(1) Fundación de Investigaciones Científicas Teresa Benedicta de la Cruz. Luján, Buenos Aires

Transcription

1 Manuscript Click here to download Manuscript: Mycoflor Soybean FINAL RV.docx Click here to view linked References Mycoflora and Mycotoxins Contamination of Soybean RR Harvested In the Pampean Region, Argentina Carolina E. Garrido,, Héctor H.L. González,, María Paula Salas,, Silvia L. Resnik,*, Ana M.Pacin * () Fundación de Investigaciones Científicas Teresa Benedicta de la Cruz. Luján, Buenos Aires () Agencia Nacional de Promoción Científica y Tecnológica () Consejo Nacional de Investigaciones Científicas y Técnicas () Facultad de Ingeniería, Universidad de Buenos Aires () Departamentos de Industrias y de Química Orgánica, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires () Comisión de Investigaciones Científicas de la Provincia de Buenos Aires *Address for correspondence: Silvia Resnik Departamento de Química Orgánica Facultad de Ciencias Exactas y Naturales Ciudad Universitaria, CEHA, Buenos Aires, Argentina Fax: 0 sresnik000@yahoo.com.ar Ana María Pacin Fundación de Investigaciones Científicas Teresa Benedicta de la Cruz Dorronzoro (B00) Luján, Buenos Aires, Argentina Fax: fundacion@ictbdelacruz.org.ar Running title: Mycoflora and Mycotoxins Contamination of Soybean RR

2 Abstract A total of freshly harvested soybean seed samples (Roundup Ready (transgenic) soybean cultivars) from the 0/0 crop season were collected from locations in the Northern Pampean Region II, Argentina. These samples were analyzed for internal mycoflora, toxin production of isolated fungi, and for a range of mycotoxins. Mycotoxin analysis of aflatoxins (AFs), zearalenone (ZEA), fumonisins (FBs) and ochratoxin A (OTA) was done by HPLC-FLD, alternariol and alternariol monomethyl ether with HPLC-UV, trichothecenes (deoxynivalenol, nivalenol, T- toxin, HT- toxin, fusarenon X, - acetyldeoxynivalenol and -acetyldeoxynivalenol were analyzed by GC-ECD. Fungal colonization was more frequently found for samples from América, Saladillo and Trenque Lauquen than for samples from General Villegas and Trenel; a total of 0 fungal isolates were obtained from the soybean seeds. The most commonly identified fungal genera were Alternaria, Sclerotinia, Chaetomium, Cladosporium, Aspergillus, Penicillium, Phomopsis and Fusarium. Alternaria alternata, A.tenuissima, Aspergillus flavus, Penicillium citrinum, Fusarium verticillioides and F.semitectum were the predominant toxigenic fungal species. Mycotoxin production was confirmed for several isolates of toxigenic species, including Aspergillus flavus, A. parasiticus, Alternaria alternata, A.tenuissima, Fusarium graminearum, F semitectum and F. verticillioides. In particular, the percentage of mycotoxigenic Alternaria alternata (0 %), A.tenuissima ( %) and aflatoxigenic strains of A. flavus (%) were remarkably high. Although none of the mycotoxins AFs, ZEA, FBs, trichothecenes and OTA were directly detected in samples of soybean seeds, the frequent presence of toxigenic fungal species indicates the risk of multiple mycotoxin contamination. Key words: Soybean, mycoflora, mycotoxins, Alternaria, Aspergillus, Fusarium, Penicillium.

3 Introduction Argentina is the third largest producer of soybean seeds (Glycine max (L.) Merrill), and the most important exporting country of soybean oil and protein meal worldwide (SAGPYA 0). Based on historical yield trends, the Argentinean soybean production is estimated at million metric tons for 0, of which approximately million tons per year are exported, which implies an income of about thousand million dollars (USDA 0). The consumption of products that include soybean as one of its components has increased in the oriental and western countries, as human or animal supplies. Food quality and safety are concepts that start from the field up to the consumer, and are based on the consumption of innocuous and good quality food that assure health and well-being. Fungal proliferation reduces the soybean quality and yield (Sinclair, Wrather et al. ), and it was shown that a diverse group of saprophytic and parasitic fungi can colonize and infect soybean seeds before the harvest (Roy et al. 00, Villarroel et al. 00). In decreasing order, Alternaria spp., Fusarium spp., Aspergillus spp., and Penicillium spp., have been found in soybean seeds in Argentina during 00 to 00 harvests (Zelaya 00). However, there are little research information in relation with mycoflora, toxigenic capacity and natural occurrence in our country (Pacin 00). The objectives of this work were: ) to identify the internal mycoflora of soybean seeds, freshly harvested from the 0/ crop season, collected from five localities of at the Buenos Aires and La Pampa provinces, ) to compare the fungal incidence on soybeans seeds between regions, with special emphasis on mycotoxigenic species, ) to determine mycotoxin levels in these soybean seeds, and ) to determine the toxigenic potential of some fungal species isolated from soybean seeds. Materials and methods Soybean samples A total of freshly harvested Roundup Ready soybean seeds samples ( kg) were obtained during the 0/0, crop season from field trials carried out to study the yield performance of commercial cultivars belonging to Don Mario Seeds Company. Harvest sampling was performed in zig-zag, without considering the edges. Soils from experimental sites did not show any physical or nutritional constraints. The experimental fields of Buenos Aires province, were América (º S, º O), General Villegas (º S, º0 O), Saladillo (º S, º O), Trenque Lauquen (º S, º O), and Trenel (º 00 S, º0 00 O) in La Pampa province, all in the agroecological Northern Pampean II region V. Randomly selected subsamples of 0 g each were submitted into paper bags to mycological analysis and stored at ºC for not more than days. The rest of the samples were stored at -ºC until mycotoxins analysis.

4 Isolation and identification of fungi For isolation of the internal mycoflora in whole seeds, subsamples of soybean from each sample were surface-disinfected in a commercial % aqueous solution of sodium hypochlorite for minute, rinsed twice with sterile distilled water and dried in a sterile laminar flow cabinet. One hundred seeds per sample were placed, 0 seeds per plate, on Yeast Extract-Glucose-Chloramphenicol Agar, Biokar Diagnostic BK00 (González et al. ). The plates were incubated in the dark at C for - days and the resulting fungal colonies subcultured onto Potato Dextrose Agar, Biokar Diagnostic BK0, and identified.where several different fungi were isolated from a single seed, all of them were recorded. Keys for fungal identification were those recommended by Nelson et al. (), Pitt and Hocking (00), Samson et al. (00), Simmons (00) and Roy et al. (00). The isolation frequency (Fr) and the relative density (RD) of genera and species were calculated according to González et al. (00): ns ni Fr(%) ( ) 0 RD(%) ( ) 0, N Ni where, ns number of samples which a certain fungal species was found; N total number of samples; ni number of isolates of a genus or species; Ni total number of fungal isolates. Some of the cultures have been deposited in the BAFC Culture Collection from the Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires in the Biodiversity and Experimental Biology Department. Toxigenic capacity In order to analyze the toxigenic capacity, one isolate of possible toxigenic species of each soybean sample were tested using the procedure described below: Slants preparation: Four isolates of F. semitectum, seven of F. verticillioides and one of F. graminearum were transferred to carnation leaf agar slants and incubated under h fluorescent light, h darkness for days at C to stimulate conidial formation (Megalla et al. ). Fourteen isolates of A. flavus, one of A. parasiticus and two of A. niger were grown on potato dextrose agar slants and incubated at C for days. Nine isolates of A. alternata and of A. tenuissima were grown on potato dextrose agar slants and incubated at C for days. Culture media: To prepare the rice media in 0 ml Erlenmeyers flask, distilled water was added up to 0% to g of polish rice, and then autoclaved at C for 0 minutes. Autoclaved flasks were shaken by hand to loosen the grains, and each flask was inoculated with ml of conidial suspension before incubation. ml of YES medium (% yeast extract, % sucrose) was prepared using ml Erlenmeyer. Conidial suspension preparation: the suspensions of all fungus were prepared by adding (.0 ml) sterile distilled water to each slant and gently scraping the agar surface with a wire loop to give a turbid suspension (x conidia/ml)

5 Inoculation and Incubation procedure: In order to study the production of deoxynivalenol (DON), nivalenol (NIV), fusarenon X (FUS-X), -acetyldeoxynivalenol (-AcDON), -acetyldeoxynivalenol (-AcDON), T- toxin (T-), HT toxin (HT-), zearalenone (ZEA), rice media was inoculated with 0 µl conidial suspension of F. semitectum and F. graminearum and were incubated at C for weeks. For testing aflatoxin B (AFB ), B (AFB ), G (AFG ) and G (AFG ), rice media was inoculated with 0 µl the conidial suspension of A. flavus and A. parasiticus and incubated one week, at C. For ochratoxin A (OTA) production, ml of the conidial suspension of A. niger were inoculated in ml of YES following the methodology described by Magnoli et al.(00), where the flasks were incubated stationary at 0 C for days in darkness. To test for the production of fumonisins B (FB ), B (FB ) and B (FB ), 00 µl of the conidial suspension of F. verticilliodes were inoculated in the rice media and incubated at C for weeks. To test for the production of alternariol (AOH) and alternariol monomethyl ether (AME), 00 µl of the conidial suspension of Alternaria spp. were inoculated in the rice media and incubated at C for weeks All toxigenic capacity studies were performed once, for each fungus. After the corresponding incubation, extraction of mycotoxins of each fungal inoculated rice or YES medium was carried out previous to High Performance Liquid Chromatographic or Gas Chromatography analysis. Materials and instruments Reagent, solvents and standards Organic solvents were HPLC grade. Acetonitrile (ACN), toluene, methanol (MeOH), ethyl acetate were purchased from Sintorgan (Argentina); hexane, sodium bicarbonate (NaHCO ), sodium azide from J.T Baker (Mexico) ;boron trifluoride from J.T Baker (USA); trifluoroacetic anhydride (TFA) from Tedia Company Inc. (USA); acetic acid, formic acid, sodium hydroxide (NaOH), potassium chloride, monoacid sodium phosphate, diacid sodium phosphate, sodium chloride, from Merck Química (Argentina), di-sodium tetraborate decahydrate,-mercaptoethanol, orthophosphoric acid, o-phthalaldehyde from Merck (Germany); -(N,N-Dimethylamine) pyridine (DMAP) and heptafluorobutyric acid anhydride (HFBA) from Sigma-Aldrich (Switzerland). Fumonisins, trichothecenes, OTA and ZEA standards were from Biopure (Austria); AFs, AME, AOH and standards were from Sigma Chemical Company (USA).The concentration of ther stock standard solutions from Biopure were: FB 0. µg ml -, FB 0. µg ml -, and FB. µg ml - ; NIV, DON and HT-,. µg ml -, -AcDON and. µg ml - ; FUS-X and T-. µg ml - ; OTA. µg ml -. ZEA from Biopure was prepared according to the method described by the Method : (AOAC, 0) for ZEA. AFs standars were prepared according AOAC Method 0. (AOAC 0) and mg of AME and AOH were diluted with ml of MeOH. HPLC quality water was prepared with a Waters Milli-Q system (Waters Associated, Milford, MA, USA). The phosphate buffer saline (PBS) was prepared with a mixture of 0. g of monoacid sodium phosphate,. g of diacid sodium phosphate,.0 g of sodium chloride, 0.0g of potassium chloride and 0. g of and then diluted to a L with bi distilled water and adjusted the ph of the solution to.. Instruments

6 The High Performance Liquid Chromatography (HPLC) equipment used (Agilent 00 series, USA), included a degasser (GA), an auto sampler (GA), a fluorescence detector (GA), ultraviolet detector (GA), quaternary pump (GA) and a temperature controller (GA). The Gas Chromatography (GC) equipment used (Agilent Technology 0A) include an autosampler (Agilent ), nitrogen as carrier and auxiliary gas and electron capture detector of Ni (ECD). Mycotoxins analysis Extraction Extraction of mycotoxins has different procedures depending on the substrate (soybean, inoculated rice and cultures of A.niger in YES medium). Aflatoxins and zearalenone For the AFs and ZEA extraction, of fungal rice culture were extracted with ml of ACN blending for min at high speed with a blender (Osterizer, USA). Approximately ml of the supernatant were passed through a Solid phase clean-up column PuriTox TC-M (Trilogy, USA). Subsequent, the filtrate was divided in aliquots, of ml (0.g of the sample), one for AFs analysis and one for ZEA analysis, and of ml (0.g of the sample) for Thin Layer Chromatography confirmation. The Method used for AFs is based in the Method 0. of AOAC (AOAC 0) and for ZEA in the Method. of AOAC (AOAC 0). Subsamples of soybeans ( g) were extracted with 0 ml of acetonotrile:water (:, v/v), blending for min at high speed with a blender. Five ml of the supernatant were applied to anextraction column PuriTox TC-M. The filtrate was divides in aliquot, of ml (0. g) and of ml (0. g). The aliquots were evaporated under vacuum in a water bath at 0 C until dryness. Fumonisins For the fumonisins analysis, fungal rice culture and soybeans were extracted as follow. Twenty- five g of fungal rice culture were extracted with 0 ml of MeOH and 0 ml of MeOH:water (:, v/v) by blending for minutes at high speed. The extract was filtered, adjusted to ph= with NaOH M and centrifuged during minutes (000 rpm).for the cleanup step, quaternary amine solid phase extraction column Strata SAX (Phenomenex, USA) was used, previously conditioned by the successive passage of ml of MeOH and ml of MeOH:water (:, v/v). Then, 0 ml of the filtered were applied to the column, washed with ml MeOH:water (:, v/v) and ml of MeOH. Fifty g of soybeans samples were extracted with 0 ml of MeOH:water (:, v/v) by blending for min at high speed. The extract was filtered through Whatman N filter (USA), adjusted to ph= (NaOH M) and centrifuged during minutes (000 rpm).for the cleanup step, quaternary amine solid phase extraction column Strata SAX (Phenomenex, USA) was used, previously conditioned by the successive passage of ml of MeOH and ml of MeOH:water (:, v/v). Then, an aliquot

7 ml of the filtered were applied to the column, washed with ml MeOH: water (:, v/v) and ml of MeOH, according to the method. (AOAC 0) The fumonisins were eluted with ml of acetic acid:meoh (:, v/v) at speed of ml/min and the elutes were collected and evaporated to dryness at 0 C under a stream of nitrogen. Ochratoxin A A.niger in YES medium and soybeans samples, were analyzed to determine the presence of OTA. For OTA analysis in YES medium, the complete volume of the medium was blended for min and the solid particles were removed by centrifugation. To 0 g of the solid particles, ml of ACN was added, the mixture blended for min and filtered. An aliquot of ml was taken and ml of buffer PBS at ph. were added and centrifuged for min (000 rpm). Twenty-five g of soybeans sample were extracted with 0 ml of ACN:water (0:0, v/v) by blending for minutes at high speed. The extract was filtered through Whatman N filter (USA). An aliquot of ml was taken and ml of buffer PBS at ph. were added and centrifuged during min (000 rpm). For both, YES medium and soybeans samples extracts, an immunoaffinity column OCHRAPREP (R-Biopharm Rhône LTD, Scotland), at a flow rate of - ml/min was used. The column was washed with 0 ml of buffer PBS ph. at a flow rate of ml/min. OTA was eluted with. ml of acetic acid: MeOH (:, v/v) by gravity, backflushing times before evaporate until dryness at 0 Cunder a stream of nitrogen (Drunday and Pacin 0). Trichothecenes For the trichothecenes analysis the method used is based in the method described by Castillo et al. (00). The fungal rice culture and soybeans were extracted as follow. Fungal rice culture and soybean samples were analyzed to determine the presence of DON, FUS-X, NIV, T-, HT-, -AcDON and -AcDON. Twenty-five g of inoculated rice were extracted with ml of ACN, blending for min at high speed. Approximately ml of the supernatant were passed through a solid phase clean-up extraction column PuriTox TC-M 0 (Trilogy, USA). Twenty-five g of soybeans subsample were extracted with 0 ml of ACN: H O (:, v/v), by blending for min at high speed. Approximately ml of the supernatant were passed through a solid phase clean-up extraction column MultiSep (Romer Labs, Austria). Subsequent, the filtrate obtained was applied to other extraction columns PuriTox TC-M 0, for a more clear purification with less interference. Finally, an aliquot of the filtrate (equivalent to 0. g) was evaporated under vacuum in a water bath at 0 C until dryness. Alternariol and Alternariol monomethyl ether For the AME and AOH extraction, the rice culture media was extracted with 0 ml of ethyl acetate:formic acid (:, v/v) by shaking at 00 rpm for min in darkness. Then the solid particles were separated from the extract centrifuging at 00 rpm during minutes. The solids where extracted two more times with ml of ethyl acetate:formic acid (:, v/v), following the

8 same steps of the first extraction. The extract were then mixed and evaporated under vacuum in a water bath at 0ºC until dryness Chromatographic analysis Detection and quantification limits (LOD and LOQ, respectively) where calculated as a relation of signal to noise or signal no noise (LOD and LOQ, respectively). The range for the calibration curves were for all the FBs between 0 to 00 µg Kg -, trichothecenes to 0 µg Kg -,OTA 0. to. µg Kg-, ZEA 0 to 000 µg Kg -, AFs from 0. to µg Kg -, AME from. to 000 µg Kg -, and AOH from 0 to 0000 µg Kg -. All the R were higher than 0.. Aflatoxins The evaporated extract ( ml) was reconstituted in 00 µl hexane and was derivatized with 0 µlof TFA; shaken during 0 seconds and after minutes neutralized with 0 µl of ACN:water (:, v/v), where the water phase was separated. For HPLC separation, a Microsorb-MV 0 C reversed phase column µm 0 mm *. mm was used (Varian, USA). The mobile phase was water:acn:meoh (0::, v/v/v), the flow rate was ml min - and the injection volume was 0 µl. Fluorescence excitation and emission wavelengths were set at 0 nm and 0 nm. LOD and LOQ, for AFs were: 0. µg Kg - and 0. µg Kg - for AFB andafg ;0.µg Kg - and 0. µg Kg - for AFB, and AFG. Zearalenone For ZEA analysis, the evaporated extract of 0. g was suspended in ml of MeOH:water (:, v/v). For HPLC separation, a Thermo-hypersil BDS C reversed phased column µm 0 mm *. mm was used (Thermo-Hypersil, Germany). The mobile phase was water:meoh:acn (0::, v/v/v) at a flow rate of ml min - and the injection volume was µl. Fluorescence excitation and emission wavelengths were set at nm and 0 nm, respectively. LOD and LOQ for ZEA were µg Kg - and µg Kg -. Fumonisins For FBs determination the residue was suspended in ml of ACN:water (:, v/v) and was derivatized online with a solution o- phthalaldehyde (0 mg o-phthalaldehyde in ml of methanol, diluted with ml of disodium tetraborate solution 0. M, and 0 µl of -mercaptoethanol). Within min of reaction time, the derivatized solution was injected into the HPLC system. For the HPLC separation, a Microsorb-MV 0 C reversed phase column µm 0 mm *. mm was used (Varian, USA). The mobile phase was MeOH:orthophosphoric acid (:, v/v) at ph=., the flow rate was ml min - and the injection volume was µl. Fluorescence excitation and emission wavelengths were set at 0 nm and 0 nm, respectively.lod and LOQ for FB were and µg Kg - ; for FB and µg Kg - and for FB and 0 µg Kg -, respectively. Ochratoxin A For OTA analysis the residue was suspended in 0 µl of mobile phase, ACN:water:acetic acid (.:.:, v/v/v). For the HPLC separation, a Hypersil BDS C reversed phased column µm mm * mm was used (Hypersil, UK). The mobile phase was set at a flow rate of ml min -, and the injection volume was 0 µl. Fluorescence excitation and emission

9 wavelengths were set at nm and 0 nm at a temperature of 0 C (Drunday and Pacin 0). LOD and LOQ for OTA were 0.0 and 0.0 µg Kg -, respectively Alternariol and alternariol monomethyl ether For AOH and AME analysis the residue was suspended in 00 µl of mobile phase, MeOH: phosphoric acid % (:, v/v). For HPLC separation, the analytical column used was a Thermo Hypersyl C reversed phased column μm, 0 mm. mm (Thermo Electrón, UK) with a guard column Hypersyl C μm mm (Thermo Scientific, UK). Flow rate from the intial time (t) to t = min was 0. ml/min, from t =. to t = min was ml min -, and from t = to t = 0 min the flow was 0. ml min - (Broggi et al. 0). Ultraviolet detection was made at nm. Detection was performed at. nm; column temperature was 0 C and sample temperature C. LOD and LOQ for AOH were and µg Kg -, and LOD and LOQ for AME were and µg Kg -, respectively. Trichothecenes For trichothecenes determination, the evaporated extract was suspended in 00 µl of ethyl acetate:meoh (:, v/v) and shaken for s using a vortex and 0 µl of this solution were transferred to a derivatization tube and evaporated to dryness under a nitrogen stream at temperature lower than 0 C. Then, 0 µl of toluene:acn (0:0, v/v) with mg ml - of the DMAP catalyst were added to the residue and shaken with vortex for s, followed by the addition of 0 µl of the HFBA, shaken again for s and put in a bath sand at 0- C during 0 min. After this time,. ml of % NaHCO solution and 00 µl of toluene were added, shaken for 0 s with vortex and let it cool at room temperature before centrifuged at 000 rpm for min. An aliquot of 00 µl from the toluene phase was separated and put into an insert to be processed. For GC separation the analytical column used was a HP- column 0 m 0. mm 0. µm (Agilent J & GC columns, USA). LOD and LOQ were and µg Kg - for DON, and µg Kg - for FUS-X, and µg Kg - for NIV, and µg Kg - for T-, and µg Kg - for HT-, and and µg Kg - for A-DON, respectively. Statistical analysis In order to compare the fungal contamination between locations, analysis of variances were performed (Conover ). Asymptotic tests for equality of proportions were used to compare relative densities (RD) of fungal genera and species between localities (Devore ). The Fisher exact test was applied to analyze possible differences in the isolation frequencies (Fr) of fungal genera and species between localities. A difference was considered significant if the corresponding p-value was less than 0.0 and highly significant when it was less than 0.0. The statistical analysis was done using Statistix version.0 (000), running under Windows. Results and discussion Fungi associated with soybean seeds

10 A total of 0 fungal isolates were obtained from Roundup Ready soybean cultivars seeds harvested in 0; more than per cent of fungi isolated were Deuteromycota and Ascomycota, and the remaining species were Zygomycota. This variety of fungal species is narrower than the one found for soybean seeds, pods and flowers in the USA (Roy et al. 00). The Fr and RD of fungal genus and species recovered from soybean seeds are shown in Table. In the present study, species belonging to Alternaria, Sclerotinia, Chaetomium, Cladosporium, Aspergillus, Penicillium, Phomopsis, and Fusarium genera were most commonly isolated from RR soybean seeds. These genera were also identified in previous investigations in Argentina (Boca et al. 00, Broggi et al. 00, Pereyra et al. 0), Ecuador (Pacin et al. 00) and USA (Roy et al. 00, Villarroel et al. 00).The high incidence level of Alternaria species in the soybean seeds is consistent with previous reports carried out on freshly harvested RR soybean samples collected from experimental fields (Boca et al. 00, Zelaya 00), where A. alternata was the most frequently isolated fungus (Fr:.% and.%, respectively). Broggi et al. (00) reported of a Fr:.% for A. Alternata in a field of Entre Ríos Province. These similar results were reported for other provinces in Argentina during two harvests seasons (00-0 and 00-0) (Barros et al. 0). With regard to fungal species from the genera Aspergillus and Penicillium, relevant in storage period, showed low Fr and RD levels, with the exception of P. citrinum (Fr:.%) and A. flavus (Fr. %). Box-plots of the distribution of the internal fungal contamination by locality, calculated using the percentage of infested seeds, are shown in Fig.. Considering their medians, the samples most infested by fungi came from América (0.%) and from Saladillo (.%), and those from Trenque Lauquen, General Villegas and Trenel were the less contaminated (.,.0 and.0%, respectively). It can be seen that the distributions are in general asymmetric, the variances are not equal (América:.; Saladillo:.; Trenque Lauquen:.; Trenel:. and Gral. Villegas:.) and several outliers are present ( for América, and one each for Saladillo, Gral. Villegas and Trenel, respectively). Applying a logarithmic transformation to the data (ln([variable + ]) in order to avoid outliers, the variances continue being different (América: 0.; Saladillo: 0.; Trenque Lauquen: 0.; Trenel: 0. and Gral. Villegas: 0.) and some distributions are still asymmetric, so to compare internal fungal contamination between locations, the median test (non parametric) was applied and significant (with p < 0.0) and highly significant differences (with p < 0.0) were found between localities (Table ). Another point was to calculate the Fr and RD of the non-toxigenic and the potential toxigenic fungal species isolated; the prevalent non-toxigenic fungal species isolated from soybean seeds in Buenos Aires and La Pampa Provinces were Chaetomium spinosum, Sclerotinia sclerotiorum, Cladosporium cladosporioides and Phomopsis sojae. Other species recorded with minor incidence were, Acremonium strictum, arthrinium phaeospermum, Aspergillus candidus, Epiccocum nigrum, Eurotium chevalieri, Mucor racemosus, Penicillium chrysogenum, Penicillium corylophilum, Phoma spp., Rhizoctonia solani, Rhizopus stolonifer and Ulocladium spp. Figure shows the Fr and RD of the fungi with potential toxicological interest isolated in the different locations. Fusarium species were isolated from most of the samples and the isolation frequencies were variable between localities. In our study, F. verticillioides, F. semitectum and F. graminearum, were the most frequently Fusarium species found. These findings are different from previous reports carried out on RR soybean seeds collected in Buenos Aires Province in (Boca et al. 00), where F. graminearum was the most frequently isolated Fusarium species (Fr:,%). In a study on soybean seeds in Entre Ríos

11 Province in 00 (Broggi et al. 00), the predominant Fusarium species were F. semitectum (Fr:.0%) and F. graminearum (Fr:.%). Zelaya (00) found that F. semitectum was the most prevalent Fusarium species and identified different species including F. equiseti, F. poae, F. sambucinum, F. subglutinans and F. lateritium. In soybean seeds from Ecuador, F. verticillioides and F. semitectum were the most frequently species isolated (Pacin et al. 00). Alternaria alternata, A. tenuissima, and Penicillium citrinum were isolated in all the localities studied. Highly significant differences (p<0.00) were observed for RD when comparing Trenel and Saladillo with the other localities for Aspergillus ochraceus and Fusarium. Mycotoxigenic capacity of fungal isolates Eight out of A. flavus isolates were able to produce AFB and AFB, the isolate CIM N also co-produced AFG (Table ) and none of the strains, inoculated in rice, were AG producers. A. parasiticus strain studied shown that AFB, AFB and AFG production is greater than in all the A. flavus tested isolates. In this work the percentage of aflatoxigenic strains of A. flavus was remarkable high (%) in comparison with Vaamonde et al. (00); these authors showed that among A. flavus strains isolated from soybean seeds, collected in Buenos Aires Province, only % ( strains) were aflatoxigenic. On the other hand, in coincidence with the present study, they also found that the 0% of A. parasiticus strains isolated from soybean seeds were AFB and AFG producers. Four of seven F. verticillioides strains analyzed (Table ) were FB producers and the CIM N 0 strain, which also produces FB and FB, is a huge fumonisins producer; two F. semitectum isolates synthesized HT- and the only F. graminearum strain studied, produced ZEA. The only two A. niger strains tested for OTA production (CIM and CIM ) did not produce this toxin. More than % of the Alternaria strains, inoculated in rice were able to produce AME and/or AOH (Table ), in a wide range of amounts, but Alternaria alternata appears to produce more than A. tenuissima of both toxins. AME has been produced in higher quantities than AOH for both species of Alternaria. Torres et al. () showed the ability to produce both toxins by strains of Alternaria alternata, from maize seeds, cultivated in rice and found a similar percentage, % strains produced AOH and % produced AME; but the levels of AOH were 00 to 00 µg Kg - and the AME production by the strains, ranged between 00 and 00 µg Kg -, which is less than the results in this work. They tested four strains of A. tenuissima which had no toxigenic capacity. Patriarca et al. (00) showed that % of Alternaria strains isolated from Argentinean wheat had a capacity to produce AOH (range µg Kg - ) and % to produce AME (range µg Kg - ), a relatively higher results than the present work on strains isolated from soybean. Other authors (Solfrizzo et al. 00) testing Alternaria alternata isolated from carrot, demonstrated the huge capacity to produce AOH (000 to 000 µg Kg - ) and AME (000 to µg Kg - ) on rice, higher than the presented results. Mycotoxins associated with soybean seeds

12 Aflatoxins, fumonisins, ochratoxin A, trichothecenes and zearalenone, were not detected in any freshly harvested RR soybean seeds sample, in spite of several toxigenic fungi were isolated. It is well known that fungal growth and the ability to produce mycotoxins is a consequence of not only environmental conditions, such as, climatic factors (Miedaner et al. 00, Munkvold 00), but also a complex interaction of several factors such as aggressiveness of fungi, host susceptibility (Barath et al., Castillo et al. 00, Schollenberger et al. 00) as well as edaphic and agronomic factors (Edwards 00). It was identified a higher occurrence of contaminants toxigenic strains of soybeans. Although AFs, FBs, OTA, trichothecenes and ZEA were not detected in any freshly harvested soybean seeds, some of the species would be able to produce mycotoxins under propitious environmental conditions. It has been showed (Oviedo et al. 0) that irradiated soybeans was a good substrate for AOH and AME production and the concentration range of both toxins varied considerably depending on water activity and temperature interactions. Oviedo et al. (0) found that 0% of 0 soybean samples from the North of Argentina were contaminated by AME and AOH; among them, % were contaminated with AOH, % with AME and of the positive samples showed co-occurrence of both mycotoxins analysed. AOH was detected in concentrations ranging from to µg Kg -, whereas AME was found in concentrations ranging from to µg Kg -. All together these findings indicate that, although several toxigenic fungi are found associated with soybean and has the potential to produce considerable amounts of mycotoxins future research is needed to identify the importance of natural occurrence of mycotoxins in soybeans taking into account other cultivated fields and different harvest years. Acknowledgements The authors acknowledge to Agencia Nacional de Promoción Científica y Tecnológica, Fundación de Investigaciones Científicas Teresa Benedicta de la Cruz, Consejo Nacional de Investigaciones Científicas y Técnicas and Comisión de Investigaciones Científicas de la Provincia de Buenos Aires and Universidad de Buenos Airesfor the financial support. The authors acknowledge with great thanks the technical assistance of Mrs. Gabriela Cano, Mrs. Vanesa Dundray and Mrs. Daniela Taglieri. Conflicts of interest: None References AOAC (0) Official Methods of Analysis th Edition. Ed: Latimer, G, J., AOAC International, Natural toxins, Chapter. Barath A, Sawinsky J, Halasz A, Borbiro G () Substrate influence on mycotoxin production of Fusarium species and their analytical detection. Cereal Res Commun : Barros G, Oviedo M, Ramírez M, Chulze S (0) Safety Aspect in Soybean Food and Feed Chains: Fungal and Mycotoxins Contamination. Soybean - Biochemistry, Chemistry and Physiology INTECH :-0 [ pp]

13 Boca RT, Pacin AM, González HHL, Resnik SL, de Souza JC (00) Soja y micotoxinas: Flora fúngica, variedades y prácticas agronómicas. Aceites y Grasas : Broggi LE, González HHL, Resnik SL, Pacin AM (00) Alternaria alternata prevalence in cereals grains and soybean seeds from Entre Ríos, Argentina. Rev Iberoam Micol :- Broggi LE, Reynoso C, Resnik SL, Martinez F, Drunday V., Romero Bernal A (0) Occurrence of alternariol and alternariol monomethyl ether in beverages from the Entre Rios Province market, Argentina. Mycotoxin Res : Castillo M, Samar MM, Molto G, Resnik S, Pacin A. (00) Trichothecenes and zearalenone production by Fusarium species isolated from Argentinean black beans. Mycotoxin Res : Conover WJ () Practical Nonparametric Statistics. New York: John Wiley & Sons Devore JL () Probability and Statistics for Engineering and the Sciences. Monterrey, California: Brooks-Cole Publishing Company Drunday V, Pacin A (0). Occurrence of ochratoxin a in coffee beans, ground roasted coffee and soluble coffee and method validation. Food Control 0:- Edwards SG (00) Influence of agricultural practices on Fusarium infection of cereals and subsequent contamination of grain by trichothecene mycotoxins. Toxicol Letters : González HHL, Resnik SL, Boca RT, Marasas WFO()Mycoflora of Argentinean corn harvested in the main production area in. Mycopathologia 0:- González HHL, Moltó GA, Pacin A, Resnik SL, Zelaya MJ, Masana M, Martínez EJ(00)Trichothecenes and mycoflora in wheat harvested in localities in Buenos Aires Province, Argentina. Mycopathologia :- Magnoli C, Violante M, Combina M, Palacio G, Dalcero A(00) Mycoflora and ochratoxin-producing strains of Aspergillussection Nigri in wine grapes in Argentina. Lett Appl Microbiol :- Megalla, SE, Bennett, GA, Ellis JJ, Shotwell OL() Production ofdeoxynivelenol and zearalenone by isoltes of Fusarium graminearum. J Food Protect 0:- Miedaner T, Schilling AG, Geiger HH. (00) Molecular genetic diversity and variation for aggressiveness in populations of Fusarium graminearum and Fusarium culmorum sampled from wheat fields in different countries. J Phytopathol : Munkvold GP. (00) Cultural and genetic approaches to managing mycotoxins in maize. Annu Rev Phytopathol : Nelson PE, Toussoun TA, Marasas WFO () Fusarium species: an illustrated manual for identification. Pennsylvania: The Pennsylvania State University Park Pacin AM (00) Existe un diagnóstico sobre micotoxinas en soja en argentina? pp- Accessed on November 0 Oviedo MS, Barros GG, Chulze SN, Ramirez ML (0) Natural occurrence of alternariol monomethyl ether in soya beans. Mycotoxin Res :-

14 Oviedo MS, Ramirez ML, Barros GG, Chulze SN (0) Influence of water activity and temperature on growth and mycotoxin production by Alternaria alternata on irradiated soya beans.int J Food Microbiol. : Pacin AM, González HHL, Etcheverry M, Resnik SL, Vivas, L, Espin S (00) Fungi associated with food and feed commodities from Ecuador. Mycopathologia :- Patriarca A, Azcarate MP, Terminiello L, Fernández Pinto V (00) Mycotoxin production by Alternaria strains isolated from Argentinean wheat. Int J Food Microbiol :- Pereyra CM, Cavaglieri LR, Chiacchiera S M, Dalcero A M(0) Fungi and Mycotoxins in Feed Intended for Sows at Different Reproductive Stages in Argentina. Vet Med Int :- Pitt JI, Hocking AD (00) Fungi and Food Spoilage. London-New York: Blackie Academic & Professional. Roy KW, Baird RE, Abney TS (00) A review of soybean (Glycine max) seed, pod, and flower mycofloras in North America,with methods and a key for identification of selected fungi. Mycopathologia 0: SAGPYA Secretaría de Agricultura, Ganadería, Pesca y Alimentos de la Nación. Argentina (0) ( November 0 Samson RA, Hoekstra ES, Frisvad JC (00)Introduction to food and airborne fungi. Centraalbureau voor Schimmelcultures, Baarn, Delft. Schollenberger M, Müller HM, Rüfle M, Suchy S, Plank S, Drochner W. (00) Natural occurrence of Fusarium toxins in grains and feedstuffs of plant origin from Germany. Mycopathologia : - Simmons EG (00)Alternaria. An Identification Manual. Utrecht: CBS Biodiversity Series Sinclair BJ () Reducing losses from plant disease. Proc. World Soybean Res. Conf. V. Chiang Mai, Thailand. February : Solfrizzo M, De Girolamo A, Vitti C, Tylkowska K, Grabarkiewicz-Szcz ęsna J, Szopińska D, Dorna H (00) Toxigenic profile of Alternaria alternata and Alternaria radicina occurring on umbelliferous plants. Food Addit. Contam : 0 0 Statistix Version.0 (000) Analytical Software. Tallahassee, Florida Torres A, González HHL, Etcheverry M, Resnik SL, Chulze S () Production of alternariol and alternariol monomethyl ether by isolates of Alternaria spp. from Argentinian maize. Food Addit. Contam :-0 USDA (0).Oilseeds and Products Annual, Global Agricultural Information Network, Argentina. %0Aires_Argentina_--0.pdf. Accessed November 0 Vaamonde G, Patriarca A, Fernández Pinto V, Comerio R, Degrossi C(00) Variability of aflatoxin and cyclopiazonic acid production by Aspergillus section flavi from different substrates in Argentina. Int J Food Microbio l:- Villarroel DA, Baird RE, Trevathan LE, Watson CE, Scruggs ML(00) Pod and seed mycoflora on transgenic and conventional soybean [Glycine max (L.) Merrill] cultivars in Mississippi. Mycopathologia :0

15 Wrather JA, Anderson TR, Arsyal DM, Gai J, Ploper LD, Porta-Pugliu A, Ram HH, Yorinory JT () Soybean loss estimates for the top soybean producing countries in. Plant Dis : Zelaya MJ(00) Study of the contaminant mycoflora in Argentinian soybean germoplasm.magister Thesis in Bromatology and Food Technology. University of Buenos Aires

16 Fig Fig. Box plots of fungal contamination (%) by location in Argentina, 0

17 Fig Fig. Fr and RD of the fungi with potentially toxicological interest isolated

18 Table Table Percentage of isolation, frequency (Fr) and relative density (RD), of fungal genus and species harvested in Argentina in 0 recovered from soybean seeds Genera Species Genera Species Fungi RD (%) Fr (%) Acremonium strictum.... Alternaria.. Alternaria alternata. 0. Alternaria infectoria.0. Alternaria tenuissima.. Arthrinium phaeospermum Aspergillus.. A. candidus 0.. A. flavus.. A. niger 0.. A. ochraceus 0.. A. parasiticus 0.. A. tamarii 0.. A. terreus 0.. Cladosporium cladosporioides Chaetomium spinosum Epicoccum nigrum Eurotium chevalieri Fusarium.. F. graminearum 0.. F. semitectum.. F. verticillioides.. Mucor racemosus Penicillium.0 0. P. chrysogenum 0.. P. citrinum.. P. corylophilum 0.. Phoma spp Phomopsis sojae Rhizoctonia solani Rhizopus stolonifer Sclerotinia sclerotiorum.... Ulocladium spp.....

19 Table Table P values of the comparison of fungal contamination median between locations in soybean seeds harvested in Argentina in 0 Location América Saladillo Trenque Lauquen Trenel Gral. Villegas América Saladillo Trenque Lauquen Trenel Gral Villegas

20 Table Table Aflatoxins production (µg Kg - ) by Aspergillus species belonging to Flavi group recovered from soybean seeds harvested in Argentina in 0 CIM N Species AFB AFB AFG A. flavus 0 nd A. flavus 0 0 nd A. flavus nd nd nd A. flavus nd nd nd A. flavus nd nd nd A. flavus nd nd nd A. flavus 0 0 nd 0 A. flavus nd nd nd A. flavus 0 A. flavus 0. nd A. flavus 0 nd A. flavus nd A. flavus 0 nd A. parasiticus 0 0 A. flavus nd nd nd CIM: Centro de Investigación en Micotoxinas nd: not detected

21 Table Table Trichothecenes, zearalenone and fumonisins production (µg Kg - ) by Fusarium species recovered from soybean seeds harvested in Argentina in 0 CIM N Species ZEA DON NIV HT- T- FusX Ac- DON Ac- DON FB FB FB F. graminearum 0 nd nd nd nd nd nd nd na na na F. semitectum nd nd nd 0 nd nd nd nd na na na F. semitectum nd nd nd nd nd nd nd nd na na na F. semitectum nd nd nd nd nd nd nd na na na F. semitectum nd nd nd nd nd nd nd nd na na na F. verticillioides na na na na na na na na 0 nd nd F. verticillioides na na na na na na na na 0 nd nd F. verticillioides na na na na na na na na nd nd nd F. verticillioides na na na na na na na na nd nd nd F. verticillioides na na na na na na na na 0 nd nd F. verticillioides na na na na na na na na nd nd nd 0 F. verticillioides na na na na na na na na CIM: Centro de Investigación en Micotoxinas nd: non detected; na: not analyzed

22 Table Table Alternariol and alternariol monomethyl ether production (µg Kg - ) by Alternaria alternata and A. tenuissima species recovered from soybean seeds harvested in Argentina in 0 Nº CIM Species AOH AME 00 A. alternata 00 0 A. alternata A. alternata A. alternata A. alternata A. alternata A. alternata 0 0 A. alternata 00 0 A. alternata 0 0 A. tenuissima A. tenuissima A. tenuissima A. tenuissima A. tenuissima A. tenuissima 0 0 A. tenuissima nd 0 0 A. tenuissima A. tenuissima nd 0 0 A. tenuissima 00 A. tenuissima 0 0 A. tenuissima 0 0 A. tenuissima 0 nd 0 A. tenuissima 0 0 A. tenuissima nd nd A. tenuissima nd 0 00 A. tenuissima 0 0 A. tenuissima A. tenuissima A. tenuissima nd 0 0 A. tenuissima 0 0 CIM: Centro de Investigación en Micotoxinas nd: non detected