http://orcid.org/0000-0003-0982-1473
Figures
Abstract
Currently beer is booming, mainly due to the steady rise of craft breweries worldwide. Previous surveys for occurrence of mycotoxins in beer, were mainly focussed on industrial produced beer. The present survey reports the presence of mycotoxins in craft beer and how this compares to industrial produced beer. More than 1000 beers were collected from 47 countries, of which 60% were craft beers. A selection of 1000 samples were screened for the presence of aflatoxin B1, ochratoxin A (OTA), zearalenone (ZEN), fumonisins (FBs), T-2 and HT-2 toxins (T-2 and HT-2) and deoxynivalenol (DON) using a mycotoxin 6-plex immunoassay. For confirmatory analysis, a liquid chromatography tandem mass spectrometry (LC-MS/MS) method was developed and applied. The 6-plex screening showed discrepancies with the LC-MS/MS analysis, possibly due to matrix interference and/or the presence of unknown mycotoxin metabolites. The major mycotoxins detected were DON and its plant metabolite deoxynivalenol-3-β-D-glucopyranoside (D3G). The 6-plex immunoassay reported the sum of DON and D3G (DON+D3G) contaminations ranging from 10 to 475 μg/L in 406 beers, of which 73% were craft beers. The popular craft beer style imperial stout, had the highest percentage of samples suspected positive (83%) with 29% of all imperial stout beers having DON+D3G contaminations above 100 μg/L. LC-MS/MS analysis showed that industrial pale lagers from Italy and Spain, predominantly contained FBs (3–69 μg/L). Besides FBs, African traditional beers also contained aflatoxins (0.1–1.2 μg/L). The presence of OTA, T-2, HT-2, ZEN, β-zearalenol, 3/15-acetyl-DON, nivalenol and the conjugated mycotoxin zearalenone 14-sulfate were confirmed in some beers. This study shows that in 27 craft beers, DON+D3G concentrations occurred above (or at) the Tolerable Daily Intake (TDI). Exceeding the TDI, may have a health impact. A better control of brewing malts for craft beer, should be put in place to circumvent this potential problem.
Citation: Peters J, van Dam R, van Doorn R, Katerere D, Berthiller F, Haasnoot W, et al. (2017) Mycotoxin profiling of 1000 beer samples with a special focus on craft beer. PLoS ONE 12(10): e0185887. https://doi.org/10.1371/journal.pone.0185887
Editor: Vijai Gupta, Tallinn University of Technology, ESTONIA
Received: December 21, 2016; Accepted: September 21, 2017; Published: October 5, 2017
Copyright: © 2017 Peters et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: Our research was partially financed by the Dutch Ministry of Economic Affairs, SenterNovem and the Province of Gelderland. There was no additional external funding received for this study. Innosieve Diagnostics provided support in the form of salary for author Ronald van Doorn, whose role was limited to the collection of craft beer samples and valuable contributions to the introduction considering craft beer. Innosieve Diagnostics did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: Author Ronald van Doorn is employed by Innosieve Diagnostics. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
Introduction
Beer production and consumption is booming like never before, mainly due to the increasing popularity of craft beer. Craft beer is produced by small, independent and traditional breweries according to the definition of the Brewers Association. The number of craft breweries continues to grow, claiming a larger market share every year. In the US alone, a few hundred new craft breweries emerge annually. The total amount of craft breweries in 2014 saw an increase of 19.4% compared to 2013. Of the 3,464 breweries operating in the US, 3,418 were classified as craft breweries [1]. The same phenomenon is seen in other parts of the world. In the Netherlands, 108 new breweries emerged just in 2015, bringing the total brewery count to 382 [2]. The reason for the popularity of craft brewers is that they tend to focus on flavour and tradition, combined with innovation rather than on large-scale and low-cost production. This development already started with the Campaign for Real Ale (CAMRA) in Britain, 43 years ago [3]. Some craft breweries also produce similar beer styles as industrial brewers (e.g. pilsner beers). The vast majority of the craft breweries however, produce ancient beer styles, adjusted classic styles or even newly invented styles. Whereas “regular” beers are brewed almost exclusively with water, malted barley, hop and yeast, craft brewers add a wide range of different ingredients to the brewing process. Some examples are coffee, cacao, tobacco, liquorice, nuts, tomatoes, chili peppers, fruit and a range of spices [4, 5]. A quick count on one of the most popular websites for craft beer [6], shows that there are currently 83 unique beer styles. Because new styles are regularly being invented, this number will likely increase in coming years.
Mycotoxins are fungal metabolites with acute and/or chronic health effects on animals and humans. These effects include diarrhoea, reduced fertility, immunosuppression, cancer and even death [7–9]. Mycotoxins contaminate a wide range of cereals, including wheat [10], maize [11] and oats [12]. Barley is one of the key ingredients in beer and is prone to mycotoxin contamination [13–15]. Occurrence of mycotoxins in beer has been extensively surveyed, utilizing both instrumental analysis as well as immunoassays. A selected overview [16–45] is presented in Table 1. The most reported mycotoxins in beers, at relevant levels, are the type B trichothecene deoxynivalenol (DON), its plant metabolite deoxynivalenol-3-β-D-glucopyranoside (D3G) and fumonisins (FBs). DON and D3G were mainly reported in European beers, while FBs were mainly reported in beers from Africa and Southern Europe. In general, very high contaminations for all mycotoxins, besides T-2 toxin (T-2) and HT-2 toxin (HT-2), were reported previously in African beers. Ochratoxin A (OTA) was mainly reported in European beers, while aflatoxins (AFs) were mainly reported in African and Asian beers. T-2 toxin (T-2), HT-2 toxin (HT-2) and zearalenone (ZEN) were rarely reported. Most beer surveys in Table 1 lack detailed information about specific beer styles, country of origin or alcohol content. The presence of mycotoxins in hops, a key ingredient in beer, has rarely been investigated [46]. Furthermore, mycotoxins may also be introduced in beer upon the addition of commodities other than cereals. The risk of mycotoxin contamination may therefore be higher in craft brewing, where a wide range of commodities are added at various stages of the brewing process [47]. The question therefore arises, whether these new and revived craft beer styles contain more, or different, mycotoxins compared to regular commercial beers. Additionally, the changing climate may contribute to altered levels of mycotoxins in field crops [48, 49] which eventually will lead to altered levels of mycotoxins in beer.
In this work we present a large-scale survey for mycotoxin occurrence in 1000 beer samples with a unique outlook on the upcoming and strongly expanding craft beer market. Beer samples of many different beer styles (representing 60% craft beers) were collected throughout the world, but with a detailed focus on European countries. This selection of 1000 samples was investigated for mycotoxin contamination and to elucidate possible differences between industrial beers and craft beers. Furthermore this survey aimed for a detailed look into the possible occurrence of conjugated (masked) mycotoxins in beer. To facilitate the fast mycotoxin multiplex screening of 1000 beer samples, a previously developed 6-plex microsphere immunoassay method [50, 51] for the detection of DON and D3G, aflatoxin B1 (AFB1), OTA, the sum of T-2 and HT-2, fumonisins (sum of fumonisin B1 (FB1), B2 (FB2) and B3 (FB3) and ZEN in barley, was modified and adapted for beer samples. To confirm the presence of mycotoxin contaminations in a subset of the screened beer samples, a dedicated multi-mycotoxin liquid chromatography tandem mass spectrometry (LC-MS/MS) method for beer was developed. This method includes several conjugated mycotoxins, as well as mycotoxin metabolites, such as aflatoxin M1 (AFM1), ochratoxin B (OTB), nivalenol (NIV) and zearalenone 14-sulfate (Z14S).
This unprecedented survey reveals the discovery of the conjugated mycotoxin Z14S in beer and confirms that high DON and D3G contaminations can specifically occur in craft beer. The sum of these DON and D3G contaminations (DON+D3G) contribute to surpassing the tolerable daily intake (TDI) of DON upon moderate beer consumption.
Materials and methods
Instrumentation
The 6-plex microsphere immunoassay was performed on a flow cytometer platform (FM3D) or on a planar bead array analyzer (MAGPIX), both running on XPONENT software (all from Luminex, Austin, USA). Mycotoxin concentrations were calculated using the xMAP dedicated Bio-Plex manager software 6.0, combined with Bio-Plex results generator 3.0 (Bio-Rad Laboratories, Veenendaal, the Netherlands). A Bio-Plex II Wash Station (Bio-Rad) with magnetic plate support was used for all washing steps. Incubation of the 6-plex assays was done on a Bühler TiMix 2 shaker (Salm en Kipp, Breukelen, the Netherlands) at room temperature (RT). Beer samples were degassed at RT using the Ultrasonic Cleaner (VWR International, Amsterdam, the Netherlands) at maximum power and centrifuged in an Eppendorf 5810R centrifuge (VWR) equipped with an A-4-62 swinging bucket rotor. All confirmatory analyses of mycotoxins in selected beer samples were done on an AB Sciex (Nieuwerkerk a/d IJssel, the Netherlands) QTRAP 5500 tandem mass spectrometer (MS/MS) equipped with an electrospray ionization (ESI) source, operated in positive and negative multiple reaction monitoring (MRM) mode. The MS system was coupled to a Shimadzu (‘s Hertogenbosch, the Netherlands) Prominence Liquid Chromatography (LC) system, equipped with a Restek (Interscience, Breda, the Netherlands) Ultra Aqueous C18 (100×2.1 mm) column (see Supporting Information (S.I.)). Integration of reconstructed MRM chromatograms was done with MultiQuant V2.0 software using the Signal Finder integration algorithm (AB Sciex). Monoclonal antibodies (mAbs) against AFB1, ZEN, T-2 and DON were purchased from Aokin AG (Berlin, Germany), while the FB1 and OTA mAbs were purchased from Soft Flow Biotechnology Ltd. (Gödöllő, Hungary). The R-Phycoerythrin (RPE)-FB1 and RPE-OTA conjugates were produced in-house using RPE from Moss (Pasadena, MD, USA). The remaining RPE-mycotoxin conjugates were synthesized by Aokin.
Chemicals
Cellstar 96-well culture microtiter plates, 10 and 50 mL tubes were from Greiner (Alphen a/d Rijn, the Netherlands). Sheath fluid (FM3D) and drive fluid (MAGPIX) were both purchased from Luminex (Austin, USA). The following mycotoxins and metabolites were purchased from Biopure (Tulln, Austria): FB1, FB2, FB3, OTA, OTB, AFB1, aflatoxin B2 (AFB2), aflatoxin G1 (AFG1), aflatoxin G2 (AFG2), aflatoxin M1 (AFM1), T-2, HT-2, DON, D3G, 3-acetyl-DON (3ADON), 15-acetyl-DON (15ADON), NIV, ZEN, α-zearalenol (α-ZEL) and β-zearalenol (β-ZEL). Zearalenone-14-β-D-glucopyranoside (Z14G), α-zearalenol-14-β-D-glucopyranoside (α-ZELG), β-zearalenol-14-β-D-glucopyranoside (β-ZELG) and zearalenone 14-sulfate (Z14S) were produced according to [52] or isolated from Fusarium inoculated rice. Syringeless filter devices (Mini-UniPrep, PTFE) for sample clean-up were purchased from GE Healthcare (Rotterdam, the Netherlands). Acetonitrile (ACN) and methanol (MeOH) were purchased from Biosolve (Valkenswaard, the Netherlands), formic acid (FA) from Merck (Amsterdam, the Netherlands) and ammonium formate (AMF) from Fluka Analytical (Steinheim, Germany). All other chemicals were purchased from VWR International (Amsterdam, the Netherlands).
Beer samples
A total of 1000 beer samples, from 42 different countries (Table 2), were selected. Sample collection was mainly random and depending on access and availability, i.e., sampling was not intended to be statistically representative. Primary goal was to collect as much as possible craft beer samples to be able to make a comparison to industrial produced beer and have a more detailed look on craft beer itself. Secondary goal was to cover many different beer styles. At the start of this research, it was still difficult to collect a representative number of international craft beers due to poor availability. Craft beers were mainly collected from bars, restaurants, supermarkets, specialized craft-beer shops and during craft-beer festivals between 2011 and 2014, while industrial beers were mainly collected in supermarkets. Craft beer samples also included vintage beers (beers produced before 2011, often cellared for conditioning). Based on local contacts, additional samples from the USA, China and several African countries were sent to the authors. African beer samples were both commercial (bought in South African supermarkets) and traditional opaque home-brews (collected on site in town villages in Northern South Africa). From each beer, 10 mL was collected in a 50 mL tube and degassed by sonication at maximum power for 10 minutes at RT. After sonication, the beer samples were transferred to a 10 ml tube, centrifuged at 3200g and stored at -20°C. The designated beer styles of these samples were grouped into 20 main beer styles (Table 3).
Screening of beer samples
The mycotoxin 6-plex paramagnetic microsphere immunoassay method used, was a new adaptation of a previously described method [51]. The adapted mycotoxin 6-plex immunoassay involved a simplified extraction method and its performance in beer samples was validated on the planar array imaging platform. The stored beer samples were defrosted, mixed by inversion, and centrifuged for 10 minutes at 3200 g to pellet yeast or other insoluble matter. The supernatant was then diluted 8 fold using methanol-water (1:9 v/v) (10% MeOH). To 40 μL of the diluted beer samples a 10 μL mixture of microspheres, previously coupled with mycotoxin specific monoclonal antibodies, was added followed by the addition of 10 μL of a mixture of mycotoxin specific reporter molecules (mycotoxins coupled to RPE). The final buffer composition in the assay was phosphate buffered saline, 0.02% Tween 20, pH 7.4. Sample and assay components were incubated for 15 minutes, to allow competition between the mycotoxin-RPE conjugates and the free mycotoxins in the samples for antibody interaction. The microspheres were then trapped by a magnet and washed followed by analysis on one of the microsphere dedicated platforms (Fig 1). Since the beer styles investigated are very diverse in composition, it is rather impossible to find a suitable common blank beer. As a practical solution, we selected a dark ale as blank beer for all screening assays, following confirmation of the absence of mycotoxins by LC-MS/MS. This blank beer was used to prepare beer-based multi-mycotoxin calibration curves for AFB1, OTA, ZEN, DON, T-2 and FB1. First, the blank beer was diluted 4-fold with 10% MeOH and subsequently mixed (1:1, v/v) with each standard of the mycotoxin used for the construction of calibration curves, resulting in a final 8-fold dilution of the matrix content and a 2-fold dilution of each standard. Using these multi-mycotoxin calibration curves (S1 Fig), the mycotoxin concentrations in 1000 beer samples were calculated. To this end, the dedicated Bio Plex manager software (Bio-Rad Laboratories, Veenendaal, the Netherlands) built for automated curve and data fitting, was used. The results of the triplicate sample measurements are displayed as a concentration range combined with a standard deviation to give a better idea about the performance of the multiplex method. The mycotoxin monoclonal antibodies (mAbs) in the 6-plex immunoassay were previously tested for cross-reactivity [50, 51]. These results showed, amongst others, cross-reactivity of D3G in the DON immunoassay.
Competition occurs between the free mycotoxins present in the beer and the added mycotoxin-reporter molecules for antibody interaction on the beads. Next, the beads are captured at the bottom of the well by magnetic force. Remaining, non-interacting, assay components are removed by washing. Beads are then measured using a red light source for mycotoxin assay classification and a green light source for quantification of the reporter signal.
Confirmation of screening results by LC-MS/MS
From the 1000 beer samples that were screened with the mycotoxin 6-plex immunoassay, 100 beer samples were selected for confirmatory LC-MS/MS analysis. The first set of samples submitted for confirmatory analysis, was based on mycotoxin contamination results revealed by the 6-plex screening assay. It included mostly high contaminations, as well as some blanks, revealed by the screening. Furthermore, the first selection was also based on covering a wide range of beer styles. Based on the confirmatory results of the first selection, a more detailed selection of new (previously screened) beers was made. The focus was on observed contamination trends in certain beer styles. Since craft beer was the main focus of this survey, we aimed for a total of 70% of craft beers in the final LC-MS/MS selection. For confirmatory analysis we adapted an existing ISO 17025 accredited LC-MS/MS method for feed. This adapted LC-MS/MS method (S1 Text) contained all the relevant mycotoxins, as well as a selection of available mycotoxin metabolites and conjugated forms, relevant to the 6-plex screening method targets (S1 Table). Since the variation in beer matrices is very diverse, especially in craft beer, we chose to use a single point standard addition method for quantification. To this end, 100 μl of degassed beer sample was diluted with 100 μl of the standard solution. The diluted sample was filtered through a syringeless filter device and 5 μl was injected. The limit of quantification (LOQ) was set at 10 times below the standard addition level. The upper quantification limit was set arbitrarily at 2 times above the standard addition level. The concentrations for standard additions can be found in the S1 Table. The limit of detection (LOD) was set at a signal-to-noise ratio of 3:1 based on the peak-to-peak noise around the retention times of the analytes in the reconstructed MRM chromatograms. To this end, we utilized MultiQuant V2.0 software (AB Sciex) using the Signal Finder integration algorithm.
Results and discussion
Scope of the survey
Until now, most published beer surveys for mycotoxins are lacking relevant information on the beer styles (Table 1). Occasionally, information about the country of origin is supplied [29] and sometimes beers are grouped on the basis of their alcohol content [43]. More recently, Varga et al [44] provided detailed information about alcohol content, country of origin and beer style categories. Because of the serious style expansion by craft brewers, we chose to elaborate even further on the beer style categories while also focusing on alcohol content and origin. In our large-scale survey of 1000 beer samples from all over the world (S2 Fig), there was a strong focus on Europe with a total of 787 beers screened. Within Europe the emphasis was on beers from the Netherlands (209), Belgium (203) and Germany (87) (S2 Fig). Furthermore, nearly 60% of all the beers analysed were craft beers. A flow chart overview of the general survey approach is given in Fig 2.
Performance of the 6-plex immunoassay as a screening method
A mycotoxin 6-plex immunoassay, previously applied as a qualitative screening assay for barley [51], was adapted for beer (Fig 1). The extraction protocol was simplified and the method was made suitable for the fast and semi-quantitative detection of mycotoxins in beer. The performance of the adapted 6-plex immunoassay was bench-marked against the previously developed, and in-house validated screening assay for barley [51], by determining the intra- and interday precision for beer samples. The intra- and interday relative standard deviation (%RSD) for multi-mycotoxin fortified samples in dark ale, based on the B/B0 values, were determined for each fortified sample and compared to the previous method (S2 Table). Also in beer, for most mycotoxins, the %RSD values were well below 10%. For T-2 the %RSD values were below 20%. This may indicate that the beer matrix interferes somewhat more with the T-2 assay than barley did. For quantification we prepared multi-mycotoxin dose-response curves in dark ale and checked the variation by comparing intra- and interday IC50s (S3 Table). Based on these dose-response curves, the mycotoxin concentrations in the multi-mycotoxin fortified samples were determined by the Bio Plex manager software. The intra- and interday quantitative precision is displayed in S4 Table. Besides the FB1 assay, which showed overestimations up to a factor 3.9, the other assays performed satisfactory with only slight overestimations with a factor ranging from 1.1 to 1.7. The overestimations in the FB1 assay were not unexpected and reported previously [50]. Moreover, the mycotoxin mAbs used in the 6-plex, in most cases, show cross-reactivity with other metabolites as was reported in detail previously [51]. Therefore, we chose to confirm selected results (see Method section for selection criteria) of the screening assay by an LC-MS/MS based method. The complete mycotoxin 6-plex screening data for all 1000 beers, grouped by beer style groups, can be found in the S5 Table.
Performance of the LC-MS/MS confirmative method
A multi-toxin ISO 17025 accredited LC-MS/MS method for feed was successfully adapted for beer. In order to make the LC-MS/MS method fit for purpose, 23 irrelevant mycotoxins and metabolites were removed from the original 40 targets and 8 new mycotoxin metabolites and conjugated forms of interest were added. The standard addition quantification method was found fit for purpose, following fortification of 5 diverse beer styles (pale lager, adjunct lager, dark ale, sour ale and imperial stout) with all mycotoxins and metabolites. This resulted in adequate detection of precursor and product ions in the respective reconstructed MRM chromatograms. Average variation of LC retention times were well below 0.2 minutes and the deviations in MS/MS product ion intensity ratios, when comparing beer samples to non-matrix-matched standards (in 10% MeOH), were typically below 30% (occasionally below 50%). For DON, higher deviations occurred mainly at low concentrations (< 10 μg/L) and were not linked to any specific beer style. For both methods, the 6-plex immunoassay and the LC-MS/MS, we chose to simply dilute the beer samples rather than to perform concentration or clean-up steps. Therefore, potential matrix effects were only reduced by dilution in both the 6-plex immunoassay and LC-MS/MS. Since the beer styles analysed vary strongly in composition and gravity, variable matrix interference and signal suppression (or enhancement) were to be expected. As a consequence, some mycotoxin standard additions (mainly in imperial stouts) were hardly visible in the reconstructed MRM chromatograms for some of the beer samples. These samples were not considered nor reported. Please note that the sample dilution used for LC-MS/MS is smaller than for the 6-plex immunoassay, since the standard addition method allows correction for matrix effects. The actual volume of the beer sample analysed in the LC-MS/MS is 2.5 μl while the 6-plex screening uses 5 μl.
Occurrence and discussion of specific mycotoxins
Aflatoxins.
The 6-plex immunoassay data suggests the presence of AFB1 in several beer samples with concentrations ranging from 0.1 to 3.7 μg/L (S5 Table). In particular imperial stouts (Table H in S5 Table), as well as some other dark beers, showed AFB1 contaminations. However, when analysed by LC-MS/MS, none out of the 28 selected AFB1 suspected beers could be confirmed. Since the AFB1 immunoassay showed almost no cross-reaction with AFB2, AFG1 or AFG2 [51], contamination with these metabolites was highly unlikely. A possible explanation for the screening results could be the presence of the structurally related sterigmatocystin (STC), whose presence in beer has been reported [53]. However, additional testing showed that STC had no cross-reactivity in the AFB1 assay. Further research is needed to elucidate the origin of the observed suspect screening results, but it seems plausible that matrix effects yielded the false positive results in the immunoassay for these beers. On the other hand, LC-MS/MS analysis revealed 5 beers positive for AFs (Table 4) that were not screened suspect with the 6-plex immunoassay. The 8 times dilution of the sample in the immunoassay may be the reason for this, compared to a 2 times dilution in the LC-MS/MS analysis. From those 5 beers, 4 were African traditional beers and one a pale lager from Zimbabwe. All positive beers were contaminated with AFB1 (0.1–1.2 μg/L) and three of them also contained AFB2 (0.1–0.2 μg/L). In one traditional beer also AFM1 was detected. This may indicate the use of milk or milk derived products in this particular beer. Milk products can be used in certain beer styles (e.g. whey in milk stouts) and therefore can be a source of AFM1 contamination. However, nowadays milk stouts are often produced by the addition of lactose to beer [5]. In this particular sample, being home-brewed, the chance of cross-contamination from other sources (which is common in African domestic brewing) may provide a plausible explanation. AFG1 or AFG2 were not detected in any beer sample. Contaminations of AFs in African traditional beer have been previously reported [19, 20], while AF contaminations in European beers is rarely reported [18]. Occurrence of AFs in beer is of the highest toxicological concern and therefore consumption should be avoided at any time, considering that the IARC classified aflatoxins as carcinogenic to humans (Group 1) [54].
T-2 and HT-2 toxins.
We obtained signals for T-2/HT-2 in many beer samples in the 6-plex immunoassay (S5 Table). However, only in 3 out of the 31 T-2/HT-2 suspected beer samples (and 4 out of all 100 beers submitted for confirmatory analysis) the presence of T-2 or HT-2 toxins was confirmed by LC-MS/MS (Table 5). The highest values being 2.3 μg/L and 3.4 μg/L respectively. These values are lower than those recently found by Rodríguez-Carrasco et al [22]. In their survey 14 (out of 154) samples contained HT-2, all with levels between 24.2–38.2 μg/L. All those samples, from 2013, came from Germany and were from the wheat beer style. In our case, the presence of T-2 and HT-2 (as determined by LC-MS/MS) seemed not style nor country/region dependent. Our results suggest that several suspect immunoassay screening results may indicate either false positives or possible modified forms of T-2 and/or HT-2. Regular consumption of these beers will not lead to exceedance of the TDI of 0.1 μg/ kg BW for the sum of T-2 and HT-2 [55] (14 μg/L for a person of 70 kg BW drinking one 0.5 L bottle of beer per day). Beer sample 356, an imperial stout, had the highest T-2/HT-2 contamination (57 μg/L) in the 6-plex immunoassay, but this was not confirmed by LC-MS/MS. In preliminary follow-up research, we analysed this sample using high resolution LC-MS/MS analysis and found indications that two forms of HT-2 glycosides may be present in this beer sample (results not shown). It has been reported previously [56] that HT-2 glycosides were present in wheat. Due to the lack of proper standards, quantification of these HT-2 glycosides was not possible. For the same reason the cross-reaction of HT-2 glycosides in the 6-plex immunoassay could not be determined, making it impossible to verify if the T-2/HT-2 suspect results are based on the presence of glycosides or other possible conjugated forms. The metabolites T-2 triol and T-2 tetraol were not detected in any beer sample and therefore not the cause of the suspect screening results.
Zearalenone.
For ZEN, concentrations up to 5.6 μg/L were found in the 6-plex immunoassay (S5 Table). From the 3 selected ZEN suspect beers, 1 was confirmed as positive when analysed further by the LC-MS/MS method. LC-MS/MS analysis of selected negative immunoassay samples revealed 6 additional ZEN contaminations, although 5 out of 6 results were below the LOQ of 0.3 μg/L (Table 6). Also β-ZEL was found in 4 samples, in all cases below the LOQ of 2 μg/L and in two cases co-occurring with ZEN. Z14S was detected in 12 beer samples in Table 6. From those samples, 3 were above the LOQ (0.5 μg/L) while 9 samples had concentrations between the LOQ and LOD. Z14S was not found in African traditional beers. Z14S co-occurred with ZEN four times and in 2 beer samples ZEN, β-ZEL and Z14S co-occurred. In a previous survey, including conjugated mycotoxins, Z14S was not detected in beer [57]. α-ZEL, α-ZELG, β-ZELG and Z14G were not detected in any beer sample. ZEN was previously detected in high concentrations (up to 426 μg/L) in African traditional beers [20]. Using immunoassays, Bauer et al [25] and Kuzdraliński et al [24] both detected ZEN in beers (concentrations up to 2.0 μg/L), but these were not confirmed by instrumental analysis. Therefore it remains unclear whether ZEN metabolites were contributing to the ZEN values reported. The ZEN mAb in our 6-plex screening assay showed no cross-reactions to Z14S. None of the contaminated samples in our survey would lead to exceedance of the TDI for ZEN, set by EFSA at 0.25 μg/kg BW [58], under normal circumstances (35 μg/L for a person of 70 kg BW drinking one 0.5 L bottle of beer per day). This also counts if we add up the sum of all ZEN metabolites in a single sample. Therefore ZEN is not a major risk factor in the confirmed beers.
Ochratoxin A.
The 6-plex immunoassay assay showed several OTA suspect samples with indicative levels ranging from 0.1–1.6 μg/L (S5 Table). From the 25 selected OTA suspect samples, 6 samples were confirmed positive by LC-MS/MS and 5 samples had OTA concentrations ranging from 0.3–0.6 μg/L (Table 7). Remarkably, from these confirmed OTA positives, 4 beers were from the same Norwegian brewery. The other beers confirmed positive were from England, but originated from different breweries. Note that sampling of all these positive craft beers had occurred in the same year (2011) and at the same craft beer festival. The OTA contaminations found in our survey were in beers from European origin. They were slightly higher than those previously found (in European beers) by Visconti et al [28] and Bertuzzi et al [29] but considerably lower than the contaminations found by Odhav en Naicker [20] in African traditional beers. OTB was not detected in any beer sample. OTA is of high toxicological concern, since it is possibly carcinogenic to humans (Group 2B) [59]. In 2006, EFSA established a Tolerable Weekly Intake (TWI) for OTA of 120 ng/kg BW per week which can be translated to an average of 17 ng/kg BW per day [60]. The beers in our survey confirmed positive for OTA do not surpass this derived TDI under normal circumstances (2.4 μg/L for a person of 70 kg BW drinking one 0.5 L bottle of beer per day).
Fumonisins.
From previous research it is known that the 6-plex immunoassay overestimates the fumonisin content [51]. As a result, several false suspects became apparent when we compared the FBs immunoassay data (S5 Table) with the LC-MS/MS data. Higher false suspect concentrations were typically found in darker style beers (e.g. imperial stouts, dark lagers and ales). Confirmatory analysis showed that pale lagers contaminated with FB1 were mainly from Spain and Italy (Table 8). In some countries, mostly for economic reasons, pale lagers often contain other cereals besides barley. The declared information on the label of some of these supermarket beers purchased in Spain and Italy, revealed that they contain maize as an adjunct. In fact, Italian and Spanish beers brewed with higher amounts of barley (or 100% barley) are often considered specialty beers in these countries. The use of maize in pale lagers (or any other beer) increases the risk of contamination with FBs. These data may suggest a Mediterranean trend. However, in a preliminary screening of available Greek commercial and craft beers in 2015, we did not find any pale lager that contained FBs (results not published). The highest LC-MS/MS FB1 contamination detected in our survey was for an Italian pale lager (51 μg/L), followed by a Spanish non-alcoholic beer that contained (28 μg/L). In only 4 beer samples we were able to detect FB3 and in 2 beer samples this was above the LOQ (1 μg/L). FB2 was not detected in any beer sample. Fumonisin contaminations of Italian beers (30 μg/L) and Spanish beers (85 μg/L) were reported previously [29, 39]. Besides pale lagers, mainly African traditional home-brews were prone to FB1 contamination. The highest contamination in that category was 36 μg/L with another two beers close to this contamination level (30 and 28 μg/L respectively) (Table 8). There are agro-ecological and cultural reasons for this. First, most of Africa is hot and humid thus ideal for Fusarium infection and growth. Further, Shepard et al [35] have shown that in the Eastern Cape, the best maize is selected for cooking while the mouldy maize is then used for brewing beer. It is believed that infected maize adds a desirable taste to the final beer. Beers from two Spanish breweries were sampled again approximately 2 years later, and analysed only for FBs using LC-MS/MS (S6 Table). Like the previous results, the cheapest pale lagers (from brewery #2) had the highest FB1 (56 μg/L) and total FB contamination (69 μg/L). Beers from brewery #1, showed lower FB contaminations this time (14 and 17 μg/L respectively) at the sampling two years later. Besides FB1, all beers contained FB2 and FB3 in this reassessment. With the FB concentrations found in our survey, the TDI is not easily exceeded. EFSA has set a group TDI of 2 μg/kg BW per day (sum of FB1, FB2 and FB3) [61]. If we take into account an average body weight of 70 kilogram for an adult [62], then a person would need to drink more than 2 litres per day of the highest contaminated beer (69 μg/L) before reaching the TDI [61]. At that consumption level, alcohol intake is definitely a more serious risk. Nevertheless, daily exposure to FB1 through beer should be avoided as much as possible, since consumers may be exposed to other dietary sources of FBs as well. FBs are of high toxicological concern, since they are possibly carcinogenic to humans (Group 2B) [59]. The incidence of human oesophageal cancer and the occurrence of Fusarium verticillioides (and its mycotoxins FBs), has been associated with regions where corn is produced and consumed as staple food [63]. Franceschi et al [64] reported significant associations, in males, between maize consumption and oral cancer in northern Italy. In a case-control study, 80% of the patients diagnosed with oesophageal cancer indicated to be regular consumers of African traditional beers. Based on these findings, Segal et al [65] concluded that the consumption of these African traditional beers was a major risk factor. However, the African traditional beers analysed in our survey, did not have that extreme FBs contaminations compared to those previously reported in literature [19, 35].
Type B trichothecenes: DON, D3G, ADONs and NIV.
The 6-plex immunoassay reports results for the sum of DON and D3G (S5 Table). This is beneficial since previous research seems to indicate [66–69] that D3G is of toxic relevance. Recently a request was made to EFSA for a scientific opinion on the risks for animal and human health related to the presence of deoxynivalenol, its metabolites and masked deoxynivalenol (D3G) in food and feed [70]. Therefore, it is plausible that D3G will be added to the total DON group (of DON and its acetylated derivatives) for risk assessment. The majority of the beers screened (60%) had contamination levels below 10 μg/L of DON+D3G, while beers with contaminations above 100 μg/L occurred less frequent (6%) (Fig 3). From the 406 beers that have DON+D3G contaminations above 10 μg/L, 73% were craft beers and these had a higher average contaminations (63 μg/L) compared to industrial produced beers (39 μg/L). The popular craft beer style imperial stout did not follow this trend. Only 17% of all imperial stouts screened had DON+D3G contamination levels below 10 μg/L, while 29% had DON+D3G contaminations above 100 μg/L (Fig 3). The highest overall DON+D3G contaminations were present in imperial stout, eisbock and stout (475, 308 and 169 μg/L, respectively) beers. The highest average group DON+D3G contaminations, based on beer style, were imperial stout, eisbock and African traditional with 86, 81 and 65 μg/L respectively, while the saison, pale lager and non/low alcohol beer styles had the lowest average contaminations (19, 23 and 23 μg/L, respectively) (S7 Table). The highest contamination incidences were found in eisbock, imperial stout and dark lager (83, 83 and 68%, respectively) (S7 Table), while the lowest contamination incidences were found in the sour ales, saison and pale lager beer styles (7, 8 and 13% respectively (S7 Table). The screening results revealed a clear correlation between the alcohol content (%ABV) and the DON+D3G contamination (Fig 4). The same positive correlation was reported in previous mycotoxin beer surveys [40, 43, 44]. For high %ABV beers, a higher input of grains is needed to deliver the fermentable sugars and with that comes a higher risk of mycotoxin contamination. LC-MS/MS data showed that, within the selected 100 beers for confirmation, 26 beers were contaminated with only DON above the corresponding LOD and 13 beers with only D3G. In contrast, in 38 beers both DON and D3G were detected. In 19 beers the concentration of D3G was higher than that of DON (Table 9). Varga et al [44] previously reported molar D3G/DON ratios (corrected for molecular mass) between 0.11 and 1.25 with an average of 0.56. In our survey, the molar D3G/DON ratios ranged from 0.10 to 2.60 with an average of 0.79. The highest ratio observed was for a pale lager from Poland. Generally, in this survey, beers having a D3G/DON ratio higher than 0.60, are almost all craft beers. DON was also present in African traditional beers, but in these beers no D3G contaminations were observed. The absence of D3G in these beers, may indicate that sorghum malt, often used in these traditional beer styles [71], does not have the potential to conjugate DON to D3G. Like the screening assay, LC-MS/MS analysis showed that beer sample 183 had the highest contamination for both DON and D3G with contaminations of 412 and 619 μg/L respectively. Since these concentrations were more than 2 times higher than the standard addition, we decided to reanalyze this particular beer sample following a 10 times dilution. Reanalysis showed that it contained 309 μg/L of DON and 535 μg/L of D3G. These concentrations reconfirmed that this imperial stout had the most extreme contamination in the entire survey. Recently, Piacentini et al [36] surveyed Brazilian craft beers using liquid chromatography with fluorescence detection and found beers with high DON contaminations (17 samples, range 127–501 μg/L). Unfortunately the surveyed beer styles were not further defined than ales and lagers. Previously, DON concentrations as high as 501 μg/L had only been reported in African traditional beers [34, 45]. The LC-MS/MS method used was not able to distinguish between 3ADON and 15ADON (ADONs) and therefore the ADON contaminations found should be considered as the sum of both (Table 9). In 5 African traditional beers ADONs were detected and in 3 of those also NIV was present. All these mycotoxin levels were below the LOQ (10 μg/L for ADONs and 5 μg/L for NIV). ADONs were also detected in one pale ale (Poland) and 2 pale lagers (Italy), again all below the LOQ. In one Dutch pale ale, NIV was detected at a concentration of 21 μg/L. These results indicate that NIV and ADONs are not frequent contaminants of beers, with the exception of African traditional beers (5 out of 14). Previously, Kostelanska et al [43] found ADONs concentrations as high as 25 μg/L and described them as common contaminants present in 50% of 176 beers that were analysed. On the other hand, Bertuzzi et al [29] could not find any ADONs in 106 beer samples analysed while Varga et al [44] did not find any 3ADON in 374 beer samples surveyed.
Taking into account that the DON mAb used in the 6-plex screening assay has 60% cross-reaction to D3G, we compared the 6-plex and the LC-MS/MS data. In 23 beers, both in 6-plex and LC-MS/MS, no DON+D3G was detected. From the 77 beers confirmed positive for DON and/or D3G (Table 9) by LC-MS/MS, 14 beers were negative in the 6-plex immunoassay. Concentrations for DON+D3G in these samples were generally low with the exception of African traditional beer sample 416 (68 μg/L) suggesting a beer specific interference in the immunoassay. For 27 beers, the 6-plex values for DON+D3G were below those found by LC-MS/MS analysis with an average factor of 0.7. For imperial stouts (9 beers) this factor was the same. For 32 beers, the 6-plex values for DON+D3G were above those found by LC-MS/MS. In average, the immunoassay values were a factor 2.1 higher. This value excludes beer 57, a fruit/beer mix, which showed a 24 times overestimation. After reanalysis, with both the 6-plex immunoassay and LC-MS/MS, this beer still showed the same overestimation. The addition of grapefruit juice to this beer seems to be responsible for high matrix interference in the 6-plex assay. For the imperial stouts (9 beers) within the group of 32 beers, the average 6-plex values for DON+D3G were a factor 2.3 higher when compared to LC-MS/MS. This was mainly attributed to a few extremes (Table 9).
Imperial stouts show the highest DON+D3G contaminations in our survey, and since both USA and European imperial stouts were well presented, a geographical comparison was made. In total, 52 imperial stouts from the USA and 74 imperial stouts from Europe were screened. The mean DON+D3G contaminations (based on beers with contaminations higher than 10 μg/L) were 93 μg/L and 64 μg/L respectively. This suggests that USA imperial stouts have higher DON+D3G contaminations compared to European ones. This may be attributed to malt usage. Imperial stout is a high gravity style and is mostly pitch-black because of the specific taste-defining malts used. These malts are often, but not limited to, brown malt, caramel malt, chocolate malt and roast malt. This may suggest that these colored malts are responsible for the high DON+D3G contributions, since the strong pale lagers (e.g. barley wines) seem to suffer less of high DON+D3G contaminations compared to the imperial stouts. When comparing identical styles divided by lighter and darker colors (pale vs. dark) (S3 Fig), dark lagers clearly have higher DON+D3G contaminations compared to pale lagers. Dark ales tend to have higher contaminations (higher than 50 μg/L) compared to pale ales, but pale ales have more contaminations higher than 25 μg/L. Comparison of strong pale lagers and strong dark lagers shows minor differences. These comparisons show that for some beer styles higher contaminations can be associated to beer color.
The DON+D3G contaminations in 27 beers from this survey, are equal to, or exceed the TDI of 1μg/kg BW for DON (140 μg/L for a person of 70 kg BW drinking one 0.5 L bottle of beer per day). Personal risk, related to exceeding the TDI for DON, based on the DON+D3G contaminations for selected beers in this survey, is presented in Table 10. Additionally, the consumption of multiple bottles and the likely additional exposure via the daily diet (bread, pasta, breakfast cereals) further increases the risk. Unlike FBs, AFs and OTA, DON is not grouped as a (possible) carcinogenic mycotoxin. Its intake causes symptoms like vomiting, nausea, growth retardation, reproductive disorders and suppression of the immune system in humans and animals [72]. More recently DON is also believed to be active at the central nervous system level (brain) causing modified neurochemistry and neuronal activity [73].
Conclusion
To our best knowledge, this survey is the largest ever performed for the occurrence of mycotoxins in beer. It is for certain the most extensive screening for mycotoxins in craft beers to date. The applied mycotoxin 6-plex screening method facilitated fast and easy screening of 1000 global beer samples, whilst the developed beer-dedicated LC-MS/MS method proved to be very useful for quantitative confirmatory analysis. The effectiveness of the 6-plex mycotoxin immunoassay screening approach, without any sample clean-up, was hampered by matrix interferences. This occurred particularly at low concentrations and certain beer styles. It caused false suspect samples for AFB1, FBs, T-2/HT-2 mainly in dark beer styles. For type B trichothecenes the chosen approach lead to over- and underestimations, particularly in a few imperial stouts. Therefore, a blank reference beer for imperial stouts is desired in the 6-plex assay. It will help to improve mycotoxin determination in this complex beer style. But still, even then large matrix background variations can be expected. Older recipes just contain malts while newer recipes show the addition of coffee, cacao and other adjuncts. For a further reduction of over- and underestimations, a suitable clean-up procedure may be considered for future 6-plex screening of mycotoxins in beer. Additionally, averaging data from replicates in the LC-MS/MS standard addition method may lead to a better quantitative comparison. Furthermore, the use of logarithmic dose-response curves in the 6-plex immunoassay, compared to a narrow linear range for the standard additions used in LC-MS/MS, will always contribute to a less accurate quantification in the 6-plex immunoassay.
Until now, there are no Maximum Levels set for the occurrence of mycotoxins in beer. We agree with the conclusion drawn previously by Varga et al [44] about their extensive DON survey in beer, stating that setting maximum levels for DON and its metabolites in beer helps to protect beer drinkers from consuming highly contaminated beers. Based on beer samples in the presented survey, exceeding the TDI for DON, future research should additionally focus on malts and/or grains used for high gravity beer styles like imperial stout. It seems that the darker malts and/or roasted malts in imperial stouts, combined with the high gravity, contribute to the high DON+D3G levels. Color comparison of similar styles, with pale and dark varieties, partially supports this hypothesis. Current malt certificates are often lacking information, therefore, it is suggested that malts used in high gravity beer styles should be analysed in more detail, especially for DON+D3G. With that detailed information brewers should be able to judge until what gravity the malts are safe to use. It is proposed that stricter maximum mycotoxin levels, or better specified levels, are applied for malts that are used for brewing high gravity beers. Furthermore, small craft breweries should consider the implementation of cheap, reliable, easy and fast on-site mycotoxin assays to control the purchased malts and adjuncts as well as their final products. In the end, on-site mycotoxin testing may not always be feasible for small starting breweries. Therefore they should be able to rely on the proper control of the purchased malts. With the craft beer market consistently expanding, and with many craft breweries producing imperial stouts, quality control management seems a necessary step.
Supporting information
S1 Fig. Multi-mycotoxin dose-response curves in dark ale.
https://doi.org/10.1371/journal.pone.0185887.s002
(PDF)
S2 Fig.
A) Global beer samples surveyed by region B) European beer samples surveyed by country.
https://doi.org/10.1371/journal.pone.0185887.s003
(PDF)
S3 Fig. Comparison of DON+D3G contaminations in similar beer styles.
https://doi.org/10.1371/journal.pone.0185887.s004
(PDF)
S1 Table. Mycotoxin specific MS/MS settings and standard addition levels.
https://doi.org/10.1371/journal.pone.0185887.s005
(PDF)
S2 Table. Intra- and interday precision of the 6-plex assay.
https://doi.org/10.1371/journal.pone.0185887.s006
(PDF)
S3 Table. Intra- and interday IC50s based on multi-mycotoxin dose-response curves.
https://doi.org/10.1371/journal.pone.0185887.s007
(PDF)
S4 Table. Intra- and interday determinations of mycotoxin concentrations using the 6-plex assay.
https://doi.org/10.1371/journal.pone.0185887.s008
(PDF)
S5 Table.
6-plex immunoassay screening data for A) African traditional beers B) bock beers C) dark ale beers D) dark lager beers E) double India pale ale beers F) eisbock beers G) fruit/vegetable/spice beers H) imperial stout beers I) India pale ale beers J) non/low alcohol beers K) pale ale beers L) pale lager beers M) saison beers N) smoked beers O) sour ale beers P) stout beers Q) strong dark ale beers R) strong pale ale beers S) strong pale lager T) wheat beers.
https://doi.org/10.1371/journal.pone.0185887.s009
(PDF)
S6 Table. Reassessment of Spanish breweries for FB contaminations by LC-MS/MS analysis.
https://doi.org/10.1371/journal.pone.0185887.s010
(PDF)
S7 Table. DON+D3G contamination per beer style group.
https://doi.org/10.1371/journal.pone.0185887.s011
(PDF)
Acknowledgments
The authors would like to thank Diana Chan, previously a student of Nottingham Trent University, for her contribution to the beer screening. Ron Hoogenboom of RIKILT Wageningen University & Research (Institute of Food Safety) and Martien Spanjer of the Netherlands Food and Consumer Product Safety Authority (NVWA) for critical reading of the manuscript. Menno Olivier of Brouwerij de Molen for constructive discussions regarding craft brewing and the supply of relevant information. Milena Zachariasova of the Institute of Chemical Technology, Prague for sending a selection of Czech beer samples.
References
- 1.
https://www.brewersassociation.org accessed December 2016
- 2.
http://www.cambrinus.nl accessed December 2016
- 3.
Protz R. CAMRA's good beer guide. 41st ed. St Albans: CAMRA books; 2014
- 4.
https://beerconnoisseur.com/articles/strange-brews-10-weirdest-beer-ingredients accessed May 2016
- 5.
Oliver G. The Oxford Companion to Beer. 1st ed. Oxford: Oxford University Press; 2011
- 6.
http://www.ratebeer.com accessed December 2016
- 7. Fink-Gremmels J (1999) Mycotoxins: their implications for human and animal health. Vet Q 21: 115–120. pmid:10568000
- 8. Peraica M, Radić B, Lucić A, Pavlović M (1999) Toxic effects of mycotoxins in humans. B World Health Organ 77: 754–766.
- 9. Zinedine A, Soriano JM, Molto JC, Manes J (2007) Review on the toxicity, occurrence, metabolism, detoxification, regulations and intake of zearalenone: an oestrogenic mycotoxin. Food Chem Toxicol 45: 1–18. pmid:17045381
- 10. Tralamazza SM, Bemvenuti RH, Zorzete P, de Souza Garcia F, Correa B (2016) Fungal diversity and natural occurrence of deoxynivalenol and zearalenone in freshly harvested wheat grains from Brazil. Food Chem 196: 445–450. pmid:26593513
- 11. Mngqawa P, Shephard GS, Green IR, Ngobeni SH, de Rijk TC, Katerere DR (2015) Mycotoxin contamination of home-grown maize in rural northern South Africa (Limpopo and Mpumalanga Provinces). Food Addit Contam Part B Surveill 9: 38–45. pmid:26600208
- 12. Fredlund E, Gidlund A, Sulyok M, Borjesson T, Krska R, Olsen M, et al. (2013) Deoxynivalenol and other selected Fusarium toxins in Swedish oats—occurrence and correlation to specific Fusarium species. Int J Food Microbiol 167: 276–283. pmid:23962918
- 13. Muller HM, Reimann J, Schumacher U, Schwadorf K (1997) Natural occurrence of Fusarium toxins in barley harvested during five years in an area of southwest Germany. Mycopathologia 137: 185–192. pmid:16283459
- 14. Mateo EM, Gil-Serna J, Patino B, Jimenez M (2011) Aflatoxins and ochratoxin A in stored barley grain in Spain and impact of PCR-based strategies to assess the occurrence of aflatoxigenic and ochratoxigenic Aspergillus spp. Int J Food Microbiol 149: 118–126. pmid:21741104
- 15. Maenetje PW, Dutton MF (2007) The incidence of fungi and mycotoxins in South African barley and barley products. J Environ Sci Health B 42: 229–236. pmid:17365338
- 16. Nakajima M, Tsubouchi H, Miyabe M (1999) A survey of ochratoxin A and aflatoxins in domestic and imported beers in Japan by immunoaffinity and liquid chromatography. J AOAC Int 82: 897–902. pmid:10444830
- 17. Mably M, Mankotia M, Cavlovic P, Tam J, Wong L, Pantazopoulos P, et al. (2005) Survey of aflatoxins in beer sold in Canada. Food Addit Contam 22: 1252–1257. pmid:16356889
- 18. Burdaspal PA, Legarda TM (2013) Survey on aflatoxin in beer sold in Spain and other European countries. World Mycotoxin J 6: 93–101.
- 19. Matumba LU, Van Poucke CU, Biswick T, Monjerezi M, Mwatseteza J, de Saeger S (2014) A limited survey of mycotoxins in traditional maize based opaque beers in Malawi. Food Control 36: 253–256
- 20. Odhav B, Naicker V (2002) Mycotoxins in South African traditionally brewed beers. Food Addit Contam 19: 55–61. pmid:11811766
- 21. Yoshinari T, Takeuchi H, Aoyama K, Taniguchi M, Hashiguchi S, Kai S, et al. (2014) Occurrence of four Fusarium mycotoxins, deoxynivalenol, zearalenone, T-2 toxin, and HT-2 toxin, in wheat, barley, and Japanese retail food. J Food Prot 77: 1940–1946. pmid:25364928
- 22. Rodríguez-Carrasco Y, Fattore M, Albrizio S, Berrada H, Mañes J (2015) Occurrence of Fusarium mycotoxins and their dietary intake through beer consumption by the European population. Food Chem 178: 149–155. pmid:25704695
- 23. Zöllner P, Berner D, Jodlbauer J, Lindner W (2000) Determination of zearalenone and its metabolites alpha- and beta-zearalenol in beer samples by high-performance liquid chromatography-tandem mass spectrometry. J Chromatogr B Biomed Sci Appl 738: 233–241. pmid:10718641
- 24. Kuzdraliński A, Solarska E, Muszyńska M (2013) Deoxynivalenol and zearalenone occurence in beers analysed by an enzyme-linked immunosorbent assay method. Food Control 29: 22–24.
- 25. Bauer JI, Gross M, Gottschalk C, Usleber E (2016) Investigations on the occurrence of mycotoxins in beer. Food Control 63: 135–139.
- 26. Okoye ZSC (1986) Zearalenone in native cereal beer brewed in Jos metropolis of Nigeria. J Food Safety 7: 233–239.
- 27. Nkwe DO, Taylor JE, Siame BA (2005) Fungi, aflatoxins, fumonisin Bl and zearalenone contaminating sorghum-based traditional malt, wort and beer in Botswana. Mycopathologia 160: 177–186. pmid:16170615
- 28. Visconti A, Pascale M, Centonze G (2000) Determination of ochratoxin A in domestic and imported beers in Italy by immunoaffinity clean-up and liquid chromatography. J Chromatogr A 888: 321–326. pmid:10949498
- 29. Bertuzzi T, Rastelli S, Mulazzi A, Donadini G, Pietri A (2011) Mycotoxin occurrence in beer produced in several European countries. Food Control 22: 2059–2064.
- 30. Gumus T, Arici M, Demirci M (2004) A survey of barley, malt and beer contamination with ochratoxin A in Turkey. J I Brewing 110: 146–149.
- 31. Medina Á, Jiménez M, Gimeno-Adelantado JV, Valle-Algarra FM, Mateo R (2005) Determination of ochratoxin A in beer marketed in Spain by liquid chromatography with fluorescence detection using lead hydroxyacetate as a clean-up agent. J. Chromatogr. A 1083: 7–13. pmid:16078682
- 32. Wu J, Tan Y, Wang Y, Xu R (2011) Occurrence of ochratoxin A in wine and beer samples from China. Food Addit Contam Part B Surveill 4: 52–56. pmid:24779662
- 33. Lasram S, Oueslati S, Chebil S, Mliki A, Ghorbel A (2013) Occurrence of ochratoxin A in domestic beers and wines from Tunisia by immunoaffinity clean-up and liquid chromatography. Food Addit Contam Part B Surveill 6: 1–5. pmid:24786617
- 34. Roger DD (2011) Deoxynivanol (DON) and fumonisins B1 (FB1) in artisanal sorghum opaque beer brewed in north Cameroon. Afr J Microbiol Res 5: 1565–1567.
- 35. Shephard GS, van der Westhuizen L, Gatyeni PM, Somdyala NI, Burger HM, Marasas WF (2005) Fumonisin mycotoxins in traditional Xhosa maize beer in South Africa. J Agric Food Chem 53: 9634–9637. pmid:16302789
- 36. Piacentini KC, Savi GD, Olivo G, Scussel VM (2015) Quality and occurrence of deoxynivalenol and fumonisins in craft beer. Food Control 50: 925–929.
- 37. Cano-Sancho G, Ramos AJ, Marin S, Sanchis V (2012) Occurrence of fumonisins in Catalonia (Spain) and an exposure assessment of specific population groups. Food Addit Contam Part A Chem Anal Control Expo Risk Assess 29: 799–808. pmid:22273497
- 38. Hlywka JJ, Bullerman LB (1999) Occurrence of fumonisin B1 and B2 in beer. Food Addit Contam 16: 319–324. pmid:10645345
- 39. Torres MR, Sanchis V, Ramos AJ (1998) Occurrence of fumonisins in Spanish beers analyzed by an enzyme-linked immunosorbent assay method. Int J Food Microbiol 39: 139–143. pmid:9562886
- 40. Papadopoulou-Bouraoui A, Vrabcheva T, Valzacchi S, Stroka J, Anklam E (2004) Screening survey of deoxynivalenol in beer from the European market by an enzyme-linked immunosorbent assay. Food Addit Contam 21: 607–617. pmid:15204540
- 41. Zachariasova M, Hajslova J, Kostelanska M, Poustka J, Krplova A, Cuhra P, et al. (2008) Deoxynivalenol and its conjugates in beer: a critical assessment of data obtained by enzyme-linked immunosorbent assay and liquid chromatography coupled to tandem mass spectrometry. Anal Chim Acta 625: 77–86. pmid:18721543
- 42. Zachariasova M, Vaclavikova M, Lacina O, Vaclavik L, Hajslova J (2012) Deoxynivalenol Oligoglycosides: New “Masked” Fusarium Toxins Occurring in Malt, Beer, and Breadstuff. J Agric Food Chem 60: 9280–9291. pmid:22897145
- 43. Kostelanska M, Hajslova J, Zachariasova M, Malachova A, Kalachova K, Poustka J, et al. (2009) Occurrence of Deoxynivalenol and Its Major Conjugate, Deoxynivalenol-3-Glucoside, in Beer and Some Brewing Intermediates. J Agric Food Chem 57: 3187–3194. pmid:19301815
- 44. Varga E, Malachova A, Schwartz H, Krska R, Berthiller F (2013) Survey of deoxynivalenol and its conjugates deoxynivalenol-3-glucoside and 3-acetyl-deoxynivalenol in 374 beer samples. Food Addit Contam Part A Chem Anal Control Expo Risk Assess 30: 137–146. pmid:23025486
- 45. Kirui MC, Alakonya AE, Talam KK, Tohru G, Bii CC (2014) Total aflatoxin, fumonisin and deoxynivalenol contamination of busaa in Bomet county, Kenya. Afr J Biotechnol 13: 2675–2678.
- 46.
Hopsteiner newsletter February 2011 http://www.mbaa.com/brewresources/foodsafety/Documents/Mycotoxins%20in%20hops.pdf accessed December 2016
- 47.
Webb T, Beaumont S, The World Atlas of Beer 1st ed. London: Octopus Publishing Group; 2012
- 48. Medina A, Rodriguez A, Magan N (2014) Effect of climate change on Aspergillus flavus and aflatoxin B1 production. Front Microbiol 5: 1–7.
- 49. Uhlig S, Eriksen GS, Hofgaard IS, Krska R, Beltrán E, Sulyok M (2013) Faces of a Changing Climate: Semi-Quantitative Multi-Mycotoxin Analysis of Grain Grown in Exceptional Climatic Conditions in Norway. Toxins 5: 1682–1697. pmid:24084167
- 50. Peters J, Thomas D, Boers E, de Rijk T, Berthiller F, Haasnoot W, et al. (2013) Colour-encoded paramagnetic microbead-based direct inhibition triplex flow cytometric immunoassay for ochratoxin A, fumonisins and zearalenone in cereals and cereal-based feed. Anal Bioanal Chem 405: 7783–7794. pmid:23760139
- 51. Peters J, Cardall A, Haasnoot W, Nielen MWF (2014) 6-Plex microsphere immunoassay with imaging planar array detection for mycotoxins in barley. Analyst 139: 3968–3976. pmid:24919458
- 52. Berthiller F, Hametner C, Krenn P, Schweiger W, Ludwig R, Adam G, et al. (2009) Preparation and characterization of the conjugated Fusarium mycotoxins zearalenone-4O-β-D-glucopyranoside, α-zearalenol-4O-β-D-glucopyranoside and β-zearalenol-4O-β-D-glucopyranoside by MS/MS and two-dimensional NMR. Food Addit Contam Part A 26: 207–213.
- 53. Veršilovskis A, de Saeger S, Mikelsone V (2008) Determination of sterigmatocystin in beer by high performance liquid chromatography with ultraviolet detection. World Mycotoxin J 1: 161–166.
- 54. WHO (1993) Some Naturally Occurring Substances: Food Items and Constituents, Heterocyclic Aromatic Amines and Mycotoxins. IARC Monogr Eval Carcinog Risks Hum 56: 246–395.
- 55. EFSA (2011) Scientific Opinion on the risks for animal and public health related to the presence of T-2 and HT-2 toxin in food and feed. EFSA J 9: 1–187.
- 56. Lattanzio VM, Visconti A, Haidukowski M, Pascale M (2012) Identification and characterization of new Fusarium masked mycotoxins, T2 and HT2 glycosyl derivatives, in naturally contaminated wheat and oats by liquid chromatography-high-resolution mass spectrometry. J Mass Spectrom 47: 466–475. pmid:22689622
- 57. Vendl O, Crews C, MacDonald S, Krska R, Berthiller F (2010) Occurrence of free and conjugated Fusarium mycotoxins in cereal-based food. Food Addit Contam: Part A 27: 1148–1152.
- 58. EFSA (2011) Scientific opinion on the risks for public health related to the presence of zearalenone in food. EFSA J 9: 1–124.
- 59. WHO (2002) IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Some Traditional Herbal Medicines, Some Mycotoxins, Naphthalene and Styrene. IARC Monogr Eval Carcinog Risks Hum 82: 1–556. pmid:12687954
- 60. EFSA (2006) Opinion of the Scientific Panel on contaminants in the Food Chain of the EFSA on a request from the Commission related to ochratoxin A in food. EFSA J 365: 1–56.
- 61.
SCF (2003) Updated opinion of the Scientific Committee on Food on Fumonisin B1, B2 and B3.http://ec.europa.eu/food/fs/sc/scf/out185_en.pdf Accessed May 2016 https://ec.europa.eu/food/sites/food/files/safety/docs/cs_contaminants_catalogue_fusarium_out185_en.pdf Accessed December 2016
- 62. EFSA (2012) Guidance on selected default values to be used by the EFSA Scientific Committee, Scientific Panels and Units in the absence of actual measured data. EFSA J 10; 1–32.
- 63. Stockmann-Juvala H, Savolainen K (2008) A review of the toxic effects and mechanisms of action of fumonisin B1. Hum Exp Toxicol 27: 799–809. pmid:19244287
- 64. Franceschi S, Bidoli E, Baron AE, La Vecchia C (1990) Maize and risk of cancers of the oral cavity, pharynx, and esophagus in northeastern Italy. J Natl Cancer Inst 82: 1407–1411. pmid:2388291
- 65. Segal I, Reinach SG, de Beer M (1988) Factors associated with oesophageal cancer in Soweto, South Africa. Br J Cancer 58: 681–686. pmid:3219281
- 66. Berthiller F, Krska R, Domig KJ, Kneifel W, Juge N, Schumacher R, et al. (2011) Hydrolytic fate of deoxynivalenol-3-glucoside during digestion. Toxicol Lett 206: 264–267. pmid:21878373
- 67. Herías MV, Hessle C, Telemo E, Midtvedt T, Hanson LÅ, Wold AE (1999) Immunomodulatory effects of Lactobacillus plantarum colonizing the intestine of gnotobiotic rats. Clin Exp Immunol 116: 283–290. pmid:10337020
- 68. Nagl V, Woechtl B, Schwartz-Zimmermann HE, Hennig-Pauka I, Moll WD, Adam G, et al. (2014) Metabolism of the masked mycotoxin deoxynivalenol-3-glucoside in pigs. Toxicol Lett 229: 190–197. pmid:24968060
- 69. Gratz SW, Duncan G, Richardson AJ (2013) The human fecal microbiota metabolizes deoxynivalenol and deoxynivalenol-3-glucoside and may be responsible for urinary deepoxy-deoxynivalenol. Appl Environ Microbiol 79: 1821–1825. pmid:23315729
- 70.
EFSA (2013) M-2013-0260 Request for a scientific opinion on the risks for animal and public health related to the presence of deoxynivalenol, metabolites of deoxynivalenol and masked deoxynivalenol in food and feed. http://registerofquestions.efsa.europa.eu
- 71. Marliére CA Pimenta RC, Cunha AC (2009) Fumonisin as a risk factor to esophageal cancer: a review. Appl Cancer Res 29: 102–105.
- 72. Pestka JJ (2010) Deoxynivalenol: mechanisms of action, human exposure, and toxicological relevance. Arch Toxicol 84: 663–679. pmid:20798930
- 73. Bonnet MS, Roux J, Mounien L, Dallaporta M, Troadec JD (2012) Advances in deoxynivalenol toxicity mechanisms: the brain as a target. Toxins 4: 1120–1138. pmid:23202308