Food Additives and Contaminants, July 2005; 22(7): 677–680
Fumonisin B1, fumonisin B2, zearalenone and ochratoxin A contamination of maize in Croatia ´ 2, DARIO IVIC ´ 2, ANA-MARIJA DOMIJAN1, MAJA PERAICA1, ZˇELJKO JURJEVIC 2 ´ & BOGDAN CVJETKOVIC 1
Unit of Toxicology, Institute for Medical Research and Occupational Health, Ksaverska c. 2, 10000 Zagreb, Croatia and 2Department of Plant Pathology, Faculty of Agronomy, University of Zagreb, Svetosˇimunska 25, 10000 Zagreb, Croatia (Received 10 January 2005; revised 1 April 2005; accepted 7 April 2005)
Abstract Mycotoxins are products of moulds that frequently contaminate maize. In this study the presence of mycotoxins fumonisin B1 (FB1), fumonisin B2 (FB2), zearalenone (ZEA) and ochratoxin A (OTA) was determined in 49 maize grain samples collected in autumn 2002. The most frequent finding was that of FB1(100%), followed by ZEA (84%) and OTA (39%), while FB2 was found only in three samples. The co-occurrence of two and three mycotoxins was found in 55 and 37% of samples, respectively. The concentrations (mean SD) of FB1, ZEA and OTA in positive samples were 459.8 310.7, 3.84 6.68 and 1.47 0.38 mg kg1, respectively, and the concentrations of FB2 in three positive samples were 68.4, 109.2 and 3084.0 mg kg1. Although such low concentrations of mycotoxins are not a significant source of exposure in countries with a European diet, a few samples with extreme values indicate that thorough control is needed.
Keywords: Fumonisin B1, fumonisin B2, HPLC, maize, mycotoxins, ochratoxin A, zearalenone
Introduction Moulds produce secondary metabolites called mycotoxins that may contaminate human and animal food. Exposure to high doses of mycotoxins may produce mycotoxicosis in animals and humans (D’Mello and MacDonald 1997; Peraica et al. 1999). Maize (Zea mays) is one of the two most common cultures in the Republic of Croatia and is often contaminated by Fusarium and Penicillium moulds (Jurjevic´ et al. 1999). Among other mycotoxins, Fusarium moulds can produce fumonisins and zearalenone (ZEA). The most frequently found fumonisins are fumonisin B1 (FB1) and fumonisin B2 (FB2). The most frequent mycotoxicoses caused by FB1 in domestic animals are pulmonary oedema in swine and leukoencephalomalacia in horses (IPCS 2000). In some regions of Africa and China, unusually frequent oesophageal cancer in humans was associated with high exposure to FB1. The International Agency for Research on Cancer (IARC) evaluated FB1, as possibly carcinogenic to humans (Group 2B) (IARC 2002). In Croatia, a single investigation of FB1and FB2 in
naturally contaminated maize revealed a high frequency of FB1-positive samples (Jurjevic´ et al. 1999). The oestrogenic mycotoxin ZEA disturbs the reproduction of domestic animals. In tropical, but also in mild climatic zones, premature menarche in girls was associated with the exposure to high doses of ZEA (Saenz de Rodriguez 1984; Szuetz et al. 1997). Maize samples collected in Croatia from individual farmers were analysed for the presence of ZEA using the thin layer chromatography (TLC) method (Balzer et al. 1977). This investigation of 191 samples showed a low percentage (2.6%) of ZEA-positive samples, probably due to the exceptionally dry season. However, the mean concentration in positive samples was high (5100 mg kg1), and the concentration ranged from 43 to 10 000 mg kg1 of ZEA. In another study performed on 26 maize samples collected in the whole of Croatia in 1985, five samples contained ZEA (Pepeljnjak and Cvetnic´ 1986). The mean ZEA concentration was 1192 mg kg1 and the concentration range was 560–3000 mg kg1. Another worldwide contaminant of cereals is ochratoxin A (OTA) (Speijers and van Egmond
Correspondence: Ana-Marija Domijan. E-mail:
[email protected] ISSN 0265–203X print/ISSN 1464–5122 online ß 2005 Taylor & Francis Group Ltd DOI: 10.1080/02652030500132927
678
A.-M. Domijan et al.
1993). This nephrotoxic compound produced by Penicillium and Aspergillus moulds has been evaluated by IARC as possibly carcinogenic to humans (Group 2B) (IARC 1993). Since the 1970s when the hypothesis of OTA involvement in the aetiology of endemic nephropathy was raised (Krogh 1974), most mycotoxin investigations in Croatia were focused on human exposure to OTA in the endemic region. A five-year study (1972–1976) of maize samples collected from the endemic region and using TLC showed that 8.3% of all samples were contaminated with OTA, and the frequency of OTA-positive samples varied from 0 to 15.2% from year to year (Pavlovic´ et al. 1979). Again using the TLC method, a much higher frequency of OTA contamination (26%) was found in maize stored for animal consumption and collected in 1975 (Balzer et al. 1977). In a more recent study, OTA was analysed in maize using high-performance liquid chromatography (HPLC) (Jurjevic´ et al. 1999) and was found in 10 and 25% of samples collected in 1996 and 1997, respectively. In another study 33.3% of OTApositive maize samples were found (Puntaric´ et al. 2001). The OTA concentration in maize samples collected from three regions in 1999 was compared with OTA concentrations in maize collected from the endemic region. The mean OTA concentration was much higher in maize from the endemic region than from other regions, but the latter did not significantly differ in OTA concentrations. Mixtures of mycotoxins are usual in cereals and other commodities. The aim of this study was to determine the frequency and the concentration of mycotoxins FB1, FB2, ZEA and OTA in maize samples collected in different parts of Croatia in a year with very high rainfall in the growing period (July and August). Materials and methods Chemicals Standards of FB1, FB2, ZEA and OTA were purchased from Sigma (St. Louis, MO, USA). Water (Merck, Darmstadt, Germany) and methanol (Kemika, Zagreb, Croatia) used for the mobile phase were of HPLC grade. All other chemicals were of pro analysi grade. Samples of maize Samples of visually mould-free maize (N ¼ 49) were collected from small farms in several maize-growing counties of Croatia, immediately after the harvest in autumn 2002. Each sample (1 kg) was stored at 20 C until it was analysed for the presence of FB1, FB2, ZEA and OTA.
Mycotoxin analysis All mycotoxin analyses were performed on a HPLC system (Varian, USA) with fluorescence detection. Chromatographic data were collected and processed using Star Chromatography Workstation software (Ver. 5.0, Varian, Walnut Creek, CA, USA). Determination of FB1 and FB2 FB1 and FB2 were extracted from 50 g of ground maize samples blended at high speed for 1 min with 100 ml of methanol:water (80:20) and 5 g of NaCl. The 10 ml of extract was filtered and diluted with 40 ml phosphate buffer/0.1% Tween-20. The diluted extract was filtered through a fibreglass membrane (1 mm) and the filtrate (10 ml) was cleaned up using immunoaffinity columns (IAC, FumoniTest, Vicam, Watertown, MA, USA) according to the producer’s instructions. Fumonisins were eluted from IAC columns with an acetic acid:methanol (1:99) solution and the eluate was evaporated under a gentle stream of nitrogen in a water bath at 60 C. Before injection on to the HPLC column, the sample was redissolved in mobile phase, and derivatized by reaction with o-phthaldehyde. The injection volume was 20 ml. The mobile phase for HPLC analysis consisted of methanol and 0.1 M NaH2PO4 2H2O (68:32) and the pH was adjusted to 3.35 with H3PO4. The flow rate was 0.8 ml min1. FB1 and FB2 were separated on the analytical column coupled with a the guard column. Analytical and guard columns were LiChrospher RP-18 (Merck, Darmstadt, Germany) with 5 mm particles and their size was 125.0 4.0 and 4.0 4.0 mm, respectively. The wavelengths of the detector were set at em336 and ex440. Standard curves for FB1 and FB2 were linear (with r2 ¼ 0.996771 and 0.996619, respectively). The detection limits for both mycotoxins were 10 mg kg1, the recovery was above 81%, and the reproducibility, expressed as RSD, was below 10%. Determination of ZEA The extraction of ZEA followed the procedure described by Llorens et al. (2002). Ground samples (4 g) were extracted three times with dichloromethane and the clean-up procedure was performed on SPE cartridges (Bond Elut C-18, Varian, Harbor City, CA, USA). ZEA was eluted from the SPE columns with methanol and the eluate was evaporated to dryness under a gentle stream of nitrogen in a water bath at 60 C. The sample was redissolved in the mobile phase, and 20 ml was injected on to the HPLC. The mobile phase consisted of methanol and water (80:20). The flow rate of the mobile phase was 0.5 ml min1. The guard column and the analytical
Mycotoxin contamination of maize column were LiChrospher RP-18 (Merck, Darmstadt, Germany) with 5 mm particles, and their sizes were 4.0 4.0 and 250.0 4.0 mm, respectively. The wavelengths of the detector were set at em274 and ex440. The linearity of the standard curve (r2) was 0.9992. The detection limit was 0.39 mg kg1, recovery 108% and reproducibility, expressed as RSD was 5.6%. Determination of OTA Ground maize samples (50 g) were blended at high speed for 1 min with 100 ml of methanol:water (80:20) and 5 g of NaCl, and filtered. The filtrate (10 ml) was diluted with 40 ml of purified water. The diluted filtrate was refiltered through a fibre glass membrane (1 mm) and 10 ml of extract were cleaned up using IAC columns (OchraTest, Vicam, Watertown, MA, USA). OTA was eluted from the IAC columns with methanol and the eluate was evaporated to dryness under a gentle stream of nitrogen in a water bath at 60 C. Prior to HPLC analysis, samples were redissolved in the mobile phase and 50 ml was injected on to the HPLC column. For OTA analysis, the mobile phase consisted of methanol, water and acetic acid (70:30:2) with the flow rate of 0.5 ml min1. The guard column and the analytical column were LiChrospher RP-18 (Merck, Darmstadt, Germany) with 5 mm particles. The sizes of the guard and analytical columns were 4.0 4.0 and 125.0 4.0 mm, respectively. The detector wavelengths were set at ex336 and em464. The standard curve was linear (r2 ¼ 0.9982). The detection limit of the method was 0.25 mg kg1, recovery 85% and reproducibility, expressed as RSD, was 6.76%. Results and discussion An earlier study showed that the most common moulds in maize in Croatia are Penicillium and Fusarium (Jurjevic´ et al. 1999). In mild climatic zones, these moulds produce mycotoxins FB1, FB2, ZEA and OTA. The optimal growth of Penicillium and Fusarium moulds and their production of mycotoxins is favoured when the temperature is above 20 C and humidity above aw 0.87 in the
679
growing period (Pitt et al. 2000), which in Croatia lasts from the beginning of July until the end of August. According to the Meteorological and Hydrological Service of Croatia this period was particularly rainy in 2002. In our study, all maize samples contained FB1 (Table I), which corroborates the study of Jurjevic´ et al. (1999) with 97 and 93% of FB1-contaminated maize samples in two consecutive years. The mean FB1 concentration in our study (459.8 mg kg1) is within the mean concentration range found for these two years (645 and 134 mg kg1). In both studies the frequency of FB2-positive samples was very low. These findings are not surprising because it was noticed earlier that contamination with FB2 is usually lower that that with FB1 (Scudamore and Patel 2000). In our study ZEA was found in 84% of samples, but the mean concentration was low (3.84 mg kg1) (Table I). The discrepancy between our results and the low frequency of positive samples found by the Balzer study could be attributed to rainfall differences in the growing period (Balzer et al. 1977). OTA was found in 39% of maize samples, and the mean concentration in positive samples was 1.47 mg kg1 (Table I). A much lower frequency of OTA-positive samples was found in studies using the TLC method to detect OTA (Balzer et al. 1977, Pavlovic´ et al. 1979). Compared to the studies of Jurjevic´ et al. (1999) and Puntaric´ et al. (2001) using the HPLC method, we have found a higher frequency of OTA-positive maize samples. This difference is probably due to the high humidity during the growing period of maize collected for our study. This is the first study on the co-occurrence of FB1, FB2, ZEA and OTA in Croatia. FB1 was found in all samples, ZEA in 84% of samples, and OTA in 39%, while only three samples (6%) were FB2 positive. Maize samples were contaminated with FB1 alone (8%), with two (55%) or three mycotoxins (37%) (Figure 1A). Various combinations of mycotoxins found in maize are presented in Figure 1B. Since maize in Croatia is frequently contaminated with Fusarium moulds, we decided to check the concentrations of FB1 and ZEA, both produced by Fusarium moulds. The co-occurrence of FB1 and ZEA was rather high (49%), although the
Table 1. Fumonisin B1, fumonisin B2, zearalenone and ochratoxin A in maize samples (N ¼ 49).
FB1 FB2 ZEA OTA
Concentration in positive samples (mean SD, mg kg1)
Range (mg kg1)
Number of positive samples
459.8 310.7 68.4, 109.2, 3084.0 3.84 6.68 1.47 0.38
142.2–1377.6 68.4–3084.0 0.43–39.12 0.9–2.54
49/49 3/49 41/49 19/49
680
A.-M. Domijan et al. 8%
A 37%
the Republic of Croatia, grants No. 0022018 and 0178024.
1 mycotoxin 2 mycotoxins 3 mycotoxins 55%
References
B 60 50 40 30 20 10 FB1+ZEA+OTA
FB1+FB2+OTA
FB1+FB2+ZEA
FB1+OTA
FB1+ZEA
FB2
FB1
0
Figure 1. Frequency of maize samples containing one, two or three mycotoxins (A) and their combinations (B) (N ¼ 49).
concentration of ZEA was low. The optimal temperature for fumonisin production is 20 C and for ZEA it is below 10 C (Le Bars et al. 1994). This explains the high concentration of fumonisins and low concentrations of ZEA in our study. The co-occurrence of fumonisins and OTA in maize in Croatia was studied earlier (Jurjevic´ et al. 1999) and our results are similar to those of this study. In Croatia, tolerable levels of grain contamination have been established for ZEA and OTA (200 and 5 mg kg1, respectively). The European Union has only set the 5 mg kg1 limit for raw cereal grain contamination with OTA. No sample exceeded this limit in our study. Although there are no EU limits for FB1, FB2 and ZEA in grain, some European countries have national regulations on these mycotoxins. Romania, Austria, France and Russia have different limits for ZEA concentration in cereals (30, 60, 200, and 1000 mg kg1, respectively). Only one of our samples, containing 39.12 mg kg1, was above the legislation limit. Only in Switzerland is the maximum allowable concentration of FB1 þ FB2 in maize 1000 mg kg1, and this limit is exceeded in five samples in our study. The mean concentration of mycotoxins found in our study does not exceed the mentioned limits, although some samples containing high concentrations of FB1 and FB2 indicate that thorough control of grain is needed.
Acknowledgements We wish to thank Mrs Mirjana Matasˇin for technical assistance. We are grateful for the financial support of the Ministry of Science, Education and Sports of
Balzer I, Bogdanic´ C, Muzˇic´ S, 1977. Natural contamination of corn (Zea mais) with mycotoxins in Yugoslavia. Annales de la Nutrititon et de l’ Alimentation 31:425–430. D’Mello JPF, MacDonald, AMC, 1997. Mycotoxins. Animal Science and Technology 69:155–166. International Agency for Research on Cancer (IARC) 1993. Some naturally occurring substances: food items and constituents, heterocyclic aromatic amines and mycotoxins. IARC Monographs on the Evaluation of Cracinogenic Risks to Humans. Vol 56. Lyon: IARC. International Agency for Research on Cancer (IARC) 2002. Some traditional herbal medicines, some mycotoxins, naphthalene and styrene. IARC Monographs on the Evaluation of Cracinogenic Risks to Humans. Vol 82. Lyon: IARC. International Program for Chemical Safety (IPCS) 2000. Fumonisin B1. WHO Environmental Health Criteria. Vol 219. Geneva: WHO. Jurjevic´ Zˇ, Solfrizzo M, Cvjetkovic´ B, Avantaggiato G, Visconti A. 1999. Ochratoxin A and fumonisins (B1 and B2) in maize from Balkan nephropathy endemic and non endemic areas of Croatia. Mycotoxins Research 15:67–80. Krogh P. 1974. Mycotoxin porcine nephropathy: a possible model for Balkan endemic nephropathy. In: Puchlev A, editor. Endemic Nephropathy. Sofia: Bulgarian Academy of Science. pp 266-270. Le Bars J, Le Bars P, Dupy J, Boudra H, Cassini R. 1994. Biotic and abiotic factors in fumonisin B1 production and stability. Journal of the AOAC 77:517–521. Llorens A, Mateo R, Mateo JJ, Jimenez M. 2002. Comparison of extraction and clean-up procedures for analysis of zearalenone in corn, rice and wheat grains by high-performance liquid chromatography with photodiode array and fluorescence detection. Food Additives and Contaminants 19:272–281. Pavlovic´ M, Plesˇtina R, Krogh P. 1979. Ochratoxin A contamination of foodstuffs in an area with Balkan (endemic) nephropathy. Acta Pathologica Scandinavica, Section B 87:243–246. Pepeljnjak S, Cvetnic´ Z, 1986. Mikolosˇka i mikotoksikolosˇka kontaminacija zˇitarica na sˇirem anefropaticˇnom podrucˇju SR Hrvatske. Posebna izdanja Akademije nauka i umjetnosti Bosne i Hercegovine 80:29–41. Peraica M, Radic´ B, Lucic´ A, Pavlovic´ M. 1999. Toxic effects of mycotoxins in humans. Bulletin of WHO 77:754–766. Pitt JI, Basilico JC, Abarca ML, Lopez C. 2000. Mycotoxins and toxigenic fungi. Medical Mycology 38(Suppl 1):41–46. Puntaric´ D, Bosˇnir J, Sˇmit Z, Sˇkes I, Baklaic´ Z. 2001. Ochratoxin A in corn and wheat: geographical association with endemic nephropathy. Croatian Medical Journal 42:175–180. Saenz de Rodriguez C. 1984. Environmental hormone contamination in Puerto Rico. New England Journal of Medicine 310:1741–1742. Scudamore KA, Patel S. 2000. Survey for aflatoxins, ochratoxin A, zearalenone and fumonisins in maize imported into the United Kingdom. Food Additives and Contaminants 17:407–416. Speijers GJA, van Egmond HP. 1993. Worldwide ochratoxin A levels in food and feeds. In: Creppy EE, Castegnaro M, Dirheimer G, editors. Human Ochratoxicosis and its Pathologies. Montrouge: John Libbey Eurotox Ltd. pp 85–100. Szuetz P, Mesterhazy A, Falkay GY, Bartok T. 1997. Early telarche symptoms in children and their relations to zearalenone contamination in foodstuffs. Cereal Research Communications 25:429–436.
