Mycoflora and Mycotoxins natural occurrence in corn from entre Rios Province, Argentina

Mycotoxin Research, Vol. 17 (2001) Mycoflora and Mycotoxins natural occurrence in corn from entre Rios Province, Argentina Pacin AMa,b,8roggi LEc, Resnik SL b,d*and González HHLe,f a Centro de nvestigación en Micotoxinas, Universidad Nacional de Luján, Luján, Argentina. b Comisión de nvestigaciones Científicas de la Provincia de Buenos Aires, Argentina. e Facultadde Bromatología,UniversidadNacionalde Entre Ríos, Argentina. d Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Argentina., e Consejo Nacional de nvestigaciones Científicas y Técnicas, Argentina. f Facultad de ngeniería, Universidad de Buenos Aires, Argentina. *Depto de ndustrias, FCEyN-UBA,Cdad Universitaria. Fax: 011-4631-1148. E-mail: resnik@di.fcen.uba.ar Abstract Corn samples were collected in 1999 from three departments of Entre Ríos province, Argentina, and were surveyed for mould contamination and natural occurrence of Fusarium mycotoxins, ochratoxin A and aflatoxins. Fusarium verticillioides was the most prevalent fungal species recorded at all departments. Zearalenone, deoxynivalenol and ochratoxin A were not found in any samples. Only one of the 52 corn samples analysed was contaminated with aflatoxin 81 (17 J,lg/kg).Fumonisin 81 was found in 58 % of samples (range of positive samples: 47-3,347 J,lg/kg), fumonisin 82 in 33.0 % (range of positive samples: 23-537 J,lg/kg)and fumonisin 83 in 25.0 % (range of positive samples: 24-287 J,lg/kg) of them. This is the first report on the natural occurrence of mycotoxins in corn from Entre Ríos province, Argentina. Levels of fumonisins were lower than detected in other Argentinian provinces. 31

ntrqduction Com (Zea mays L.) is the main cereal produced in Argentina, it represents c.a. 50 % of total exported cereal grains and oilseeds. n Entre Ríos, com is the m.ost important cereal grain produced (30 % of total cereal grains and oilseeds is harvested in this province), being an important income for the regional economy (2). Natural oécurrence of atlatoxins, zearalenone; trichothecenes and fumonisins in Argentinian cereals has already been detected (3, 4, 6, 10, 13, 15). Simultaneous occurrence of fumonisins and zearalenone in com from Buenos Aires, Córdoba and Santa Fe provinces was reported in the 1995 harvest (5). There are not available data about fungal and mycotoxins contamination in com harvested at Entre Ríos province. During sowing time until harvest period (September to March), in the three localities studied, the average monthly precipitations was 94.4 mm, the average of maximum temperatures was 26.2 C and the mean of minimum temperatures was 15.6 C. The aim of this work was to assess the fungal contamination and natural cooccurrence of fumonisins (FB), atlatoxins (AF), zearalenone (ZEN), deoxynivalenol (DON) and ochratoxin A (OA) in the 1999 com harvest at Entre Ríos province, Argentina. ~ Materials and Methods Corn samples During the crop season in 1999, fifty-two com samples were colected at random from three different departments of Entre Ríos province: Diamante (\ 5 samples), Paraná (\ 6 samples) and Gualeguaychú (21 samples). These three departments represent c.a. 50.0 % ofthe provincial com production (2). At the harvest time, when the trucks arrived to store plants subsamples of com were taken in six different positions of the trucks. Samples of 10 kg of randomised com were undertaken as described by Apro el al. (1). They were mixed and subsamples of 1 kg were dried (average moisture content: 13.5 % w/w dry basis). Each sample was prepared for analyses by grinding in a laboratory mili (Romer Labs lnc., MO, USA) and approximately 250 g subsamples were maintained at -18 C prior mycotoxin analysis. Mycology solation of fungi. For isolation of the intemal mycotlora, subsamples of com kemels from each sample were treated according to González el al. (7). dentification of fungi. solates of fungi were identified according to the folowing authorities. Fusarium spp. according to Nelson et al. (8); Penicil/ium spp., Aspergil/us spp. and other fungi according to Pitt and Hocking (12). The relative density (RD) of species was calculated according to González el al. (7) as folows: 32

Mycotoxin Research, Vol. 17 (2001) Where m. RD(%) = ( Ni) x 100 ni = Number of isolates of a genus or speeies Ni = Total number offungi isolated Mycotoxins Analysis Aflatoxins, Deoxyniva/eno/ and Zeara/enone ana/ysis. The samples were analysed for the presenee of these myeotoxins aecording to the method of Romer N MY8402s (9). Briefly, subsamples (50 g) were extraeted with 100 ml of aeetonitrile : water 84:16 (v/v) by blending for 3 min at high speed with an Osterrizer blender. The extraet was filtered through Whatman N 4 paper. Approximately 8 ml of the filtrate were plaeed in a 15 x 85 mm eulture tube and 4 ml of the filtrate were passed through a Myeosep 226 eolumn (Romer Labs. ne., MO, USA). Then, the purified extraet was transferred to another tube and evaporated in a 60 C water bath under vaeuum (17). The residue was redisolved in toluene:aeetonitrile 95:5 (v/v) and a 50 /ll aliquot was applied on Merek siliea gel TLC plates (No 1.05553) and developed in toluene:aeetone 1:1 (v/v). Sample spots of AF and ZEN were visualised by fluoreseence (AF: 366 nm, ZEN: 254 nm). Then, plates were air dried, sprayed with 20 % AC3ethanol:water 1:1 (v/v) solution and heated at 150 C for 7 minothe levels of DON were visualy estimated (Rf ==0.5) by eomparison with standard (Sigma, USA) spots under long wave UV lighí. The visual deteetion limit was 1 /lglkg for AF and 25 /lglkg for ZEN and DON. Average reeoveries for each aflatoxin were greater than 95 % at a level of 10 /lglkg. The average reeoveries for DON and ZEN were greater than 90 % at a level of 100 /lg/kg. Fumonisins ana/ysis. Coro subsamples were analysed to determine the presenee of FBh FB2 and FB3 aeeording to the method of Sydenham et a/. (6). Briefly, aqueous methanol extraets were prepared for eaeh one of the samples, whose aliquots were applied to eonditioned strong anion exehange (SAX) solid-phase extraetion eartridges. The cartridges were washed to remove other compounds, and the fumonisins seleetively eluted with aeetic aeidlmethanol solution, the eluates were collected and dried. 0- Pthaldialdehyde (OPA) derivatives of eaeh purified sample extraet were prepared and analysed with HPLC utilising fluoreseenee deteetion with a LC eolumn - stainless steel, 25 em x 4 mm id, packed with Liehrosorb 5 /lm C18 reversed - phase material (Merek & Co., Darmstadt, Germany). dentifieation and quantifieation of the fumonisins were made by eomparison of the retention times and peak areas in the samples with those observed for fumonisins standards (Medical Researeh Council, Tygerberg, South Afriea). Deteetion limits were: FB): 10 nglg; FB2: 50 nglg and FB3: 10 ng/g (Signal -to - noise ratio = 3:1). Average recoveries for FB and FB3 were greater than 94 % at a level of 100 /lglkg of spiked sample and for FB2 greater than 90% at the same leve\. Ochratoxin A ana/ysis Coro subsamples were analysed to determine the presence of oehratoxin A aceording to the method of Romer Labs N MY8501. Briefly, 25 g of sample were extracted 33

with 12,5 ml of H3P040.1 M and 125 ml of dichlorometaneby blending for 1 minute at high speed with an Osterrizer blender. The organic extract was filtered through Whatman N 4 paper, 4 ml of N-hexane were added to 5 ml of the filtrate and it was passed through a MycoSep 212 ROMER column. The filtrate was discarded and the column was washed three times with 5 1'1Lof dichlormetane. Ochratoxin A was eluted with 20 ml of dichlormethane:formic acid 99: r (v/v). Then, the purified extract was evaporated in a 60 C water bath under vacuum. The residue was dissolved in 100 ~L of toluene:acetic acid '99:1 (v/v) and 50 ~L aliquot was applied on Merck silica gel 60 TLC plates (No 1.05553) and developed in toluene:methanol: acetic acid 18:1:1 (v/v). OA spots of samples were visualised by fluorescence under UV light (360 nm) and the levels of OA were visually estimated (Rf ==0.3) by comparison with standard (Sigma, USA) spots. Detection limit is 2 ~g/kg. The average recovery for OA was greater than 96 % at a level of lo~g/kg. Statistic analysis Statistix for Windows (14) software was used to compare fumonisins B> B2 and B3 results in coro samples at the three departments by using descriptive statistics. Results and Discussion The mycological examination ofthe 52 field coro samples is shown in Figure l. t can be seen that Fusarium vertici/lioides was the prevalent toxigenic fungi present in all departments. This result is lower than those observed in coro harvested at the provinces ofbuenos Aires, Santa Fé and Córdoba (88.8 % RD) by González el al. (7). n Paraná department was observed the highest RD for Alternaria alternata (6.9 %) and for Aspergi/lus jlavus (8.2 %). Other toxigenic fungi recorded at low incidence levels were Penici/lium funiculosum, F graminearum, F proliferatum, F semitectum and P citrinum. González et al.(7), observed on freshly harvested coro from Buenos Aires, Santa Fé and Córdoba RD values of91.9 % for A alternata, 85 % for Ajlavus, 7.1 % for Pfuniculosum and 2.4 % for P citrinum. Aflatoxins, ochratoxin A, zearalenone and deoxynivalenol contamination levels are presented in Table l. OA, ZEN and DON were not detected in any sample, but one sample was contaminated with aflatoxin B' No significant differences were found in fumonisins contamination level among departments and it is observed in Figure 2, the distribution of the fumonisins contamination were not normal. Linear regressions were observed between the FB contamination in each sample. The best adjustment was found between FB and FB2 contamination levels (r2 = 0.9497) and between FB2 and FB3 (r2 = 0.9336). These findings allow us to caculate the relationship among fumonisins in the coro samples from Entre Ríos and to indicate that FB contamination would be approximately five times higher than FB2 contamination, and eight times than FB3 contamination. González et al. (5), found in coro harvested at Buenos Aires, Córdoba and Santa Fe provinces, that FB level was approximately three times higher than FB2 coritamination. 34

Mycotoxin Research, Vol. 17 (2001) 60,0 50,0 i 40,0 i "PARANA "DAMANTE..GUALEGUAVCHÚ RD % 30,0 i 20,0 10,0 0,0 l h,.,:, a.....,,..,..,.. ""... A B. C D E G H -- ~. ti. A- Altemaria alternata, B- Aspergillus flavus, C- Fusarium graminearum, D- F verticillioides, E- F pro/iferatum, F- F semitectum, G- Penicil/ium citrinum, H- P funiculosum Figure 1 - Toxigenic fungi recovered from corn at three departments in Entre Ríos province, Argentina, 1999 harvest. 4600 Outlaver values * 'Oí! 3600 ~2700 S E ] 1800 Ci! 15 E- o o * * loo opaa~a D_TE GCHU GCHU: Gualeguaychú Figure 2 - Box and Whisker plots of the total fumonisins in corn from three departments of Entre Ríos province, Argentina, 35

Table 1 - Levels of aflatoxins, ochratoxin A, zearalenone and deoxynivalenol in corn samples from three departments of Entre Ríos province, Argentina in the 1999 harvest. Department +/N samples AF (l1g/kg) ZEN (l1g/kg) DON (l1g/kg) AO (l1g/kg) Diamante 0/15 ND ND ND ND Gualeguaychú 0/21 ND ND ND ND Paraná 1/16 16.8 ND ND ND ND: not detected Table 2 - Levels of fumonisin B, (l1g/kg) in corn samples from three departments of Entre Ríos province, Argentina in the 1999 harvest. Department +/total Mean of Mean of Median of Range of positive samples total positive positive samples samples samples samples Diamante 9/15 732.9 1,221.5 602.0 47.0-3,347.1 Gualeguaychú 11/21 170.1 324.8 320.0 77.0-670.0 Paraná 10/16 522.5 836.0 418.5 53.0-2 627.9 Table 3 - Levels of fumonisin B2 (l1g/kg) in corn samples from three departments of Entre Rios province, Argentina in the 1999 harvest. Department +/total Mean of Mean of Median of Range of positive samples total positive positive samples Diamante 8/15 samples samples samples 102.4 192.0 99.8 36.0-537.4 Gualeguaychú 4/21 11.3 59.1 54.1 36.0-92.3 Paraná 5/16 51.7 165.5 150.4 23.0-356.8 Table 4 - Levels of fumonisin B3 (l1g/kg) in corn samples from three departments of Entre Ríos province, Argentina in the 1999 harvest. Department +/total Mean of Mean of Median of Range of positive samples total positive positive samples samples samples samples Diamante 7/15 53.4 114.4 67.6 24.0-287.4 Gualeguaychú 3/21 6.0 41.9 35.0 30.9-59.7 Paraná 3/16 21.0 111.7 93.7 71.6-169.9 36

Mycotoxin Research. Vol. 17 (2001) Fumonisins B, B2and B3 levels for the three departments are presented in Tables 2,3 and 4; these data were lower than the ones found in Buenos Aires, Córdoba and Santa Fe provinces in 1995, where levels of fumonisins in positive samples, ranged trom 460 to 9,950 ~g/kg for FB, trom 140 to 3,060 ~g/kg for FB2 and between 90 and 1,070 ~glkg for FB3, with a total fumonisins concentration ranging from 460 to 14,080 ~glkg. (5). Moreover, the percentage of fumonisins contaminated samples (58% with FB, 33 % with FB2 and 25 % with FB3) was lower than those observed previously by González el al. (5) in Argentina, who found 100 % of corn samples contaminated. They also found co-occurrence of fumonisins and zearalenone. n Uruguay (neighbour country of Entre Ríos province) with almost equal metheorological conditions, similar results in corn were observed in 1997 (11), not only in the range (Uruguay: 165-3,688 ~g/kg, Entre Ríos, Argentina: 47-4,053 ~g/kg for total fumonisins) but also in percentage of contamination (Uruguay: 50 %; Entre Ríos, Argentina: 58 %). As conclusion, it is important to point out the negative co-occurrence of OA, ZEN and DON except one AFB positive sample; and the low fumonisins contamination levels not only in comparison with other Argentinian regions but also with other worldwide contamination corn. Moreover this work is the first report about contaminant mycotlora and mycotoxins occurrence in treshly harvested corn trom Entre Ríos province, Argentina. Acknowledgments The authors are indebted to Ms. M. E. Módena, G. Cano and D. Taglieri for their competent technical assistaflce and also acknowledge to the Consejo Nacional de nvestigaciones Científicas y Técnicas, Comisión de nvestigaciones de la Provincia de Buenos Aires, Universidad de Buenos Aires, Universidad Nacional de Luján, Universidad Nacional de Entre Ríos and Agencia Nacional de Promoción Científica y Tecnológica for the financial support. References 2 3 Apro N, Resnik SL, Ferro Fontán C (987) Representative samples for mycotoxin analysis. Anales de la Asociación Química Argentina 75: 501-510 Secretaría de la Producción de la Gobernación, Subsecretaria de Desarrollo Agropecuario, Economías Regionales y Recursos Naturales Cámara de Cereales de Entre Rios (2000) Entre Ríos, Argentina Broggi LE (998) "Estudio de la distribución de la micotlora contaminante de arroz recién cosechado. Riesgo potencial de aparición de micotoxinas en la Región Arrocera Argentina". M. S. Thesis. University of Buenos Aires, Argentina 37

.. -.-.,---, 4 5 6 7 8 9 lo 11 12 13 14 15 16 17 Chulze S, Bertinetti C, Dalcero A, Etcheverry M, Famochi C, Torres A, Rizzo 1, Varsavsky E (1989) ncidence of Aflatoxin, Zearalenone and Deoxynivalenol on Com in Argentina. Mycotoxin Research 5: 9-12 González HHL, Martínez EJ, Pacin AM, Resnik SL, Sydenham EW (1999) Natural co-occurrence of fumonisins, deoxynivalenol, zearalenone and aflatoxins in field trial com in Argentina. Food Additives and Contaminants 16:565-569. González HHL, Pacin A, Resnik SL, Martinez EJ (1996), Deoxynivalenol and contaminant mycoflora in freshly harvested Argentinian wheat in 1993. Mycopathologia 135: 129-134 González HHL, Resnik SL, Boca RT, Marasas WFO (1995) Mycoflora of Argentinian com harvested in the main production area in 1990. Mycopathologia 130: 29-36 Nelson PE, Toussoun TA, Marasas, WFO (1983) Fusarium species: an illustrated manual for identification. The Pennsylvania State University Press, University Park, Pennsylvania Official Methods of Analysis of the Association of Official Analytical Chemists ntemational, 1995, Volume 11,chapter49, pp 1-49 Pacin AM, Resnik SL, Neira MS, Moltó G, Martínez E (1997) Natural occurrence of deoxynivalenol in wheat, wheat flour and bakery products in Argentina. Food Additives and Contaminants 14: 327-331 Piñeiro MS, Silva GE, Scott PM, Lawrence GA, Stack ME (1997) Fumonisin Levels in Uruguayan Com Products. Joumal of AOAC ntemational 80: 825-828 Pitt J, Hocking, AD (1997) Fungi and Food Spoilage. Blackie Academic & Professional, London-New York Resnik S, Neira S, Pacin A, Martínez E, Apro N, Latreite S (1996) A survey of the natural occurrence of aflatoxins and zearalenone in Argentine field maize: 1983-1994. Food Additives and Contaminants 13: 115-120 Statistix for Windows version 1.0 (1996) Analytical Software. Tallahassee, Florida Sydenham EW, Shepard GS, Thiel PG, Marasas WFO, Rheeder JP, Peralta Sanhueza CE, González HHL, Resnik SL (1993) Fumonisins in Argentinian Field-Trial ComoJoumal of Agricultural and Food Chemistry,. 41: 891-895 Sydenham EW, Shephard GS, Thiel PG (1992) Liquid chromatographic determination of fumonisins B.. B2 and B3 in foods and feeds. Joumal of the Association ofofficial Analytical Chemists 75: 313-318 Wilson n, Romer TR (1991) Use of the Mycosep multifunctional column for liquid chromatographic determination of aflatoxin in agricultural products. JAOAC 74: 951-956 Manuscript received June 01, 2001; accepted June 23, 2001. 38