lable at ScienceDirect

Food Microbiology 27 (2010) 187e198

Contents lists avai

Food Microbiology

journal homepage: www.elsevier .com/locate/ fm

Review

A review on ochratoxin A occurrence and effects of processingof cereal and cereal derived food products

S.C. Duarte*, A. Pena, C.M. LinoGroup of Health Surveillance, Center of Pharmaceutical Studies, University of Coimbra, Pólo III, 3000-548, Coimbra, Portugal

a r t i c l e i n f o

Article history:Received 15 September 2009Received in revised form25 November 2009Accepted 26 November 2009Available online 2 December 2009

Keywords:Ochratoxin AAspergillusPenicilliumCerealHarvestStorageProcessing

* Corresponding author. Tel.: þ351239488400; fax:E-mail address: sofiacanceladuarte@gmail.com (S.

0740-0020/$ e see front matter � 2009 Elsevier Ltd.doi:10.1016/j.fm.2009.11.016

a b s t r a c t

Ochratoxin A (OTA) continues to grab global attention and concern for the hazard and impact thatembody for both human and animals, based on its toxicity and occurrence. Despite OTA has beendescribed in a myriad of foodstuffs, cereal and its derivatives remain the major contributors to OTAexposure. For that reason, a critical review on OTA occurrence reported by recent studies worldwidefocusing on unprocessed and processed cereal foodstuffs is made in this work. Special attention is drawnto the major cereal derived products, namely flour, bread, breakfast cereals, baby/infant foods and theinherently involved technological food processing methods and its influence on the redistribution andchemical modification of OTA.

The paper further examines the factors that influence the OTA content of cereal and its derivedproducts, explicitly the different ecological niches of the ochratoxigenic mycobiota e Aspergillus spp. andPenicillium verrucosum, the agricultural practice involved, harvest procedures and storage conditions, thetype of grain, and the nature and extent of technological processing as well as the ultimate stages ofanalytical quality level of the sampling and analysis of the suspected ingredients or foods.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

The global importance of cereal crops to the human diet andmoreover to the written history of man and agriculture cannot beoverstated (FAO, 1999). The importance ranges from the historical,mythological, cultural, religious, and economical aspects that eventoday subsist and renders them classified as the most importantgroup of food crops produced in the world.

The sustenance provided by cereals can be seen in the Bible butis still corroborated by current statistics. The nearly ubiquitousconsumption of cereals all over the world gives them an importantposition in human nutrition. Besides the high starch content asenergy source, cereals provide dietary fibre, nutritious protein andlipids rich in essential fatty acids (Dewettinck et al., 2008). It hasbeen estimated that global cereal consumption directly providesabout 50 percent of protein and energy necessary for the humandiet, with cereals providing an additional 25 percent of protein andenergy via livestock intermediaries (FAO, 1999).

Cereal grains are the edible seed (grain) of plants belonging tothe grass familye Gramineae (FAO,1999). The contaminated cerealscan represent a direct source of human exposure, by its direct

þ351239827126.C. Duarte).

All rights reserved.

consumption, or an indirect source through the consumption ofproducts derived from animals fed with contaminated feed.

Cereals have a variety of uses as foods, in an assortment thatincludes the usage of different technological processing methods,industrial or domestic/traditional. Bread is one of the mostimportant, being specially made out of wheat (Indian roti; Frenchbaguette), rye (German pumpernickel) and maize (Portuguesebroa). Breakfast cereals are another main food product that isincreasingly consumed worldwide, just as bakery products, likecookies and cakes. Another common usage of cereals is in thepreparation of alcoholic drinks such as whiskey and beer (barley;sorghum), vodka (wheat), American bourbon (rye), Japanese sake(rice), etc. A variety of unique, indigenous fermented foods (Turkishboza; Ethiopian injera; Ghana kenkey), other than leavened breadsand alcoholic beverages, are also produced in regions of the worldthat rely mainly on plant sources of protein and calories.

Due to the global importance of cereals in the diet it is a con-cerning fact their susceptibility to be invaded by molds and, incertain climatic conditions, the production of mycotoxins, that canfurthermore persist from the crops to the final products (Moliniéet al., 2005). In spite of many years of research and the introductionof good agricultural practices (GAP) in the food production, andgood manufacturing practices (GMP) in the storage and distribu-tion chain, mycotoxins continue to be a problem. The impact ofthese natural toxins in both human and animal health and welfare

S.C. Duarte et al. / Food Microbiology 27 (2010) 187e198188

is wide-ranging. OTA was described as one of the first group offungal metabolites that are toxic to animals, which, with the afla-toxins (AFLA), launched the distinctive and individualised scienceof mycotoxicology in the 1960s (Zinedine et al., 2009). OTA(C20H18O6NCl; Fig. 1) is the most toxic member of the ochratoxinsgroup, that also includes its methyl ester, its ethyl ester also knownas ochratoxin C (OTC), 4-hydroxyochratoxin A (4-OH OTA), ochra-toxin B (lacking a chlorine atom on C5 of the dihydro-methyl-isocoumarin ring system) and its methyl and ethyl esters andochratoxin a (OTa; where the phenylalanine moiety is missing).OTA, highly soluble in polar organic solvents, slightly soluble inwater and soluble in aqueous sodium hydrogen carbonate, presentsthe melting points of 90 and 171 �C, when recrystallized frombenzene (containing 1 mol benzene/mol) or xylene, respectively.OTA exhibits ultraviolet adsorption at lMeOH

max (nm; 3) ¼ 333(6400). The fluorescence emission maximum is at 467 nm in 96%ethanol and 428 nm in absolute ethanol. The infrared spectrum inchloroform includes peaks at 3380, 1723, 1678 and 1655cm�1

(Ringot et al., 2006).From the more than 300 mycotoxins isolated and described to

the present day, OTA is one of the most important, based on theobserved teratogenic, embryotoxic, genotoxic, neurotoxic, immu-nosuppressive, carcinogenic (IARC group 2B), and nephrotoxiceffects (JECFA, 2001).

Even if not yet definitely demonstrated, OTAwas suspected to beone of the main etiological agents in Balkan Endemic Nephropathy(BEN) described in the late fifties as a typical human pathology inBalkan areas where OTA-contaminated food was often found. It isalso troublesome that the BEN disease is very similar to endemicporcine nephropathy which was clearly related to the ingestion ofOTA by pigs as depicted in Denmark (Grosso et al., 2003). Asidefrom the health and toxicological perspective, OTA exposure alsofeatures an economical facet in livestock animals, in which thedecrease in productivity (milk, eggs, weight loss) and the increaseof mortality rate are problematic. Among these, ruminants, such assheep and cows, are less susceptible, conversely to monogastricanimals, because the protozoan fraction of rumen fluid is capable ofenzymatic degradation of OTA into a less toxic metabolite, ochra-toxin-a (Matrella et al., 2006). Hence, OTA concentration in bovinemilk is expectedly small, although the excretion rate may varyaccording with OTA ingestion. Among farm animals, pigs areknown to be particularly sensitive to the toxin (Chiavaro et al.,2002). In pigs just as for poultry, the carry-over effect of OTA fromfeed to animal can occur. Swine and poultry diets contain a cerealsand cereal by-products fraction of up to 50e60% on a dry matterbasis, and these raw materials are the preferred substrate for theochratoxigenic fungi growth (Schiavone et al., 2008). Animalderived products can then become contributors to OTA humaneexposure.

Nevertheless, according to the most recent assessment of OTAintake by European consumers (Miraglia and Brera, 2002) and inearlier investigations, cereals have been found to be the most

O

OOHO

Cl

NH

O OH

CH3

Fig. 1. Chemical structure of OTA.

important dietary source of this mycotoxin, contributing to circa50% of the intake. The contribution of cereals and derivativeproducts to OTA exposure is probably equally noteworthy in otherparts of the world, according to recent reports of OTA foodincidence.

The different part is the producing fungi, whose species differ intheir ecological niches, in the commodities affected, and in thefrequency of their occurrence in different geographical regions(Uysal et al., 2009). In brief, in cool to temperate conditions Peni-cillium verrucosum is responsible, while in tropical regions Asper-gillus ochraceus is probably the main source (Zinedine et al., 2009).

For the reasons above pointed out, one of the first maximumlimits established for OTA in the EU was specifically for cereals andderivatives, in 2002. These limits, laid down by EC regulation 1881/2006, are 5 mg/kg for raw cereal grains (including raw rice andbuckwheat) and 3 mg/kg for all products derived from cereals(including processed cereal products and cereal grains intended fordirect human consumption) (CEC, 2006).

This paper aims to bring up to date the current status of OTAcontamination of the worldwide cereal crop and cereal basedproducts destined to human consumption. Additionally, a review ofdata concerning the effects on the technological food processing onthe redistribution and chemical modification of OTA is made,regarding the processing techniques involved in the production ofcereal based commodities.

The knowledge of these factors, their importance and signifi-cance, can be used as a tool in an attempt to predict the OTA contentof these cereal products or even to prevent mycotoxin productionand hence their deleterious effects on humans' and animals' healthand wellbeing.

2. OTA occurrence in cereal grains

The differences in OTA contamination between cereals aremultifactorial. Hence, it is especially delicate to establish a directrelationship between an individual factor and OTA content, andthat is why it is very difficult to anticipatewith all certainty the OTAcontent of each type of cereal (González-Osnaya et al., 2007). That isone of the drawbacks, besides the ones reviewed by Garcia et al.(2009), associated with the incipient development of predictivemycotoxicology, which would be of paramount importance in theprevention of food spoilage.

OTA was originally described as a secondary metabolite ofA. ochraceus by Van der Merwe et al. (1965), and later asa secondary metabolite of several other Aspergillus and Penicilliumspp. (Cabañes et al., 2002). Although the widespread occurrence ofochratoxigenic species has been confirmed, each shows differentbehaviours in respect to ecological niches, the products (substrates)affected and their geographical occurrence. So, the origin ofa certain cereal crop is important since the mycotoxin-producingfungal species differ in their ecological niches.

P. verrucosum grows optimally below 30 �C and down to wateractivity (aw) values of 0.80, and is therefore usually found in cooltemperate regions of northern Europe and Canada, though occa-sionally in Mediterranean sea localities, with temperate climateslike Italy, Spain, France and Portugal (Logrieco et al., 2003; Cabañaset al., 2008; JECFA, 2008). It appears to be uncommon, indeedalmost unknown, in warm climates or in other kinds of foods(JECFA, 2008). This species, slow-growing under any condition, hasbeen reported almost exclusively in cereal and cereal derivedproducts (Cabañes et al., 2002). P. verrucosum is more frequentlyisolated where cooler damp harvesting conditions exist, and soinefficiently drying of grain can result in pockets of growth by thisspecies in storage (Aldred et al., 2008). Recently, Penicillium nordi-cum, formed with some strains isolated mainly from fermented

S.C. Duarte et al. / Food Microbiology 27 (2010) 187e198 189

meat, cheese and other dried proteinaceous foods, split from thelatter specie, although expectedly presenting similar physiologicalcharacteristics. At present, these two species are the only OTAproducers known and accepted in the Penicillium genera, althoughP. nordicum appears to be a minor source of OTA in foodstuffs incomparison with P. verrucosum (Cabañas et al., 2008; JECFA, 2008).

A. ochraceus and its other mesophilic xerophile kindred species,A. westerdijkiae and Aspergillus steynii, pertain to the Aspergillussection Circundati, characterized by the golden brown colouredconidia. They present optimal growth at warm temperatures of24e31�C (range 8e40�C), high aw values of 0.95e0.99 (down to0.80) and pH between 3 and 10. Although they can sporadicallyaffect cereals, other plant products and stored food, they are mostimportant in stored dried fruits, nuts, coffee and cocoa beansduring sun-drying (Logrieco et al., 2003; Cabañas et al., 2008;JECFA, 2008).

The second Aspergillus ochratoxigenic group e the black Asper-gillus species e are classified under Aspergillus section Nigri, ofwhich A. carbonarius is the main representative. Some authors haveindicated that the very closely related species A. carbonarius andA. niger, are found many times in association, and are the source ofOTA in tropical and sub-tropical foods, namely in maturing fruits,especially grapes and dried fruits for the recognized high resistanceto sunlight and ultraviolet light, as well as to relatively hightemperatures. They are also very acid tolerant and prefer a some-what reduced water activity (Cabañes et al., 2002; Logrieco et al.,2003; JECFA, 2008). A. carbonarius, produces larger spores, grows atrather lower temperatures than A. niger, with an optimal temper-ature condition at 30 �C The ability to grow at reduced aw is alsomore restricted: germination occurs down to 0.85 aw at 25 and30 �C. A. niger aggregates are often found in warm and tropicalclimates, because it grows optimally at the relatively hightemperatures of 35e37 �C (range 6e47 �C). A. niger is a xerophile,with germination reported at 0.77 aw at 35 �C. Besides nuts, freshfruit, dried wine fruit and some vegetables (Magnoli et al., 2007),A. niger in association with A. carbonarius is also isolated, althoughnot as frequently, from tropical cereals. It is important to underlinethat only a few of the A. niger isolates are believed to be able toproduce OTA in commercially grown crops (JECFA, 2008).

So, since each product tends to host a specific OTA-producingmold, the environmental conditions and factors that encourage thesubsequent formation of OTA need to be understood (Scudamore,2005).

Furthermore, the composition of the fungal population of a crop,and hence the potential occurring mycotoxins, will depend on thelength and conditions of storage so that the species present atharvest may decline or thrive with time, while the “storage” fungican increase rapidly (Domijan et al., 2005). Historically, fungi havebeen divided into two groups - the first includes the toxigenic fungiwhich invade and produce their toxins before harvest, and isknown as the “field fungi”. The second, which becomes a problemafter harvest, is known as the “storage fungi” group. However, theoriginal source of the fungi in both circumstances is the field.(Miller, 1995). The ochratoxigenic species are considered “storage”fungi, because they are isolated at increased frequency as storageprogresses.

So, OTA produced under particular environmental conditions isconsidered to be a problem of storage and does not normally occurbefore harvest (Magnoli et al., 2006). However, attention should bepaid to the harvest period. In a recent Danish survey, Elmholt andRasmussen (2005) found that most of the harvested grain samplescontained P. verrucosum prior to drying, suggesting that much grainis contaminated prior to storage. Once established, this competitivefungal species is able to dominate under conductive environmentalconditions in stored grain (Cabañas et al., 2008).

As Bhattacharya and Raha (2002) demonstrated, during storageseeds were easily invaded by storage fungi, resulting in loss ofgerminability and degradation, decreasing their value for sowingand for food and feed, thus posing a serious safety, quality andeconomical problem to producers. So, beside the risk of producingOTA, fungi growing on stored grains can reduce the germinationrate along with loss in the quantum of carbohydrate, protein andtotal oil content, and induce increased moisture and free fatty acidcontent, enhancing other biochemical changes.

Nevertheless, the presence of the mold is not always indicativethat OTA occurs. The production of secondary metabolites is notessential to the synthesizing organism but it is regulated by severaloften interwoven environmental signals (Mühlencoert et al., 2004).Furthermore, it is known that optimal conditions of aW andtemperature for mycotoxin production are more restrictive thanthose for fungal growth (Magnoli et al., 2007). As a generalphenomenon, stress is frequently mentioned as a cause for myco-toxin synthesis (Birzele et al., 2000). The contrary situation is alsotrue: since OTA is generally stable, it might be detected long afterthe producing fungi have died out or have been outgrown by otherspecies (Ayalew et al., 2006). The presence of the ochratoxigenicfungi may however be regarded as an indicator of OTA formation.For example, according to Lund and Frisvad (2003), more than 7%infestation of P. verrucosum indicates OTA contamination.

The composition of the fungal population is furthermoreimportant regarding the potential interaction and competitionbetween mycotoxigenic species and other spoilage fungi in cerealgrain substrates. For example, the xerophilic spoilage fungusA. ochraceus is an important coloniser of maize and inevitablyinteracts and competes with other contaminant Fusarium, Asper-gillus and Penicillium species for the maize grain niche. It has beenrecently demonstrated that niche overlap and dominance byA. ochraceus is influenced by water availability and temperature,and that in vitro interaction and competition markedly influencedthe production of OTA by this species. Nevertheless, the sameauthors (Lee and Magan, 2000) suggest that, to a large extent,A. ochraceus is not as competitive as some other spoilage fungi inprimary resource capture on maize grain at aw of 0.95 or above,although it may modify resource quality and influence secondarycolonization by other species under the appropriate conditions.

The production and occurrence of OTA in cereal grains isconsidered to depend first and foremost on the condition of thegrain at harvest, how carefully the grain is dried and the quality ofthe storage facilities (Eskola, 2002). In the specific cases of NorthernandWestern Europe, Canada and other temperate areas, cereals areat greater risk for the formation of OTA because grain is oftenharvested at high moisture content, sometimes above 20% (Scu-damore et al., 2003). The occurrence of OTA in such grains isattributed to the insufficient drying or over-long storage beforedrying (Uysal et al., 2009). Given favourable conditions, e.g. cerealsharvestedwith a high content of water, inefficient drying or storageunder humid conditions, high levels of OTA in cereals can occur(Jørgensen and Jacobsen, 2002). For example rice, one of the mostimportant crops worldwide, is an aquatic plant and is usually har-vested at very high moisture levels, between 35 and 50% (Zinedineet al., 2007). As a result, mycotoxin-producing molds or sporescould infect rice crops in the field and at harvest. Infected grainmaylater contaminate those products already in storage and OTAwill beproduced if environmental conditions are favourable (Pena et al.,2005). The conditions at harvest also determine higher OTAcontamination of wild rice, when compared to conventional rice. Inthe first type a highermoisture level at the time of harvest, which ismaintained for one or two weeks to accomplish fermentation, andbecause of a further stimulation of OTA production due to the highfree amino acid content inwild rice, which possesses twice as much

S.C. Duarte et al. / Food Microbiology 27 (2010) 187e198190

protein and amino acid than white rice (Gonzalez et al., 2006). Thisis in agreement with the previous study of Pena et al. (2005) inPortugal, in which none of the analysed white rice samples werefound to be contaminated, in contrast with the brown, basmati,aromatic and wild rice samples. Furthermore, if inadequate storageconditions exist, proliferation of toxigenic fungi might occur,regardless of the season in question. Indeed, Nguyen et al. (2007)determined an overall higher OTA contamination in samplescollected from Vietnamese local markets during the dry seasonwhen compared to the ones commercialized during the rainyseason.

Although it is impossible to completely eliminate all sources ofmold infection, it is possible to lessen or even avoid some particularconditions conductive to mold growth. In simple terms, the cerealmust be dried below 15% (a water activity of about 0.8) and kept atthis level throughout storage. Delays in drying then put the grain atrisk, which can lead to subsequent problems during storage (Scu-damore, 2005).

So, since the weather conditions vary between different harvestyears, the year in which the sampling and analysis is made is alsoinfluential. That was observed by Czerwiecki et al. (2002a, b) whenanalysing two different consecutive harvest years. In average,regardless of the type of cultivation and the kind of cereal grain, thefrequency and contamination levels of OTA in cereal from 1998were substantially higher than those from 1997. The authorsjustified the results with a higher mean level of precipitation inPoland in the year of 1998. Similar relation was described byJørgensen et al. (1996) in Denmark, when studying wheat and ryefor OTA contamination. The study reported that harvests from theyears of 1986e1987, characterized by normal to wet climate,featured higher average contamination levels than the followingyears, considered dry or very dry. It was also perceived that thecontamination level was not affected during the cereals' storageperiod, but instead during the period immediate following theharvest, before drying would diminish water activity. If storageconditions are not correct, after a long period OTA production canoccur. In the same study, no differences in OTA levels between branand internal grain were found. Additionally, only 10e50% of OTAcould be extracted and removed from the surface of the grain,indicating that the mold is not limited to the surface, because ofa deeper penetration of the hyphae of themold. Nevertheless, otherstudies (Juan et al., 2008a) report higher OTA frequency for thewhole-grain cereal samples when compared the non-whole-graincereal samples analysed (33% versus 14%). In this study, themaximum value was detected in an organic sample of whole-grainrye (27 ng/g). The fact of the analysed samples being only kernels orconversely whole grain is important sincemost fungi are located onthe surface of the grain, so it is where the higher contamination isexpected. So, whole grain is likely to present higher levels, which iswhy the study of the effect of technological processing is soimportantewhen elevated levels are detected in processed cereals,it could indicate the existence of much higher levels inwhole grain.This was observed by Park et al. (2005), in the Korean polished ricesamples analysed that presented levels ahead of the EU ML. Thedegree of technological processing of the cereal grain is alsoa determining aspect, as pointed out below.

The differences in nutritional composition of the grain, that varyaccording to the type analysed is another factor that might influ-ence the contamination levels determined is the type of cerealgrain studied. Some authors observed higher incidence ofcontamination onmaize, such as Araguás et al. (2003) that reportedthat one of the contaminated corn sample exceeded the EU ML.Nevertheless, several authors have suggested that glutamic acidplays an important factor in the incidence, because it is indirectlyinvolved in the production of OTA in culture and that proline could

be substituted. A study of the metabolism of glutamic acid duringthe production of OTA indicated that portions of it were incorpo-rated into the mycotoxin. A high content of this amino acid incereals, such as wheat, could be a cofactor for the presence of OTA(González-Osnaya et al., 2007).

Other studies reveal a higher OTA concentration in rye grains(Czerwiecki et al., 2002a; Jørgensen and Jacobsen, 2002).

The chemical nature of the seeds influences the seed moisture.Seeds with high oil content possess lower moisture than thosewithhigh protein or starch. It is well known that water availability,defined as water activity (aw) plays an important role in the dete-rioration of stored seeds. The aw may increase as a result ofabsorption of water from the internal seed atmosphere in order toreach an equilibrium with the prevailing high relative humidity ofthe storage atmosphere, particularly during rainy months (Bhat-tacharya and Raha, 2002).

The type of agriculture practices involved in crop production isalso believed to be determinant. In the study of Juan et al. (2008a),the organic cereal samples showed the highest incidence ofcontamination. Likewise, the 400 polish cereal grain samplesstudied by Czerwiecki et al. (2002a, b) from the 1997 and 1998harvests, obtained from conventional and ecological farms wereinvestigated for the presence of OTA. In the 1997 crop, thefrequency of OTA in the samples of all types of cereal grain origi-nating from ecological farms was substantially higher than thatfrom conventional farms. Specifically, OTA contamination ofecological rye was over six times more frequent than that fromconventional cultivation. An analogous relation was observed forthe rye samples analysed by Jørgensen and Jacobsen (2002), pre-senting a multiyear mean concentration of OTA higher for organicproduction than for the conventional production, both for the rawgrain and flour. This was not observed for the wheat samples. In thecase of rice samples the higher contamination from productsproceeding from organic practice was also observed by Gonzalezet al. (2006). According to the later authors, this fact can beattributed to the cultivation practices and limited use of chemicalproducts like fertilizers, fungicides in organic crop growing.

Another factor that might influence the reported contaminationof cereal products is the testing and quality of the measurements,all the way through the sampling, sample preparation, detectionand interpretation of results (Bao et al., 2006). In the absence ofharmonized guidelines or directives, contamination reportsconstantly reveal variability that is not always a reflection of thereal contamination status. When evaluating assorted cerealsamples in Finland, Eskola (2002) did not detect OTA (n ¼ 115), andin Spain, Blesa et al. (2004) reported an almost absence ofcontamination (n ¼ 43). The authors suggested a low sensitivity ofthe method employed as an explanation.

Furthermore, the lack of regulation and/or control programs onOTA occurrence, at national or regional level, is also affecting theconditions allowed to exist in grain harvest, storage and commer-cialization and that pose a risk to human and animal population.This situation is particularly concerning in the developing coun-tries, which justifies the need of surveys and exposure assessments.So, although considered a natural contaminant of grain primarily innorth-temperate areas, recent cereal surveys in Africa haveencountered high levels of incidence and average concentration.According to Riba et al. (2008), in North African countries the foodsmost susceptible to OTA contamination are precisely cereals, locallyproduced or imported, of which Durum wheat is of paramountimportance in the dry Mediterranean regions of North Africa, asdemonstrated by the cultural tradition of use in the production ofcouscous, pasta, traditional bread and frik.

Ayalew et al. (2006) reported maximum levels that surpassedthe ones reported inmost European surveys. The authors suggested

S.C. Duarte et al. / Food Microbiology 27 (2010) 187e198 191

low quality of the storage facilities (practice of grain storage inunderground pits) and the weather conditions (mild sub-tropicalclimate in most of the sampling areas) to justify the valuesencountered. From all the mycotoxins studied (AFLA B1, OTA,deoxynivalenol, nivalenol, zearalenone, fumonisins), OTA was themost widespread mycotoxin in the barley, sorghum, wheat and teff(a common and staple cereal in Ethiopia) grains surveyed. Most ofthe remaining African surveys refer to Mediterranean basin coun-tries, leaving the epidemiology of the rest of the continent not quiteknown. In Turkey, Terken et al. (2005) detected 96% of cereal-basedflour samples with co-occurrence of AFLA/OTA. In the samecountry, the seed-, pulses-, and cereal-flours (wheat, barley, potato,oat, rye, vetch, corn, rice, lentil, soy) samples randomly collectedfrom Ankara supermarkets and traditional bazaars by Baydar et al.(2005) showed a 100% contamination frequency. In Morocco,Zinedine et al. (2006) registered 5% of OTA/zearalenone (ZEA)contamination. Furthermore, Zinedine et al. (2006) detectedbetween 40% and 55% of contamination in wheat, corn and barley,in linewith the neighbouring survey results (40% of contamination)on Algerianwheat (Riba et al., 2008). All the positive cereal samplesinvestigated by Zaied et al. (2009) in Tunisia surpassed the level ofEuropean regulation (5 mg/kg). The determined high incidence of

Table 1Occurrence of OTA in unprocessed cereals.

Cereal sample Country (Year) Inciden

Cereal grain: UK 2000 52/32Wheat 32/20Barley 20/10Oats 0/1

Conventional production grain: Poland 1997 4/11Rye 3/5Wheat n.d.Barley 1/2

Ecological production grain: 24/12Rye 18/4Wheat 3/3Barley 3/4

Conventional production grain: Poland 1998 23/11Rye 4/3Wheat 18/3Barley 2/3

Ecological production grain: 15/9Rye 5/4Wheat 8/3Barley 2/1

Conventional wheat kernels Denmark 1993e1999 217/40Organic wheat kernels 6/1Conventional rye kernels 257/40Organic rye kernels 14/1Maize Italy 1999e2000 19/7Wheat 6/7Cereal grain (Wheat, Barley and Corn) Spain 58/11Barley Korea 2003 5/2Maize Croatia 19/4Barley Ethiopia 1999 27/10Sorghum 17/7Teff 9/3Wheat 25/10Maize (1998) Ivory Coast 16/1Maize (2001) 15/1Corn Morocco 40%Wheat 40%Barley 55%Organic cereals Spain/Portugal 2005 13/4Non-organic cereals 5/4Rye Japan 2003e2004 9/1Wheat Tunisia 42/11Barley 41/10Sorghum 43/11

(n.d. non-detected).

contamination agrees with the previous national results of Maar-oufi et al. (1995). In the same country, but more recently, theaverage levels of OTA found by Ghali et al. (2008), were 1.9, 3.0 and3.5 mg/kg for barley, wheat, corn and their derivatives, respectively.Twenty six percent of the cereal and cereal based samples wereabout the European maximum permitted level (3 mg/kg). Thehigher incidences were registered for wheat (60.7%), sorghum(52.9%) and barley (52%). These last two showed the highest inci-dence of AFLA and OTA co-occurrence, with 47 and 28%, respec-tively. Sorghum also registered the higher co-occurrence of AFLA,OTA and ZEA (23.5%). An equally higher co-contaminationpercentage was determined in OTA positive rice samples with AFLA(21.6%) and citrinin (61.5%) in Vietnam (Nguyen et al., 2007). Thishigh occurrence alone or with other mycotoxins should bea concern, especially in developing countries, like the onesmentioned above, in which the best quality cereals are exportedwhile the produce with poorer quality is consumed in the home-land. Hence the citizens in developing countries, mainly living inrural areas, are especially sensitive to adverse health effects ofmycotoxins due to the malnutrition and low standards of living(Eskola et al., 2001). It is generally assumed that the mycotoxinproblem is more serious in developing countries where the climatic

ce rate Contamination range(mean) (mg/kg)

Reference

0 (16.25%): 0.3e231: Prickett et al., 20001 (15.9%) 0.3e2316 (18.9%) 0.3e1173 (0%) n.d.0 (3.6%): 0.30e2.50 (1.11): Czerwiecki et al., 2002a2 (5.8%) 0.82e2.5 (1.38)

n.d.6 (3.9%) 0.307 (18.9%): 0.21e57.0 (5.70):8 (37.5%) 0.21e10.0 (3.17)9 (7.7%) 0.48e1.20 (0.83)0 (19.9%) 6.7e57.0 (25.73)0 (20.9%): 0.60e1024 (202): Czerwiecki et al., 2002b7 (10.8%) 4.73e8.80 (6.75)7 (48.6%) 0.60e1024 (267)6 (5.5%) 1.20e9.70 (5.45)7 (15.5%): 0.80e35.3 (7.92):6 (10.9%) 2.0e35.3 (14.5)4 (23.5%) 0.8e1.60 (1.17)7 (11.8%) 1.43e35.3 (18.4)5 (53.6%) n.d.e32 (0.3) Jørgensen and Jacobsen, 20024 (42.9%) n.d.e1.6 (0.3)5 (63.5%) n.d.e63 (0.9)7 (82.4%) n.d.e45 (3.9)0 (27.1%) n.d.e5.2 (1.7) Palermo et al., 20020 (8.6%) n.d.e1.4 (1.47)5 (50.4%) 0.219 Araguás et al., 20032 (22.7%) n.d.e0.9 (0.8) Park et al., 20059 (38.7%) n.d.e2.54 (1.47) Domijan et al., 20053 (26.2%) n.d.e164 (17.2) Ayalew et al., 20068 (21.8%) n.d.e2106 (174.8)3 (27.3%) n.d.e80 (32.7)7 (23.4%) n.d.e66 (19.6)6 (100%) 27e64 (44) Sangare-Tigori et al., 20065 (100%) 3e1738 (266)

n.d.e7.22 (1.08) Zinedine et al., 2006n.d.e1.73 (0.42)0.04e0.80 (0.17)

1 (31.7%) n.d.e27.10 (1.64) Juan et al., 2008a2 (11.9%) n.d.e0.90 (0.05)0 (90%) n.d.e1.59 (1.05) Kumagai et al., 20080 (38%) n.d.e250 (55) Zaied et al., 20093 (40%) n.d.e940 (96)3 (38%) 8e950 (117)

Table 2Reported occurrence of OTA in rice grains and derived foodstuffs worldwide.

Cereal sample Country Year Incidence rate Contamination range (mean) (mg/kg) Reference

Rice Vietnam 2/25 (8%) n.d.e26.2 (23.75) Trung et al., 2001Polished rice Korea 2003 5/60 (8.3%) n.d.e6.0 (1.0) Park et al., 2005Rice Portugal 6/42 (14.3%) n.d.e3.52 Pena et al., 2005Rice and rice products:Non-Organic Spain 5/63 (7.9%) n.d.e27.3 (13,44) Gonzalez et al., 2006Organic 6/20 (30%) n.d.e7.1 (3.6)

Rice Vietnam 35/100 (35%) n.d.e2.78 (0.75) Nguyen et al., 2007Rice Morocco 2005 18/20 (90%) n.d.e32.4 (4.15) Zinedine et al., 2007Rice Tunisia 2004e2005 4/16 (25%) n.d.e2.3 (1.4) Ghali et al., 2008Rice Tunisia 27/96 (28%) n.d.e150 (44) Zaied et al., 2009Rice Chile 2006 13/31 (42%) n.d.e12.5 Vega et al., 2009

(n.d. non-detected; a e includes: white, brown, basmati, aromatic and wild rice).

S.C. Duarte et al. / Food Microbiology 27 (2010) 187e198192

conditions and the agricultural and storage practices are consid-ered conductive to fungal growth and toxin production (Magnoliet al., 2007).

On the contrary, in South American countries, several studiesreport the scarce incidence of OTA in foodstuffs for humanconsumption, conversely to the incidence in feedstuffs. Forexample, in Brazil, Caldas et al. (2002) analysed, from July 1998 toDecember 2001, the presence of OTA in several food samples,including maize and maize-derived products (popcorn andkernels). None of the samples were found positive.

The worldwide occurrence of OTA in unprocessed cereals and inrice grains is shown in Tables 1 and 2, respectively.

3. OTA occurrence in cereal derived products

OTA occurrence is not confined to raw materials, but is alsofound in processed products, due to the mycotoxin's thermosta-bility (Czerwiecki et al., 2002a). Accordingly, it can be stated thatthe factors that determine the incidence and contamination level ofcereal grains are echoed in greater or lesser extent in OTA content oftheir derived products, like flour.

Wheatflour inparticular is an ingredient used inmany foods andis one of the most important foods in European and Americanculture. Bread, pasta, crackers, many cakes, amongst many otherfoods and cooking recipes, are made using flour or include thisingredient. Flour is the cleanest end product of the milling processand is generally regarded as a microbiologically safe product as it isa low water activity commodity. However, P. verrucosum conidiapresent inflourmay survive for several years and for this reason careshould be taken in the storageofflour. Althoughwater activity of dryflour is too low to support growth or OTA production, changes inmoisture contents of 1% or 2% may be sufficient to promote mold

Table 3Reported occurrence of OTA in cereal flour worldwide.

Cereal sample Country (Year) Incidence rate

Wheat conventional Denmark 1992e1999 108/156 (69.2%)Wheat organic 101/120 (84.2%)Rye conventional 138/165 (83.6%)Rye organic 140/155 (90.3%)Wheat Turkey 2002e2003 12/12 (100%)Wheat Korea 2003 0/35 (0%)Maize Turkey __/10White __/22Whole meal __/14Wheat Japan 2003e2004 28/50 (56%)Oat 10/20 (50%)Buckwheat 8/10 (80%)Wheat Chile 2006 21/30 (70%)

growth andmycotoxin production (Cabañas et al., 2008). In the caseof warmer countries, the problem is not associated to the presenceof P. verrucosum, but rather to that of Aspergillus spp.. But the situ-ation is similar, as stated byRiba et al. (2008). The northAfricanfloursamples analysed by these authors contained a high frequency ofochratoxigenic species of Aspergillus. Despite there being no linearcorrelation between OTA content and the number of contaminatingochratoxigenic species, the authors highlight thatwhenever there isa problem in the processing or in the storage of wheat and wheat-based feeds that allows fungal growth, the risk of OTA contamina-tion should be taken into account.

White flour for baking contains much lower concentrations ofOTA than whole meal flour because the bran and offal containinghigh levels of this mycotoxin have been removed. Further lossduring the baking stage is small (Cabañas et al., 2008). The presenceof OTA in different flour types is widespread, as shown in Table 3.

Bread is one of the most important sources of carbohydrates, inthe form of starch, in the human diet (Dewettinck et al., 2008), sothe fact that OTA is commonly associated with bread products is animportant safety issue. Several studies (Table 4) have pointed outbread as one of themain sources of daily intake of OTA, especially inthe European southernmost countries (Legarda and Burdaspal,2001; Bento et al., 2009; Duarte et al., 2009). The formation of OTAon moldy bread is believed to be a risk for human health eitherdirectly, as a result of people eating moldy bread, or indirectly, asa result of consumption of products of animals fed with moldybread. The presence of OTA in bread mainly comes from the wheatflour used for its manufacture because its presence in wheat grainor in flour is only partly eliminated during the bread makingprocess (Arroyo et al., 2005).

Regarding the agricultural practices, González-Osnaya et al.(2007) determined a slightly higher incidence of OTA in bread of

Contamination range (mean)(mg/kg)

Reference

n.d.e16 (0.3) Jørgensen and Jacobsen, 2002n.d.e19 (0.5)n.d.e30 (0.8)n.d.e68 (1.8)(0.75) Baydar et al., 2005n.d. Park et al., 20053.86e15.67 (6.39) Cengiz et al., 20071.87e7.44 (6.89)1.07e16.70 (9.30)n.d.e0.48 (0.09) Kumagai et al., 2008n.d.e0.18 (0.09)n.d.e1.79 (0.51)n.d.e2.1 Vega et al., 2009

Table 4OTA occurrence bread worldwide.

Bread type Country (Year) Incidence rate Contamination range(mean) (mg/kg)

Reference

Wheat bread Spain 93/93 (100%) 0.45 Legarda and Burdaspal, 2001Holland 29/29 (100%) 0.39U.S.A. 24/24 (100%) 0.41Switzerland 20/20 (100%) 0.07Brazil 15/15 (100%) 0.09France 14/14 (100%) 0.25Italy 12/12 (100%) 0.34Germany 11/11 (100%) 0.35Ireland 9/9 (100%) 0.36Austria 9/9 (100%) 0.08Tunes 9/9 (100%) 0.30Belgium 7/7 (100%) 0.23

Bread Germany 897/986 (91%) 0.17/0.19 Miraglia and Brera, 2002Bread Spain 2/20 (10%) 2.55e1.82 (2.19) Osnaya et al., 2006Maize bread Turkey __/10 4.08e5.28 (4.94) Cengiz et al., 2007White bread __/25 n.d.e11.40 (3.36)Whole meal bread __/14 3.54e12.61 (7.84)Organic wheat bread Spain 4/20 (20%) 0.03e0.81 González-Osnaya et al., 2007Other organic cereals bread 2/6 (33.3%) 0.39e0.81Conventional wheat bread 14/67 (19%) 0.04e10.81Other conventional cereals bread 1/7 (14.3%) 2.59Maize bread Portugal 9/15 (60%) 0.43 Juan et al., 2007Wheat bread Morocco 2006 48/100 (48%) 0.14e149 (13) Zinedine et al., 2007Maize bread Portugal (Coimbra) 21/30 (70%) 0.44 Juan et al., 2008bWheat bread 4/31 (12.9%) 0.02Wheat bread Portugal (Bragança) 13/20 (65%) n.d.e0.43 (0.3) Bento et al., 2009

Portugal (Algarve) 24/30 (80%) n.d.e0.49 (0.2)Maize bread Portugal (Lisbon) 4/5 (80%) n.d.e0.36 (0.28) Duarte et al., 2009Wheat bread 19/24 (79.2%) n.d.e0.41 (0.21)Mafra wheat 5/12 (41.7%) <0.1

(n.d. e non-detected; OR-organic; CV-conventional; IPM-integrated pest management).

S.C. Duarte et al. / Food Microbiology 27 (2010) 187e198 193

organic productionwhen compared to conventional production, 23versus 20%. Still, there were five out of 74 non-organic samples thatexceeded the EU maximum level. These included several types ofwheat bread such as precooked, loaves of bread and breadcrumbs.In many bakery products, preservatives and humectants are oftenadded to prevent growth of spoilage fungi; however, in recentyears, because of a consumer pressure to reduce the use of suchpreservatives, suboptimal concentrations are used, which maystimulate the growth of some spoilage ochratoxigenic fungi, andultimately lead to stimulation of mycotoxin production (Arroyoet al., 2005; González-Osnaya et al., 2007).

Furthermore, given that different bread products have differentaw and pH levels, it has been suggested that these features mightinfluence OTA production, with a high strain-dependence accord-ing to the food matrices, pH and aw (Arroyo et al., 2005).

Moreover, differences in the cereal grains used in bread makingresult in differences in OTA content. For example, breadmade out ofmaize is more frequently contaminated than wheat bread,according to Portuguese (Juan et al., 2008b; Duarte et al., 2009) andTurkish (Cengiz et al., 2007) surveys.

In the study of González-Osnaya et al. (2007), the authorsencountered higher OTA content in bread made only out of wheatthan that made of mixed cereals. These data do not concur withOTA contamination found by other authors (Jørgensen et al., 1996;Rafai et al., 2000; Roscoe et al., 2008; Kabak, 2009) that foundhigher concentrations of OTA in barley and oats than in othercereals. Additionally, the white wheat bread presented a higherincidence of OTA when compared to the whole-wheat breadsamples, which is contrary to most studies' findings and the factthat, in theory, the outer layers of whole wheat grains are morecontaminated, but are removed during scouring processes (Scu-damore et al., 2003; González-Osnaya et al., 2007). In fact, it is alsocontrary to what is observed in breakfast cereals, another impor-tant cereal derivative currently surpassing and substituting bread

consumption in many countries. For example in France, Moliniéet al. (2005) detected the maximum concentration in a samplecontaining dry fruit and bran. About 25% of the analysed samplessurpassed the regulatory limit of 3 mg/kg. The risk factors pointedby the work were chocolate, raisins and bran incorporated in thebaking mixture. These risk factors were confirmed with a subse-quent Greek study conducted by Villa andMarkaki (2009), inwhichthe most contaminated samples were breakfast cereals with fibres(0.64 mg/kg), dried fruits (0.87 mg/kg) and chocolate (0.51 mg/kg).These preliminary observations are also corroborated by theTurkish study of Kabak (2009), with wheat bran and dried raisins asthe ingredients of the most contaminated samples. In Spain, Ara-guás et al. (2005) demonstrated a statistically significant differencebetween high-fibre and normal-fibre breakfast cereals, with inci-dence and contamination level higher in the former (Table 5). Theseresults could be explained by the fact that OTA accumulates on anddirectly beneath the epidermis of grain seeds (Osborne et al., 1996;Rafai et al., 2000). Breakfast cereals have also been recentlysurveyed in a nationwide 3-year study in Canada. The breakfastcereals more contaminated were, in decreasing order of incidence,oat-, multigrain-, wheat-, corn- and rice-based (Table 5). Amongstthe corn-based breakfast cereals, two were sweetened with grapejuice and two contained oats as a second grain ingredient (Roscoeet al., 2008). Also in breakfast cereals, besides the type of grainused, the abiotic factors such as water availability, temperature and,when the grain is moist, gas composition, are powerful featuresthat determine the growth of spoilage fungi and OTA production(Magan and Aldred, 2005; Zinedine et al., 2009).

The cereal based baby and infant foods are another cerealderived commodity of prime importance given the fact that theyare among the first solid foods eaten by a vulnerable group, char-acterized by a higher consumption in relation to body weight anda somewhat restricted diet. The type of grain used as main ingre-dient also determines the degree of contamination. In the study of

Table 5OTA occurrence in breakfast cereals worldwide.

Sample type Country (Year) Incidence rate Contamination range (mean) (mg/kg) Reference

Breakfast cereals with: Spain 19/21 (90.5%): 0.265: Araguás et al., 2005High-fibre content 11/11 (100%) 0.132e0.975 (0.362)Normal fibre content 8/10 (80%) n.d.e0.368 (0.158)

Breakfast cereals France 31/45 (69%) n.d.e8.8 Molinié et al., 2005Breakfast cereals Greece 2006e2007 33/55 (60%) n.d.e0.87 (0.11) Villa and Markaki, 2009Breakfast cereals Canada (1999e2001) 53/156 (35%) n.d.e1.4: Roscoe et al., 2008corn-based 6/34 n.d.e0.15 (0.12)multigrain-based 16/36 n.d.e1 (0.33)oat-based 17/27 n.d.e1.4 (0.61)rice-based 3/29 n.d.e0.22 (0.13)wheat-based 11/29 n.d.e0.64 (0.3)other (buchwheat) 0/1 n.a.

Breakfast cereals Turkey 2007 9/24 (37.5%) n.d.e1.84 (0.752) Kabak, 2009Breakfast cereals Morocco 4/48 (8.3%) n.d.e224.6 Zinedine et al., 2009

S.C. Duarte et al. / Food Microbiology 27 (2010) 187e198194

Araguás et al. (2005) the multi-cereal based samples (mostly cornand rye) presented a statistically significant higher OTA content,than the “gluten-free” samples (mainlywheat, barley, oats and rye).Conversely to the results of Araguás et al. (2005), in Canada, Lom-baert et al. (2003) reported an evenly distributed occurrenceamong the oat-, barley-, soy-based, multi-grain cereals and theteething biscuits (Table 6). In Italy, an early study by Beretta et al.(2002) found none of the samples of the maize/tapioca based infantfood to be OTA contaminated, irrespective of the agriculturalpractice involved (conventional, integrated and organic). In thelatter, from all the samples analysed, the ones from integrated pestmanagement were the least contaminated, in line with thefollowing survey of Biffi et al. (2004). The disagreement in resultsbetween different surveys might result precisely from the fact thatwhen analysing OTA contamination of a cereal grain or derivedproduct, one cannot expect to justify or even predict a certain levelbased upon a single factor, as pointed out above.

4. Effects of cereal processing on OTA content

Dietary exposure to OTA varies considerably depending ondifferent factors, among which food-processing systems must beconsidered. These systems are often traditional and characteristicof the different geographical regions and, to this day, their influenceon the mycotoxin content of the food finished products ready forconsumption has been scarcely studied (Valle-Algarra et al., 2009).

Mycotoxins, in general, are stable compounds, and OTA, inparticular, is a moderately heat stable molecule that can survivemost food processing operations and, therefore, it appears in finaland derived products (Bullerman and Bianchini, 2007). A workingdefinition of “processing”, as given by Scudamore (2005) refers to

Table 6OTA occurrence in baby and infant cereal food in selected reports.

Cereal-based product Country (Year) Incidence rate

Cereal based baby food: Spain 14/20 (70%):Gluten free 0/5 (0%)Multicereals 14/15 (93%)

Infant cereals Turkey, 2007 4/24 (16.7%)Infant cereals Morocco 0/20 (0%)Type: Canada (1997e1999) 42/161 (26.1%):- Soy formulas 0/1 (0%)- Teething biscuits 1/5 (20%)- Multi-grain cereals 21/72 (29.2%)- Rice based cereals 1/8 (12.5%)- Soy-based cereals 7/22 (31.8%)- Barley-based cereals 10/47 (21.3%)- Oat-based cereals 2/6 (33.3%)

“the application of any combination of chemical, biological orphysical methods used to produce the final consumer food oranimal feed”.

The main foods for which some studies have endeavoured toquantify the conditions and the extent under which OTA isdegraded are cereals and coffee, although some work has beencarried out on other foods as well (JECFA, 2001; Araguás et al.,2005). In technological food processing, and as far as grains areconcerned, a more accurate relationship between the amounts ofOTA and the microbiological status of the cereals have to take theentire mycobiota into account, as many complex interactions mayoccur and influence the amount of OTA produced (Riba et al., 2008).In a general way, it can be stated that OTA is relatively stable onceformed, but under certain conditions of high temperature, acidic oralkaline conditions or in the presence of enzymes breakdown canoccur (Scudamore, 2005; Valle-Algarra et al., 2009).

OTA levels can be reduced initially by cleaning to remove dustand broken grains. However, the reduction is small and probablydepends on the condition of the grain when received (González-Osnaya et al., 2007). In the study of Scudamore et al. (2003), onlya 2e3% reduction of OTA content in barley was achieved bycleaning. Removal of the surface layers by abrasive scouring orpolishing and milling to remove outer layers for white flourproduction lowers OTA levels, since the mycotoxin tends to beconcentrated in the outer bran layers of cereals. Because in thisprocess of milling there is no stage or operation that destroysmycotoxins, which may only be redistributed or concentrated incertain mill fractions, this raises the question of redistributionduring milling so that both reduction and increase in concentrationcan occur depending on the milled fraction examined (González-Osnaya et al., 2007). It also supports the concern about the use of

Contamination range (mean) (mg/kg) Reference

0.187: Araguás et al., 2005n.d.0.035e0.740 (0.245)n.d.e0.374 (0.221) Kabak, 2009n.d. Zinedine et al., 2009n.d.e6.9: Lombaert et al., 2003e

n.d.e0.3 (0.28)n.d.e0.9 (0.40)n.d.e2.4 (2.40)n.d.e0.9 (0.47)n.d.e6.9 (1.00)n.d.e0.4 (0.37)

S.C. Duarte et al. / Food Microbiology 27 (2010) 187e198 195

these by-products in feed (because of the carry-over effect) and infood (like in bran-based breakfast cereals, or whole-cereal bread).Several authors described a higher contamination level of OTA inwhole wheat (Scudamore et al., 2003; Juan et al., 2008a) and wholespelt (Biffi et al., 2004). Conversely, other authors reported anabsence of significant differences in mold counts between wholewheat and white wheat flour (Weidenbörner et al., 2000) or a evenhigher incidence of OTA inwhite wheat bread than inwhole-wheatbread samples (González-Osnaya et al., 2007). These results areparadoxical given that the physical processes of scouring andmilling, where the inner bran coats are removed, decrease thelevels of this mycotoxin in white flour as demonstrated by Osborneet al. (1996) and Subirade (1996). Therefore, in flour manufacture,given that some parts of the wheat grain are removed, there isa reduction of OTA concentrations in flour and subsequent products(Osborne et al., 1996). As a result, in theory white flour for bakingcontainsmuch lower concentrations thanwholemeal flour becausethe bran and offal containing high levels of OTA have been removed(González-Osnaya et al., 2007). As Table 7 displays, whole mealflour and bread shows much smaller reductions in the concentra-tion of OTA during processing, as might be expected, because less ofthe grain is discarded (Osborne et al., 1996).

Furthermore, separate studies cited by González-Osnaya et al.(2007) showed that OTAwas redistributed by milling into the bran,where the highest OTA content is found, and into the white flourfraction with the lowest concentration. Moreover during mixing offood ingredients, flour can serve as a source of fungal contamina-tion to the atmosphere of the processing establishment, leading toa recontamination and a possible mycotoxin contamination of theproducts after baking.

Although OTA is relatively heat stable, the thermal processes canexert different effects, according to the temperature reached.For example, OTA is stable during bread baking, with no loss orreduction of its concentration (Scudamore et al., 2003). However,baking of biscuits resulted in about two-thirds of the toxin beingdestroyed or immobilized (Bullerman and Bianchini, 2007). Thehigher diminishing of OTA content in biscuits can be explained bythe higher temperature reached when compared to bread and forthe lower water content. The same happens in breakfast cerealproduction (Subirade, 1996; JECFA, 2001).

Autoclaving oatmeal with 50% water resulted in a loss of 74% ofthe OTA content, while applying the same process to dry oatmeal orrice cereal effected greater reductions (86e87.5%)(Bullerman andBianchini, 2007). The extrusion processing, largely used in break-fast cereal production, can also reduce the levels of OTA. Scudamoreet al. (2004) studied the stability of OTA during extrusion ofcontaminated whole meal wheat flour, observing that a highertemperature andmoisture content lead to a bigger OTA breakdown.Degradation was also increased by longer residence time, whenlower mass flow rates were applied, because the time the productspent in the extruder was increased. However, the maximum lossobserved was no greater than 40% of the initial amount of OTA(Bullerman and Bianchini, 2007).

In the specific case of the effects of the bakery processing onOTA, the studies are scarce. Furthermore, because baking is done inso many different ways, according to the geographic regions and

Table 7Effect of flour and bread manufacture on the concentration of OTA (mg/kg) in wheat. Per

WHEAT TYPE Initial concentration Cleaning process After cleaning

Hard 618 Clean 624 (0%)Clean and scour 256 (59%)

Soft 643 Clean 473 (27%)Clean and scour 179 (72%)

the cultural tradition, and hence includes different cereal grainsand ingredients, different levels of mycotoxins according to theorigin and storage of the grain and flour, different temperatures,sizes, etc, the research is further complicated. In a study carried byValle-Algarra et al. (2009), the different bread making conditionscustomary in Spain were assessed in relation to the levels ofseveral mycotoxins present in the flour used to bake bread,including OTA. The results indicated that during dough fermen-tation a significant reduction, ranging from 29.8 to 33.5%, of theOTA level takes place, depending on the OTA amount added to thewheat flour. The variation in the toxin reduction compared toprevious studies (Scudamore et al., 2003) may be related to thedependence on the particular strain involved in fermentation,which in turn might be regarded as a possible mean to control OTAin food and beverages frequently OTA contaminated (González-Osnaya et al., 2007). In respect to the factor temperature/timecombination no significant differences were found, with OTAfeaturing the lesser reduction (32.9%) in comparison with the alsostudied type B trichothecenes.

The same study determined a higher and statistically significantreduction in OTA stability in the bread crust (ranging from 20.4 to51.3%) than in the centre (7.3e38.2%). The size of the bread cantherefore, according to the same authors, have a relevant effect andexcuse the conflicting results reported by different authors. Valle-Algarra et al. (2009) underscore that the heating and boiling canthemselves produce new toxic compounds, and so this reductionmight not be innocuous. Furthermore, they remind of other littlestudied aspects of processed foods, such as the binding of toxins tomatrix compounds such as proteins or carbohydrates, may bepossible and bear the risk that during digestion the toxin is releasedas reported for fumonisins.

In the specific case of rice, the unprocessed samples, collecteddirectly from the cultivars, are the ones most prone to the highestcontamination, in agreement with the study of Gonzalez et al.(2006). Rice in this stage is called “paddy rice” and it is the wholegrain taken off the plant at harvest. White rice grain that is nor-mally consumedmakes up less than three-quarters of the weight ofa paddy rice grain, which also includes the hull and bran. Taking thelatter under consideration, one could explain the high OTAconcentration found in this type of sample given that it is in thehulls where the highest concentration of mycotoxins and myco-toxigenic molds is found on most cereals. Once the grain is pro-cessed into rice products, a decrease of OTA contamination can beexpected. Therefore, it is very likely that mycotoxin-producingmolds could also be present and produce important quantities ofOTA during this period, and, even though OTA concentration maypossibly diminish with rice processing, final concentration couldreach important levels, as is the case with the wild rice samplesanalysed (Gonzalez et al., 2006).

A schematic outline of the main operations and conditions thatalong the cereal processing chain (from the grain harvest to thefinished product) can influence the concentrations of OTA in thefinal cereal-based product is given in Fig. 2. It should however beremembered that, as stated above, a number of cultivation prac-tices, preceding the harvest or even the storage phase are evenlycrucial, as the strategic choice of the cereal variety to cultivate in

centages of reduction are given in parenthesis (modified from Osborne et al., 1996).

White flour White bread Wholemeal flour Wholemeal bread

209 (66%) 140 (77%) 555 (10%) 531 (14%)60 (90%) 72 (88%) 127 (80%) 226 (63%)

389 (40%) 224 (65%) 553 (9%) 476 (26%)111 (83%) 94 (85%) 189 (71%) 160 (75%)

Fig. 2. Practices and conditions that can influence OTA concentrations in cereals along the production and processing chain.

S.C. Duarte et al. / Food Microbiology 27 (2010) 187e198196

a given geographic region and the type of agriculture employed,namely integrated, organic or conventional.

The knowledge of the processing effects is not only important torecognize their effects on OTA distribution and/or chemical modi-fication, but can also be employed in the decontamination orreduction of contamination efforts, such as segregation ofcontaminated from non-contaminated kernels, milling, cleaning orwashing, sieving, dehulling, and in a more judicious selection onthe yeast strains used, for example, for bread fermentation.

Finally, as recommended by Scudamore (2005), despite thevaluable data gathered so far on the topic, considerable care mustbe observed when comparing experimental laboratory results withthose obtained in commercial practice. The exact recipes and pro-cessing conditions used by industry are often of a highly sensitivenature and are rarely made public.

Acknowledgements

This study was supported by the Portuguese Science and Tech-nology Ministry, through the FCT under the project heading PTDC/AGR-ALI/65528/2006. The authors are also gratefully recognized toFCT for a PhD fellowship granted to Sofia C. Duarte, SFRH/BD/37409/2007.

References

Aldred, D., Cairns-Fuller, V., Magan, N., 2008. Environmental factors affect efficacy ofsome essential oils and resveratrol to control growth and ochratoxin Aproduction by Penicillium verrucosum and Aspergillus westerdijkiae on wheatgrain. J. Stored Prod. Res. 44, 341e346.

Araguás, C., González-Peñas, E., López de Cerain, A., Bello, J., 2003. Acerca de la posiblecontaminación por ocratoxina A en alimentos. I: cereales cultivados en diversaszonas geográficas de la comunidad foral de Navarra. Alimentaria 3, 23e29.

Araguás, C., González-Peñas, E., López de Cerain, A., 2005. Study on ochratoxin A incereal-derived products from Spain. Food Chem. 92, 459e464.

Arroyo, M., Aldred, D., Magan, N., 2005. Environmental factors and weak organicacid interactions have differential effects on control of growth and ochratoxin Aproduction by Penicillium verrucosum isolates in bread. Int. J. Food Microbiol. 98,223e231.

Ayalew, A., Fehrmann, H., Lepschy, J., Beck, R., Abate, D., 2006. Natural occurrence ofmycotoxins in staple cereals from Ethiopia. Mycopathol 162, 57e63.

Bao, L., Krska, R., Goto, T., Arcinas, M., Morales-Diaz, A., Baldi, J., Poms, R.E., 2006.Impacts of mycotoxin regulations on world trade. Retrieved May 7, 2009 from:http://en.engormix.com/MA-mycotoxins/articles/impacts-mycotoxin-regulatio-ns-world_139.htm.

Baydar, T., Engin, A.B., Girgin, G., Aydin, S., Sahin, G., 2005. Aflatoxin and ochratoxinin various types of commonly consumed retail ground samples in Ankara,Turkey. Ann. Agr. Environ. Med. 12, 193e197.

Bento, J.M.V., Pena, A., Lino, C.M., Pereira, J.A., 2009. Determination of ochratoxin Acontent in wheat bread samples collected from the Algarve and Bragançaregions, Portugal: winter 2007. Microchem. J. 91, 165e169.

Beretta, B., Domenico, R., Gaiaschi, A., Ballabio, C., Galli, C.L., Gigliotti, C., Restani, P.,2002. Ochratoxin A in cereal-based baby foods: occurrence and safety evalua-tion. Food Addit. Contam. 19, 70e75.

Bhattacharya, K., Raha, S., 2002. Deteriorative changes of maize, groundnut andsoybean seeds by fungi in storage. Mycopathol 155, 135e141.

Biffi, R., Munari, M., Dioguardi, L., Ballabio, C., Cattaneo, A., Galli, C.L., Restani, P.,2004. Ochratoxin A in conventional and organic cereal derivatives: a survey ofthe Italian market, 2001e02. Food Addit. Contam. 21, 586e591.

Birzele, B., Prange, A., Krämer, J., 2000. Deoxynivalenol and ochratoxin A in Germanwheat and changes of level in relation to storage parameters. Food Addit.Contam. 12, 1027e1035.

Blesa, J., Berrada, H., Soriano, J.M., Moltó, J.C., Mañes, J., 2004. Rapid determinationof ochratoxin A in cereals and cereal products by liquid chromatography.J. Chromatogr. A. 1046, 127e131.

Bullerman, L.B., Bianchini, A., 2007. Stability of mycotoxins during food processing.Int. J. Food Microbiol. 119, 140e146.

Cabañas, R., Bragulat, M.R., Abarca, M.L., Castellá, G., Cabañes, F.J., 2008. Occurrenceof Penicillium verrucosum in retail wheat flours from the Spanish market. FoodMicrobiol. 25, 642e647.

Cabañes, F.J., Accensi, F., Bragulat,M.R., Abarca,M.L., Castellá, G., Minguez, S., Pons, A.,2002. What is the source of ochratoxin A in wine? Int. J. Food Microbiol. 79,213e215.

Caldas, E.D., Silva, S.C., Oliveira, J.N., 2002. Aflatoxinas e ocratoxina a em alimentos eriscos para a saúde humana. Rev. Saúde Públ 3, 319e323.

CEC (Commission of the European Communities), 2006. Commission Regulation(EC) No. 1881/2006 of 19 December 2006 setting maximum levels for certaincontaminants in foodstuffs. O.J.E.U L-364, 5e24.

Cengiz, M., Oruç, H.H., Uzuno�glu, I., Sonal, S., 2007. Ochratoxin A levels in differenttypes of bread and flour. Uludag Univ. J. Fac. Vet. Med. 26, 7e10.

Chiavaro, E., Lepiani, A., Colla, F., Bettoni, P., Pari, E., Spotti, E., 2002. Ochratoxin Adetermination in ham by immunoaffinity clean-up and a quick fluorometricmethod. Food Addit. Contam. 19, 575e581.

Czerwiecki, L., Czajkowska, D., Witkowska-Gwiazdowska, A., 2002a. On ochratoxinA and fungal flora in polish cereals from conventional and ecological farms e

part 1: occurrence of ochratoxin A and fungi in cereals in 1997. Food Addit.Contam. 19, 470e477.

Czerwiecki, L., Czajkowska, D., Witkowska-Gwiazdowska, A., 2002b. On ochratoxinA and fungal flora in polish cereals from conventional and ecological farms.Part 2: occurrence of ochratoxin A and fungi in cereals in 1998. Food Addit.Contam. 19, 1051e1057.

Dewettinck, K., Van Bockstaele, F., Kühne, B., Van de Walle, D., Courtens, T.M.,Gellynck, X., 2008. Nutritional value of bread: influence of processing, foodinteraction and consumer perception. J. Cereal. Sci. 48, 243e257.

Domijan, A.M., Peraica, M., Jurjevic, Z., Ivic, D., Cvjetkovic, B., 2005. Fumonisin B1,fumonisin B2, zearalenone and ochratoxin A contamination of maize in Croatia.Food Addit. Contam. 22, 677e680.

Duarte, S.C., Bento, J.M.V., Pena, A., Lino, C.M., 2009. Ochratoxin A exposureassessment of the inhabitants of Lisbon during winter 2007/2008 throughbread and urine analysis. Food Addit. Contam. 26, 1411e1420.

S.C. Duarte et al. / Food Microbiology 27 (2010) 187e198 197

Elmholt, S., Rasmussen, P.H., 2005. Penicillium verrucosum occurrence and ochra-toxin A contents in organically cultivated grain with special reference to ancientwheat types and drying practice. Mycopathologia 159, 421e432.

Eskola, M., Parikka, P., Rizzo, A., 2001. Trichothecenes, ochratoxin A and zearalenonecontamination and Fusarium infection in Finnish cereal samples in 1998. FoodAddit. Contam. 8, 707e718.

Eskola, M., 2002. Study of Trichothecenes Zearalenone and Ochratoxin A in FinnishCereals: Occurrence and Analytical Techniques (Dissertation). EELA Publica-tions, Helsinki.

FAO (Food and agriculture organization), 1999. Fermented Cereals. A GlobalPerspective. FAO Agricultural Services Bulletin 138. Food and AgricultureOrganization of the United Nations, Rome. Retrieved June 30, 2009 from: http://www.fao.org/docrep/x2184E/x2184e00.HTM.

Garcia, D., Ramos, A.J., Sanchis, V., Marín, S., 2009. Predicting mycotoxins in foods:a review. Food Microbiol.. doi:10.1016/j.fm.2009.05.014.

Ghali, R., Hmaissia-khlifa, K., Ghorbel, H., Maaroufi, K., Hedili, A., 2008. Incidence ofaflatoxins, ochratoxin A and zearalenone in tunisian foods. Food Control 19,921e924.

Gonzalez, L., Juan, C., Soriano, J.M., Moltó, J.C., Manes, J., 2006. Occurrence and dailyintake of ochratoxin A of organic and non-organic rice and rice products. Int. J.Food Microbiol. 107, 223e227.

González-Osnaya, L., Soriano, J.M., Moltó, J.C., Mañes, J., 2007. Dietary intake ofochratoxin A from conventional and organic bread. Int. J. Food Microbiol. 118,87e91.

Grosso, F., Saïd, S., Mabrouk, I., Fremy, J.M., Castegnaro, M., Jemmali, M., Dragacci, S.,2003. New data on the occurrence of ochratoxin A in human sera from patientsaffected or not by renal diseases in Tunisia. Food Chem. Toxicol. 41, 1133e1140.

JECFA (Joint FAO/WHO Expert Committee on Food Additives), 2001. Safety Evalu-ation of Certain Mycotoxins in Food. WHO Food Additives Series 47; FAO Foodand Nutrition Paper 74. Retrieved August 10, 2008 from: http://www.inchem.org/documents/jecfa/jecmono/v47je01.htm.

JECFA (Joint FAO/WHO Expert Committee on Food Additives), 2008. Safety Evalu-ation of Certain Mycotoxins in Food. WHO Food Additives Series 59. RetrievedNovember 18, 2009 from. http://whqlibdoc.who.int/publications/2008/9789241660594_eng.pdf.

Jørgensen, K., Rasmussen, G., Thorup, I., 1996. Ochratoxin A in Danish cereals1986e1992 and daily intake by the Danish population. Food Addit. Contam. 13,95e104.

Jørgensen, K., Jacobsen, J.S., 2002. Occurrence of ochratoxin A in Danish wheat andrye, 1992e1999. Food Addit. Contam. 19, 1184e1189.

Juan, C., Lino, C.M., Pena, A., Moltó, J.C., Mañes, J., Silveira, I., 2007. Determination ofochratoxin A in maize bread samples by LC with fluorescence detection. Talanta73, 246e250.

Juan, C., Moltó, J.C., Lino, C.M., Mañes, J., 2008a. Determination of ochratoxin A inorganic and non-organic cereals and cereal products from Spain and Portugal.Food Chem. 107, 525e530.

Juan, C., Pena, A., Lino, C., Moltó, J.C., Mañes, J., 2008b. Levels of ochratoxin A inwheat and maize bread from the central zone of Portugal. Int. J. Food Microbiol.127, 284e289.

Kabak, B., 2009. Ochratoxin A in cereal-derived products in Turkey: occurrence andexposure assessment. Food Chem. Toxicol. 47, 348e352.

Kumagai, S., Nakajima, M., Tabata, S., Ishikuro, E., Tanaka, T., Norizuki, H., Itoh, Y.,Aoyama, K., Fujita, K., Kai, S., Sato, T., Saito, S., Yoshiike, N., Sugita-Konishi, Y.,2008. Aflatoxin and ochratoxin A contamination of retail foods and intake ofthese mycotoxins in Japan. Food Addit. Contam. 9, 1101e1106.

Lee, H.B., Magan, N., 2000. Impact of environment and interspecific interactionsbetween spoilage fungi and Aspergillus ochraceus on growth and ochratoxinproduction in maize grain. Int. J. Food Microbiol. 61, 11e16.

Legarda, T.M., Burdaspal, P.A., 2001. Occurrence of ochratoxin A in samples of breadmarketed in Spain and twelve other countries. Alimentaria 321, 89e96.

Logrieco, A., Bottalico, A., Mulé, G., Moretti, A., Perrone, G., 2003. Epidemiology oftoxigenic fungi and their associated mycotoxins for some Mediterranean crops.Eur. J. Plant Pathol. 109, 645e667.

Lombaert, G.A., Pellaers, P., Roscoe, V., Mankotia, M., Neil, R., Scott, P.M., 2003.Mycotoxins in infant cereal foods from the Canadian retail market. Food Addit.Contam. 20, 494e504.

Lund, F., Frisvad, J.C., 2003. Penicillium verrucosum in wheat and barley indicatespresence of ochratoxin A. J. Appl. Microbiol. 95, 1117e1123.

Maaroufi, K., Achour, A., Betbeder, A.M., Hammami, M., Ellouz, F., Creppy, E.E., 1995.Foodstuffs and human blood contamination by the mycotoxin ochratoxin A:a correlation with chronic interstitial nephropathy in Tunisia. Arch. Toxicol. 69,552e558.

Magan, N., Aldred, D., 2005. Conditions of formation of ochratoxin A in drying,transport and in different commodities. Food Addit. Contam. (Suppl 1),10e16.

Magnoli, C., Hallak, C., Astoreca, A., Ponsone, L., Chiacchiera, S., Dalcero, A.M., 2006.Occurrence of ochratoxin A-producing fungi in commercial corn kernels inArgentina. Mycopathol 161, 53e58.

Magnoli, C.E., Astoreca, A.L., Chiacchiera, S.M., Dalcero, A.M., 2007. Occurrence ofochratoxin A and ochratoxigenic mycoflora in corn and corn based foods andfeeds in some South American countries. Mycopathol 163, 249e260.

Matrella, R., Monaci, L., Milillo, M.A., Palmisano, F., Tantillo, M.G., 2006. OchratoxinA determination in paired kidneys and muscle samples from swines slaugh-tered in southern Italy. Food Control 17, 114e117.

Miller, J.D., 1995. Fungi and mycotoxins in grain: implications for stored productresearch. J. Stored Prod. Res. 31, le16.

Miraglia, M., Brera, C., 2002. Assessment of Dietary Intake of Ochratoxin A by thePopulation of EU Member States. Reports on Tasks for Scientific Cooperation.Reports of experts participating in Task 3.2.7. Directorate-General Health andConsumer Protection, Rome, Italy.

Molinié, A., Faucet, V., Castegnaro, M., Pfohl-Leszkowicz, A., 2005. Analysis of somebreakfast cereals on the French market for their contents of ochratoxin A,citrinin and fumonisin B1: development of a method for simultaneousextraction of ochratoxin A and citrinin. Food Chem. 92, 391e400.

Mühlencoert, E., Mayer, I., Zapf, M.W., Vogel, R.F., Niessen, L., 2004. Production ofochratoxin A by Aspergillus ochraceus. Eur. J. Plant Pathol. 110, 651e659.

Nguyen, M.T., Tozlovanu, M., Tran, T.L., Pfohl-Leszkowicz, A., 2007. Occurrence ofaflatoxin B1, citrinin and ochratoxin A in rice in five provinces of central regionin Vietnam. Food Chem. 105, 42e47.

Osborne, B.G., Brown, G.L., Petagine, F., Scudamore, K.A., Banks, J.N., Hetmanski, M.I.,Leonard, C.T., 1996. The effects of milling and processing on wheat contami-nated with ochratoxin A. Food Addit. Contam. 13, 141e153.

Osnaya, L.G., Castillo, J.M.S., Cortés, J.C.M., Vinuesa, J.M., 2006. Extraction andanalysis of ochratoxin A in bread using pressurised liquid extraction and liquidchromatography. J. Chromat. A. 1113, 32e36.

Palermo, D., Pietrobono, P., Palermo, C., Rotunno, T., 2002. Occurrence of ochratoxinA in cereals from Puglia (Italy). Ital J Food Sci 14, 447e453.

Park, J.W., Chung, S., Kim, Y., 2005. Ochratoxin A in Korean food commodities:occurrence and safety evaluation. J. Agric. Food Chem. 53, 4637e4642.

Pena, A., Cerejo, F., Lino, C., Silveira, I., 2005. Determination of ochratoxin A inPortuguese rice samples by high performance liquid chromatography withfluorescence detection. Anal. Bioanal. Chem. 382, 1288e1293.

Prickett, A.J., MacDonald, S., Wildey, K.B., 2000. Survey of mycotoxins in stored grainfrom the 1999 harvest in the U.K. Home-Grown Cereals Authority (HGCA)Project report no. 230.

Rafai, P., Bata, A., Jakab, L., Ványi, A., 2000. Evaluation of mycotoxin-contaminatedcereals for their use in animal feeds in Hungary. Food Addit. Contam. 17,799e808.

Riba, A., Mokrane, S., Mathieu, F., Lebrihi, A., Sabaou, N., 2008. Mycoflora andochratoxin A producing strains of Aspergillus in Algerian wheat. Int. J. FoodMicrobiol. 122, 85e92.

Ringot, D., Chango, A., Schneider, Y., Larondelle, Y., 2006. Toxicokinetics and tox-icodynamics of ochratoxin A, an update. Chem. Biol. Interact. 159, 18e46.

Roscoe, G., Lombaert, A., Huzel, V., Neumann, G., Melietio, J., Kitchen, D., Kotello, S.,Krakalovich, T., Trelka, R., Scott, P.M., 2008. Mycotoxins in breakfast cerealsfrom the Canadian retail market: a 3-year survey. Food. Addit. Contam. 25,347e355.

Sangare-Tigori, B., Moukha, S., Kouadio, H.J., Betbeder, A.M., Dano, D.S., Creppy, E.E.,2006. Co-occurrence of aflatoxin B1, fumonisin B1, ochratoxin A and zear-alenone in cereals and peanuts from Côte d'Ivoire. Food Addit. Contam. 23,1000e1007.

Schiavone, A., Cavallero, C., Girotto, L., Pozzo, L., Antoniazzi, S., Cavallarin, L., 2008.A survey on the occurrence of ochratoxin A in feeds and sera collected inconventional and organic poultry farms in Northern Italy. Ital. J. Anim. Sci. 7,495e503.

Scudamore, K.A., 2005. Prevention of ochratoxin A in commodities and likely effectsof processing fractionation and animal feeds. Food Addit. Contam. Part A 22,17e25.

Scudamore, K.A., Banks, J., MacDonald, S.J., 2003. The fate of ochratoxin A in theprocessing of whole wheat grains. Part 1, milling and bread production. FoodAddit. Contam. 20, 1153e1163.

Scudamore, K.A., Banks, J.N., Guy, R.C.E., 2004. Fate of ochratoxin A in the pro-cessing of whole wheat grain during extrusion. Food Addit. Contam. 21,488e497.

Subirade, I., 1996. Fate of ochratoxin A during breadmaking. Food Addit. Contam. 13(Suppl), 25e26.

Terken, B., Ayse, B.E., Gozde, G., Sevtap, A., Gonul, S., 2005. Aflatoxin and ochratoxinin various types of commonly consumed retail ground samples in Ankara,Turkey. Ann. Agric. Environ. Med. 12, 193e197.

Trung, T.S., Bailly, J.D., Querin, A., Lebars, P., Guerre, P., 2001. Fungal contaminationof rice from South Vietnam, mycotoxigenesis of selected strains and residues inrice. Rev. Méd. Vét 152, 555e560.

Uysal, Ü.D., Öncü, E.M., Berikten, D., Yilmaz, N., Tuncel, N.B., Kıvanc, M., Tuncel, M.,2009. Time and temperature dependent microbiological and mycotoxin(ochratoxin-A) levels in boza. Int. J. Food Microbiol. 130, 43e48.

Valle-Algarra, F.M., Mateo, E.M., Medina, Á., Mateo, F., Gimeno-Adelantado, J.V.,Jiménez, M., 2009. Changes in ochratoxin A and type B trichothecenes con-tained in wheat flour during dough fermentation and breadbaking. Food Addit.Contam. - Part A 26, 896e906.

Van der Merwe, K.J., Steyn, P.S., Fourie, L., 1965. Ochratoxin A, a toxic metaboliteproduced by Aspergillus ochraceus. Nature 205, 1112e1113.

Vega, M., Muñoz, k., Sepúlveda, C., Aranda, M., Campos, V., Villegas, R., Villarroel, O.,2009. Solid-phase extraction and HPLC determination of ochratoxin A in cerealsproducts on Chilean market. Food Control 20, 631e634.

Villa, P., Markaki, P., 2009. Aflatoxin B1 and ochratoxin A in breakfast cereals fromathens market: occurrence and risk assessment. Food Control 20, 455e461.

Weidenbörner, M., Wieczorek, C., Appel, S., Kunz, B., 2000. Whole wheat and whitewheat flourdthe mycobiota and potential mycotoxins. Food Microbiol. 17,103e107.

S.C. Duarte et al. / Food Microbiology 27 (2010) 187e198198

Zaied, C., Abid, S., Zorgui, L., Bouaziz, C., Chouchane, S., Jomaa, M., Bacha, H., 2009.Natural occurrence of ochratoxin A in Tunisian cereals. Food Control 20,218e222.

Zinedine, A., Blesa, J., Mahnine, N., El Abidi, A., Montesano, D., Mañes, J., 2009.Pressurized liquid extraction coupled to liquid chromatography for the analysisof ochratoxin A in breakfast and infants cereals from Morocco. Food Control.doi:10.1016/j.foodcont.2009.04.009.

Zinedine, A., Brera, C., Elakhdari, S., Catano, C., Debegnach, F., Angelini, S., DeSantis, B., Faid, M., Benlemlih, M., Minardi, V., Miraglia, M., 2006. Naturaloccurrence ofmycotoxins in cereals and spices commercialized inMorocco. FoodControl 17, 868e874.

Zinedine, A., Soriano, J.M., Juan, C., Mojemmi, B., Moltó, J.C., Bouclouze, A., Cherrah, Y.,Idrissi, L., El Aouad, R., Mañes, J., 2007. Incidence of ochratoxin A in rice and driedfruits from Rabat and Salé area, Morocco. Food Addit. Contam. 24, 285e291.