2), thus allowing better comparison between the respective assays. Altogether, the tested strains showed reproducible, strain-dependent profiles of AOH, AME, ALT and TA production, under the chosen culture conditions. Complex extracts with up to several milligrams of these major toxins were obtained, which was the primary objective of the cultivation experiments.

AME

4

0

0. 2

Fig. 1. Profiles of toxin formation by different strains of A. alternata on moist, autoclaved long grain rice, 21 days after inoculation. The rice cultures were extracted with ethyl acetate and analyzed for their major toxins via HPLC with external calibration. Displayed are mean values and standard deviations of four separate inoculations, cultivations and extractions for every strain.

2

1

50

DSM 62010

SO

DSM 62006

µM

DSM 12633

M

DSM 1102

100

D

0

TA AOH

20

1000

tail intensity [%]

toxin concentration in rice culture [µg/g]

C. Schwarz et al. / Toxicology Letters 214 (2012) 46–52

m

**

B 90 80 70 60 50 40 30 20 10 0

**

no enzyme treatment FPG treated

*

***

*

**

***

*

**

*

***

###

m

en

ad DM io SO ne 20 µM 0. 1 µM 1 µM 10 µM 50 µ 0. M 01 µM 0. 1 µM 1 µM 10 µM 50 µM

tail intensity [%]

In the comet assay, TA did not induce DNA strand breaks in HT29 cells during 1 h of incubation over the entire concentration range (0.2–200 ␮M TA) (Fig. 3A). The DSM 1102 extract, however, induced a significant increase in DNA strand breaks at the highest concentration (containing 200 ␮M TA). Furthermore, a strong increase in FPG sensitive sites was observed at the highest extract concentration, which was not apparent in the TA-treated samples (Fig. 3A). Lesion which are detected by the bacterial repair enzyme FPG are often 8-oxo desoxyguanidine or formamido pyrimidine (guanosine with opened imidazole ring) (Coste et al., 2004) and are indicative for oxidative damage to the DNA (Fig. 2). The difference between AOH as a single compound and the AOH-containing DSM 12633 extract is even more pronounced. Previous work showed AOH and AME to possess moderate DNA strand breaking properties in HT29 cells (Fehr et al., 2009). These effects for AOH were confirmed in the present study with tail intensities of about 5% after treatment with 50 ␮M AOH (Fig. 3B). The AOH-containing DSM 12633 extract, however, showed a considerably stronger DNA strand breaking potential, producing tail intensities over 14% and 41% at the second highest and highest extract concentrations (containing 10 ␮M and 50 ␮M AOH, respectively). Moreover, the extract potently induced FPG sensitive sites, exceeding by far the effect of AOH as a single compound. This effect was significant from the second lowest extract concentration onward (0.1 ␮M AOH content) and cannot be explained by the content of AOH (0.1 ␮M), AME (0.04 ␮M) or TA (0.16 ␮M). At the highest extract concentration the DNA of the FPG treated duplicates was damaged to such a high extent (70–90%), that reliable scoring

DSM 1102 extract standardized for TA content

TA

en

3.2. Comet assay

AOH

DSM 12633 extract standardized for AOH content

Fig. 3. Comet assay of the standardized fungal extracts from rice cultures of A. alternata. The highest concentration of DSM 1102 extract (200 ␮M TA) represents 14.4 mg moldy rice per mL incubation medium, the highest concentration of DSM 12633 extract (50 ␮M AOH) represents 19.7 mg moldy rice per milliliter incubation medium, respectively. Dilutions of both extracts are compared to their respective single major toxin TA or AOH. Results are given as mean values of tail intensities with standard deviations from three separate incubations, using aliquots of the same extract as in the DCF assays, for better comparability (see Table 2). Quantification of the FPG treated duplicate was not possible at the highest concentration of DSM 12633 extract, as the tail intensities were too high (###). Significances indicated display the significance level as compared to the respective control calculated by Student’s t-test (*p < 0.05; **p < 0.01; ***p < 0.001).

3.4. Toxicity-guided fractionation To narrow down possible extract compounds, responsible for the DNA strand breaking effects and the induction of FPG sensitive sites, toxicity-guided fractionation of a DSM 12633 extract was performed, by manually collecting 8 fractions of the HPLC eluate at the detector outlet while additionally monitoring the detector signal. The comet assay with parallel FPG treatment is a particularly suitable toxicological endpoint for this purpose, requiring relatively little substance for incubation. In accordance with the results with the single compounds (Fig. 3A + B), fraction #2, containing AOH (∼20 ␮M), did not exhibit pronounced DNA strand breaking quality, neither did fraction #3 containing TA (∼30 ␮M) (Fig. 5A). Fraction #4 induced pronounced tail intensities (11%), but did not contain any of the analyzed toxins. The chromatogram of the fractionation

400 350 300 250 200 150 100 50 µM

µM

µM

0 20

20

2

µM

µM 0. 2

µM

20 0

µM

20

2

m en ad

io

ne

D

M SO 20 µM 0. 2 µM

0

TA

***

TA in DSM 1102 extract

800 700 600

200

*

**

*

300

***

***

400

**

500

100 0

ad

i o DM n e SO 20 µ 0. M 1 µM 1 µM 10 µM 50 µ 0. M 1 µM 1 µM 10 µM 50 µ 0. M 1 µM 1 µM 10 µM 50 µM

B

en

As FPG sensitive sites in the DNA are often interpreted as an indicator for oxidative stress, the DCF assay was used to determine whether the intracellular redox level of HT29 cells are affected by the extracts. Incubation time and extract/toxin concentrations were chosen according to the comet assay. No tested concentration of DSM 1102 extract or TA as a single compound showed significant differences to the negative control DMSO (Fig. 4A). It is therefore unlikely for the FPG sensitive sites in the respective comet assay to be caused by oxidative stress. The DSM 12633 extract enhanced the DCF signal in a concentration-dependent manner to nearly 3-fold at the highest applied concentration (50 ␮M AOH content). Analogously, treatment of HT29 cells with AOH or AME resulted in an increase of the DCF signal, up to 2.5-fold at 50 ␮M (Fig. 4B). This increase was significant for AOH and AME concentrations of 10 ␮M and higher, for the single toxins and the extract dilutions with the same toxin content. The increase in DCF signal caused by the DSM 12633 extract therefore matched the effect of respective concentrations of AOH and AME as single compounds. However, in the case of AOH and AME the increase of the DCF signal was not reflected by an increased amount of FPG sensitive sites in the comet assay (Fig. 3B). The lack of additional FPG sensitive sites after incubation with AOH or AME is in accordance with previous studies (Fehr et al., 2009). Taken together, it can be excluded that AOH, AME or TA account for the pronounced increase in FPG sensitive sites in the comet assay by the complex extracts.

450

m

3.3. DCF assay

A relative fluorescence (T/C) [%]

and analysis were impossible (marked ### in Fig. 3B). To check, whether possible synergisms between AOH, AME and TA might be responsible for the effects observed with the extract, a reconstituted toxin mix, containing the same concentrations as the DSM 12633 extract dilutions (10 ␮M, 5 ␮M and 15 ␮M or 1 ␮M, 0.5 ␮M and 1.5 ␮M of AOH, AME and TA, respectively) was prepared and also tested in the comet assay. The tail intensities of these incubations did not exceed the negative control DMSO (data not shown), far from showing the strong effects produced by the complex fungal extract at these concentrations. It is therefore unlikely, that the pronounced increase in DNA strand breaks and the potent induction of FPG sensitive sites by the extracts are caused by their major toxins AOH, AME and TA or a combination thereof. The highest extract concentrations of both extracts, used for incubation correspond to 14.4 mg/mL for DSM 1102 and 19.7 mg/mL for DSM 12633, calculated as mg of fungal culture contained in 1 mL of incubation medium. Though these concentrations are rather similar, the DNA damage observed with DSM 12633 was more pronounced, indicating this strain to be more genotoxic under the applied culture conditions.

***

C. Schwarz et al. / Toxicology Letters 214 (2012) 46–52

relative fluorescence (T/C) [%]

50

AOH

AME

AOH in DSM 12633 extract

Fig. 4. DCF assays of A. alternata extracts DSM 1102 and DSM 12633 compared to their major toxins at the same concentrations. Results are given as mean values with standard deviations of four separate incubations of cells from different passages with the same two extracts used in the comet assays (see Table 2). Each incubation comprises 6 replicates on a 96 well plate. Significances indicated display the significance level as compared to the negative control DMSO, calculated by Student’s t-test (*p < 0.05; **p < 0.01; ***p < 0.001).

of the DSM 12633 extract showed only two small, unknown peaks at about 13 min of the run (Fig. 5B). Fraction #5, containing AME, showed the most potent DNA strand breaking effect. However, the tail intensities of about 16% greatly exceeded the effect to be expected with respective AME concentrations (∼3–4%) (Fehr et al., 2009). None of the other fractions exhibited any DNA strand breaking potential in the comet assay. Regarding the induction of FPG sensitive sites, only fractions #4 and #5 increased tail intensities after FPG treatment. The significant increase induced by fraction #5 is comparable to the unfractioned extract (diluted 1:2, in order not to exceed the range of quantification). An increase in FPG-sensitive sites was also indicated by fraction #4, but did not reach statistical significance. In order to examine the role of AME in the DNA damaging fraction #5, further sub-fractionation was performed by separately collecting the eluate containing the AME peak as well as the small peak immediately following AME (in the following labeled peak 5.2) at the detector outlet. The AME subfraction mediated tail intensities

C. Schwarz et al. / Toxicology Letters 214 (2012) 46–52

**

***

***

**

***

30

20

***

*** ***

25 ***

tail intensity [%]

no enzyme FPG treated

**

**

A

15 10 5

un

F8 fr ac (1 t. :2 )

F7

F6

F5

F4

F3

F2

V

F1

U

M SO D

fraction 5

fraction 4

B 30 25

AME

278 nm 340 nm

20

mAU

around 5% with no difference between FPG treated and untreated duplicates (Fig. 6). This is in accordance with previous observations (Fehr et al., 2009). The fraction containing peak 5.2, however, led to mean tail intensities of 13.5% with a significant increase to 25% mean tail intensity when treated with FPG, analogous to the effects by the entire fraction #5 (Fig. 5B). Peak 5.2 therefore comprises most of the DNA strand breaking properties observed with fraction #5, including the significant induction of FPG sensitive sites, while AME contributes only little to the overall effect. 4. Conclusion

0

15

peak 5.2

10 5 0 -5

51

12

13

14

15

16

17

18

min Fig. 5. (A) The comet assay used as endpoint for the toxicity-guided fractionation of A. alternata DSM 12633 extract. The results are mean values and standard deviations of three separate incubations with fractions, each time freshly prepared from 10 ␮L of the same complex extract, containing approximately 20 ␮M AME, 10 ␮M AOH and 30 ␮M TA. The unfractioned extract was further diluted 1:2, in order not to exceed the quantification range. Significances indicated display the significance level as compared to the respective control calculated by Student’s t-test (*p < 0.05; **p < 0.01; ***p < 0.001). (B) Detail of the DSM 12633 fractionation chromatogram, focusing on fraction #4 with its two unknown peaks, and fraction #5 with AME and peak 5.2.

The crude fungal extracts from two A. alternata strains (DSM 1102 and DSM 12633) exhibited significant DNA damaging properties to human colon carcinoma cells in the comet assay, far greater than to be expected according to their content of their major toxins AOH, AME and TA. Neither can the significant increase in additional damage to DNA bases, measured as FPG sensitive sites, be explained by these toxins, representing the major Alternaria toxins from the quantitative point of view. These DNA damaging effects were narrowed down to an extract fraction containing two unknown peaks but no AOH, AME or TA. Another DNA damaging extract fraction contained AME and a third, unknown peak. While most of the DNA strand breaking potential of this fraction was localized within this small unknown peak, AME did contribute only marginally to the overall effect of the fraction. The extracts increased the DCF signal in human colon carcinoma cells in a concentration dependent manner, analogously to their content of AOH or AME. A possible induction of FPG sensitive sites in the cellular DNA through oxidative stress is, nevertheless, unlikely. The same increase in DCF signal is caused by AOH or AME, yet no increase in FPG sensitive sites was observed with the single toxins. Moreover, the potential to induce FPG sensitive sites was localized within fractions/peaks other than those of the major toxins. Although these unknown peaks are yet to be identified, the data presented demonstrate that a considerable genotoxic potential is associated with A. alternata contamination, beyond that of the major toxins AOH, AME and TA. Funding

no enzyme treatment FPG treated

45

**

30

***

35

Conflict of interest statement

20

*

25

*

tail intensity [%]

* ***

40

The study was support by the State of Baden-Württemberg within the State Research Initiative “Food and Health”, research program “Mycotoxins”. The funding source had no involvement at all in study design, analysis, interpretation of data, writing of the report or decision to submit the paper.

15

None declared. References

10 5 0

DMSO

UV

peak AME peak 5.2

Fig. 6. The comet assay used as endpoint for the toxicity-guided subfractionation of fraction #5. The results are mean values and standard deviations of three separate incubations with three separate fractionations, that were prepared analogously as for the testing of all fractions (Fig. 5A), but collecting only the AME peak and peak 5.2 at the detector outlet. Significances indicated display the significance level as compared to the respective control calculated by Student’s t-test (*p < 0.05; **p < 0.01; ***p < 0.001).

Andersen, B., Frisvad, J.C., 2004. Natural occurrence of fungi and fungal metabolites in moldy tomatoes. Journal of Agricultural and Food Chemistry 52, 7507–7513. Aresta, A., Cioffi, N., Palmisano, F., Zambonin, C.G., 2003. Simultaneous determination of ochratoxin A and cyclopiazonic, mycophenolic, and tenuazonic acids in cornflakes by solid-phase microextraction coupled to high-performance liquid chromatography. Journal of Agricultural and Food Chemistry 51, 5232–5237. Barkai-Golan, R., 2001. Postharvest Diseases of Fruits and Vegetables: Development and Control. Elsevier, Amsterdam. Barkai-Golan, R., 2008. Alternaria mycotoxins. In: Barkai-Golan, R., Paster, N. (Eds.), Mycotoxins in Fruits and Vegetables, vol. 1. Academic Press, Elsevier, pp. 185–203. Brugger, E.M., Wagner, J., Schumacher, D.M., Koch, K., Podlech, J., Metzler, M., Lehmann, L., 2006. Mutagenicity of the mycotoxin alternariol in cultured mammalian cells. Toxicology Letters 164, 221–230.

52

C. Schwarz et al. / Toxicology Letters 214 (2012) 46–52

Burroughs, R., Seitz, L.M., Sauer, D.B., Mohr, H.E., 1976. Effect of substrate on metabolite production by alternaria-alternata. Applied and Environmental Microbiology 31, 685–690. Coste, F., Ober, M., Carell, T., Boiteux, S., Zelwer, C., Castaing, B., 2004. Structural basis for the recognition of the FapydG lesion (2,6-diamino-4-hydroxy-5formamidopyrimidine) by formamidopyrimidine-DNA glycosylase. Journal of Biological Chemistry 279, 44074–44083. Davis, V.M., Stack, M.E., 1994. Evaluation of alternariol and alternariol methyl ether for mutagenic activity in Salmonella typhimurium. Applied and Environment Microbiology 60, 3901–3902. Dong, Z.G., Liu, G.T., Dong, Z.M., Qian, Y.Z., An, Y.H., Miao, J.A., Zhen, Y.Z., 1987. Induction of mutagenesis and transformation by the extract of Alternaria alternata isolated from grains in Linxian, China. Carcinogenesis 8, 989–991. Fehr, M., Pahlke, G., Fritz, J., Christensen, M.O., Boege, F., Altemoller, M., Podlech, J., Marko, D., 2009. Alternariol acts as a topoisomerase poison, preferentially affecting the II alpha isoform. Molecular Nutrition and Food Research 53, 441–451. Griffin, G.F., Chu, F.S., 1983. Toxicity of the Alternaria metabolites alternariol, alternariol methyl ether, altenuene, and tenuazonic acid in the chicken embryo assay. Applied and Environment Microbiology 46, 1420–1422. Haggblom, P., Niehaus, W.G., 1986. Light effects on polyketide metabolism in alternaria-alternata. Experimental Mycology 10, 252–255. Hasan, H.A.H., 1995. Alternaria mycotoxins in black rot lesion of tomato fruit – conditions and regulation of their production. Mycopathologia 130, 171–177. Keston, A.S., Brandt, R., 1965. Fluorometric analysis of ultramicro quantities of hydrogen peroxide. Analytical Biochemistry 11, 1–5. Koch, K., Podlech, J., Pfeiffer, E., Metzler, M., 2005. Total synthesis of alternariol. Journal of Organic Chemistry 70, 3275–3276. Lehmann, L., Wagner, J., Metzler, M., 2006. Estrogenic and clastogenic potential of the mycotoxin alternariol in cultured mammalian cells. Food and Chemical Toxicology 44, 398–408. Li, F.Q., Yoshizawa, T., 2000. Alternaria mycotoxins in weathered wheat from China. Journal of Agricultural and Food Chemistry 48, 2920–2924. Liu, G.T., Qian, Y.Z., Zhang, P., Dong, W.H., Qi, Y.M., Guo, H.T., 1992. Etiologic role of alternaria-alternata in human esophageal cancer. Chinese Medical Journal Peking 105, 394–400.

Magan, N., Cayley, G.R., Lacey, J., 1984. Effect of water activity and temperature on mycotoxin production by alternaria-alternata in culture and on wheat-grain. Applied and Environment Microbiology 47, 1113–1117. Meronuck, R.A., Steele, J.A., Christen, Cm., Mirocha, C.J., 1972. Tenuazonic acid, a toxin produced by alternaria-alternata. Applied Microbiology 23, 613. Müller, M., 1991. Alternaria infestation of corn silage and hay. Zentralblatt fur Mikrobiologie 146, 481–488. Panigrahi, S., 1997. Alternaria toxins. In: D’Mello, J.P.F. (Ed.), Handbook of Plant and Fungal Toxicants. CRC Press, Elsevier, Boca Raton, NY, pp. 319–337. Pero, R.W., Posner, H., Blois, M., Harvan, D., Spalding, J.W., 1973. Toxicity of metabolites produced by the “Alternaria”. Environmental Health Perspectives 4, 87–94. Robiglio, A.L., Lopez, S.E., 1995. Mycotoxin production by alternaria-alternata strains isolated from red-delicious apples in argentina. International Journal of Food Microbiology 24, 413–417. Soderhall, K., Svensson, E., Unestam, T., 1978. Light inhibits production of alternariol and alternariol monomethyl ether in alternaria-alternata. Applied and Environment Microbiology 36, 655–657. Solfrizzo, M., De Girolamo, A., Vitti, C., Visconti, A., van den Bulk, R., 2004. Liquid chromatographic determination of Alternaria toxins in carrots. Journal of AOAC International 87, 101–106. Stack, M.E., Prival, M.J., 1986. Mutagenicity of the Alternaria metabolites altertoxins I, II, and III. Applied and Environment Microbiology 52, 718–722. Tice, R.R., Agurell, E., Anderson, D., Burlinson, B., Hartmann, A., Kobayashi, H., Miyamae, Y., Rojas, E., Ryu, J.C., Sasaki, Y.F., 2000. Single cell gel/comet assay: guidelines for in vitro and in vivo genetic toxicology testing. Environmental and Molecular Mutagenesis 35, 206–221. Tournas, V.H., Katsoudas, E., 2005. Mould and yeast flora in fresh berries, grapes and citrus fruits. International Journal of Food Microbiology 105, 11–17. Vinas, I., Palma, J., Garza, S., Sibilia, A., Sanchis, V., Visconti, A., 1994. Natural occurrence of aflatoxin and alternaria mycotoxins in oilseed rape from catalonia (Spain) – incidence of toxigenic strains. Mycopathologia 128, 175–179. Zhen, Y.Z., Xu, Y.M., Liu, G.T., Miao, J., Xing, Y.D., Zheng, Q.L., Ma, Y.F., Su, T., Wang, X.L., Ruan, L.R., 1991. Mutagenicity of Alternaria alternata and Penicillium cyclopium isolated from grains in an area of high incidence of oesophageal cancer – Linxian, China. IARC Scientific Publications, 253–257.