Embed Size (px)
Citation preview
This article was downloaded by: [University of Tennessee At Martin]On: 08 October 2014, At: 10:53Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK
Journal of Occupational and Environmental HygienePublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/uoeh20
Toxigenic Fusarium spp. as Determinants ofTrichothecene Mycotoxins in Settled Grain DustAnne Straumfors Halstensen a , Karl-Christian Nordby b , Sonja Sletner Klemsdal c , Oleif Elenc , Per-Erik Clasen d & Wijnand Eduard aa Department of Occupational Hygiene , National Institute of Occupational Health , Oslo ,Norwayb Department of Occupational Medicine , National Institute of Occupational Health , Oslo ,Norwayc Department of Plant Pathology , Norwegian Crop Research Institute , Ås , Norwayd Department of Feed and Food Hygiene , National Veterinary Institute , Oslo , NorwayPublished online: 03 May 2012.
To cite this article: Anne Straumfors Halstensen , Karl-Christian Nordby , Sonja Sletner Klemsdal , Oleif Elen , Per-Erik Clasen& Wijnand Eduard (2006) Toxigenic Fusarium spp. as Determinants of Trichothecene Mycotoxins in Settled Grain Dust, Journalof Occupational and Environmental Hygiene, 3:12, 651-659, DOI: 10.1080/15459620600987431
To link to this article: http://dx.doi.org/10.1080/15459620600987431
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions
Journal of Occupational and Environmental Hygiene, 3: 651–659ISSN: 1545-9624 print / 1545-9632 onlineCopyright c© 2006 JOEH, LLCDOI: 10.1080/15459620600987431
Toxigenic Fusarium spp. as Determinants of TrichotheceneMycotoxins in Settled Grain Dust
Anne Straumfors Halstensen,1 Karl-Christian Nordby,2 Sonja SletnerKlemsdal,3 Oleif Elen,3 Per-Erik Clasen,4 and Wijnand Eduard1
1Department of Occupational Hygiene, National Institute of Occupational Health, Oslo, Norway2Department of Occupational Medicine, National Institute of Occupational Health, Oslo, Norway3Department of Plant Pathology, Norwegian Crop Research Institute, As, Norway4Department of Feed and Food Hygiene, National Veterinary Institute, Oslo, Norway
Trichothecenes are immunosuppressive mycotoxins pro-duced mainly by Fusarium spp. and often are detected asnatural contaminants of grain and other agricultural products.Exposure to trichothecenes through inhalation during grainwork may represent possible health risks for grain farmers.We aimed, therefore, to investigate the level of Fusarium spp.and trichothecenes in settled grain dust collected during workon 92 Norwegian farms. Mycotoxins were determined by gaschromatography-mass spectrometry, whereas the Fusariumspp. were identified and quantified both by species-specificsemiquantitative polymerase chain reaction (PCR) and bycultivation. All potential trichothecene-producing molds in thegrain dust were quantified using a PCR assay specific for tri5,the gene coding for trichodiene synthase that catalyzes thefirst step in the trichothecene biosynthesis. We performed cor-relation analysis between mold-DNA and mycotoxins to assesswhether the PCR-detected DNA could be used as indicators ofthe mycotoxins. The methodological problem of detecting smallamounts of airborne mycotoxins during grain work may thenbe avoided. Whereas the trichothecene-producing Fusariumspecies in grain dust could not be identified or quantified toa sufficient extent by cultivation, all investigated Fusariumspp. could be specifically detected by PCR and quantifiedfrom the DNA agarose gel band intensities. Furthermore, weobserved a strong correlation between the trichothecenes HT-2 toxin (HT-2) or T-2 toxin (T-2) and DNA specific for tri5(r = 0.68 for HT-2 and r = 0.50 for T-2; p < 0.001), F.langsethiae (r = 0.77 for HT-2 and r = 0.59 for T-2; p <0.001), or F. poae (r = 0.41 for HT-2 and r = 0.35 for T-2; p <0.001). However, only a moderate correlation was observedbetween the trichothecene deoxynivalenol (DON) and thecombination of its producers, F. culmorum and F. graminearum(r = 0.24, p = 0.02), and no significant correlation wasobserved between DON and tri5. PCR clearly improved thedetection of toxigenic Fusaria as potential sources of healthrisks for farmers inhaling grain dust during work, but the use ofFusarium-DNA as indicators for trichothecenes should be usedcautiously.
Keywords Fusarium, grain dust, occupational exposure, PCR,trichothecenes, tri5
Address correspondence to: Anne S. Halstensen, Department ofOccupational Hygiene, National Institute of Occupational Health,P.O. Box 8149 Dep. Oslo N-0033, Norway; e-mail: [email protected].
INTRODUCTION
F usarium spp. are phytopathogenic molds that cause plantdiseases and lead to crop losses in grain cultivation.
Several species also are able to produce a number of my-cotoxins that are highly toxic to humans and animals.(1,2)
The trichothecene mycotoxins are especially important, asthey are potent inhibitors of eukaryotic protein synthesis.(3)
Deoxynivalenol (DON), T-2 toxin (T-2), and HT-2 toxin (HT-2) are the most frequently found trichothecenes in Norwegianoats and barley.(4) Mycotoxin synthesis depends not only onthe mold genus but also on the fungal species and strain, andthe environmental conditions.(5) The most prominent DONproducers are F. graminearum and F. culmorum, which causeFusarium head blight in grain.(2) T-2 and HT-2 are producedby F. sporotrichioides,(6) F. equiseti,(2) F. acuminatum,(7) F.langsethiae,(6,8) and some strains of F. poae.(6) Other tri-chothecenes found in Norwegian grain are nivalenol (NIV),which is produced by F. graminearum, F. culmorum, F.cerealis, F. equiseti, and F. poae;(9−12) diacetoxyscirpenol(DAS), which is produced by F. sporotrichioides, F. poae, andF. equiseti;(11,12) and monoacetoxyscirpenol (MAS), which isproduced by F. poae.(11)
The most prominent route of exposure to mycotoxins is viaingestion of contaminated food, and the effects of ingestion arerelatively well-known through animal studies and epidemio-logic evidence.(2,13) However, exposure through skin contactand the inhalation of dust originating from contaminatedgrains or food also may pose significant risks.(14−16) Studiesin rodents have shown that inhaled T-2 toxin may be moretoxic than dermally, orally, or intraperitoneally administrated
Journal of Occupational and Environmental Hygiene December 2006 651
Dow
nloa
ded
by [
Uni
vers
ity o
f T
enne
ssee
At M
artin
] at
10:
53 0
8 O
ctob
er 2
014
mycotoxins.(17−19) Working with grain may generate signifi-cant amounts of airborne dust particles consisting of fragmentsof fungal hyphae and grain, bacteria and fungal spores, plantpollen, and silica and other inorganic particles.(20,21) Fungalspores may contain significant amounts of mycotoxins.(13,22)
Spores of many species have aerodynamic diameters of lessthan 5 µm and are thus able to reach the alveoli.
Consequently, grain dust represents a potential source ofmycotoxin exposure to grain handlers. Acute and chronicrespiratory diseases such as pulmonary mycotoxicosis, silofiller’s disease, toxic mold syndrome, or organic dust toxicsyndrome have been reported after inhalation of mycotoxinsor toxigenic molds, including Fusarium.(23) Inhalation ofmycotoxin-containing grain dust has been suggested as thecause of acute kidney failure,(24) late abortions, and adversereproductive outcomes in epidemiologic studies.(25,26)
To assess the health risk, it is important to investigate thelevel of molds and mycotoxins in the grain dust. Exposuremeasurements using personal sampling, i.e., sampling thebioaerosol in the breathing zone of a farmer while working,would give the most precise exposure assessment. However,the amount of dust obtained by personal sampling is too smallfor mycotoxin quantification by the methods presently avail-able, and because many mold species are no longer cultivableafter aerosolization, the conventional cultivation methods maynot be suitable for airborne dust. High diversity, intraspeciesvariability and conflicting taxonomy in the Fusarium genusadd to the complexity.
Species identification of mycotoxin-producing Fusariumhas been of great importance not only for studies of Fusariumepidemiology, for chemical control, and for risk assessmentof cereal grain for human or animal consumption but also inrelation to Fusarium head blight in small-grain cereals and earrot in maize, which lead to losses due to lower crop yieldsand grain quality. This has resulted in the development ofmolecular techniques such as polymerase chain reaction (PCR)for the identification and differentiation of Fusarium, basedmostly on random DNA fragments(27−29) or rDNA.(27) PCRassays have been developed for the species-specific detectionof F. avenaceum, F. culmorum, F. graminearum, F. poae,F. langsethiae and F. sporotrichioides.(27,29−31) Also, structuralgenes such as the trichodiene synthase gene (tri5), which codesfor the enzyme that catalyzes the first step in the trichothecenebiosynthesis, have been used to detect all trichothecene-producing Fusarium spp. in plants and grain.(32,33) The PCRtechniques have led to revisions of the taxonomy of the moldsand have facilitated investigations of Fusarium disease inplants. Allowing specific and sensitive detection of target DNAmolecules in a complex mixture regardless of the cultivabilityof the molds, PCR offers an alternative to the microbiologicalmethods and may have the potential for identifying toxigenicmolds in dust samples.
The purpose of the present study was to examine thepresence of Fusarium spp. and trichothecenes in settled graindust from which it is possible to collect a reasonable amountfor analysis and which has the same origin as airborne dust.
We determined the level of Fusarium mycotoxins and sixdifferent Fusarium spp. in settled grain dust. Fusarium spp.were identified by both cultivation and PCR to determine themost suitable method, and the identification of Fusaria asdeterminants of trichothecenes was evaluated.
METHODS
Selection of Grain FarmsEleven of the most important grain producing municipalities
in Norway, each with more than 340 grain farms, wereidentified according to the Census of Agriculture and Forestryof 1989.(34) They were grouped in three geographically andclimatically different districts according to their vicinity tothe River Glomma, located in the eastern part of Norway;Lake Mjøsa, located in the eastern inland; or the TrondheimFjord, located in Mid-Norway. All three districts are locatednorth of 60 degrees northern latitude. A list of active cerealfarmers in these districts was supplied by the Norwegian GrainCorporation, and farmers were contacted by telephone andinvited to participate in the study. All the farmers contactedwho had available grain for threshing or storage work agreedto participate.
Sampling of Grain DustWe collected 109 samples of newly settled grain dust
released during the threshing or storage work of the 1999and 2000 crops of spring wheat, oats, and barley. Grain dustsamples of 0.1–3 g were collected from the surfaces of thegrain container, drier, elevator, combine harvester, or graintrailer using paper filter cassettes (ALK Abello, Horsholm,Denmark) in dust collector nozzles (ALK Abello) fitted toBlack & Decker HC431 dust busters (Slough, Berkshire, U.K.).The collection of large particles (>400 µm), e.g., whole grain,was avoided by use of 400-µm stainless steel mesh that coveredthe filter cassettes. We collected settled dust originating fromthe grain that was handled at the time of sampling by avoidingsurfaces that were visibly contaminated with dust before thework started, whenever possible. Samples were transferred toclean polypropylene tubes (Nunc AS, Roskilde, Denmark) andstored at −20◦C until analysis.
Quantification of Trichothecenes Type Aand Type B by GC-MS
Mycotoxins were extracted from aliquots of a maximum0.4 g grain dust by mixing with 8 mL of acetonitrile:water(84:16, vol/vol) for 1 hour in ambient temperature. Extractswere purified, and derivatized mycotoxins were analyzed bygas chromatography-mass spectrometry (GC-MS) as previ-ously described.(10) 15-Acetyl-DON has never been foundin Norwegian cereals, as most North European isolates ofF. graminearum and F. culmorum belong to the chemotypeIA that produces DON and 3-acetyl DON.(10) 15-Acetyl-DONwas thus not expected to be found in the grain dust, so it wasused to control the purification step and to compensate for thevariation in the instrument response during the run by adding it
652 Journal of Occupational and Environmental Hygiene December 2006
Dow
nloa
ded
by [
Uni
vers
ity o
f T
enne
ssee
At M
artin
] at
10:
53 0
8 O
ctob
er 2
014
to the sample extracts before purification on the Mycosep 225column (Romer Labs Inc., Union, Mo.). An external standardmade of standard solutions of each toxin was added to toxin-free wheat extract and used to make a standard calibrationcurve for the quantification of the trichothecenes in the samplesextracts.
The detection limits (DL) were 50–100 µg/kg for T-2;50 µg/kg for nivalenol (NIV); 40 µg/kg for fusarenon-X;30 µg/kg for HT-2; 20 µg/kg for DON, 3-acetyl-DON, and4,15-diacetoxyscirpenol (DAS); and 10 µg/kg for monoace-toxyscirpenol (MAS). The recovery of each toxin was deter-mined by spiking toxin-free wheat with a known concentrationof each toxin before extraction. The mean recovery was 102%for DAS; 101% for T-2; 96% for HT-2; 84% for NIV, fusarenon-X, and MAS; 83% for 3-acetyl-DON; and 81% for DON. Amean value of 780 µg/kg of DON with a relative standarddeviation of 16% was obtained for the reference material (BCRWheat RM 379, certified value 670 µg/kg of DON, ReportEUR 14407 EN, Commission of the European Communities,Directorate-General XIII, Luxembourg 1993 (Cat. No CD-NA-14407 EN-C).
Cultivation of MoldsFrom each sample, 10 mg grain dust was mixed with
10 mL sterile water in a small bottle and shaken by mixer for15 sec (Fisons whirlimixer, WM/220/F, Rich-Mond, Wigan,Derbyshire, U.K.). The suspension was diluted 1:10 and500 µL of each suspension was plated on each of five petridishes with a modified Czapeks-Dox agar.(35) The mediumused for the 30 samples from 1999 was as described byAbildgren et al.(35) (CZID). In the second year, our laboratorychanged the medium slightly when iprodion was replaced by0.375 mg propiconazole/1000 mL and 1.125 mg fenpropi-morph/1000 mL (3 mg Tilt Top/1000 mL) (CZPD). Iprodionis unstable in agar compared with the other two fungicides.CZPD was found to be more stable over time than CZID and,therefore, easier to handle. Otherwise, no differences betweenthe two media were detected (data not shown).
The number of Fusarium colonies were counted andpresented as the number of colony forming units/mg dust(CFU/mg). The colonies were grouped according to theappearance of the mycelium on the agar. Similar coloniesconstituted a specific group, and mycelium from a few coloniesfrom each group were transferred to Spezieller Nahrstoffarmeragar (SNA)(36) and potato dextrose agar (PDA). The plates werethen incubated at 25◦ for 7 days under combined black light andcold daylight, alternating with darkness in 12-hour periods; thefungus was then identified at the species level.
DNA ExtractionDNA was extracted from 10-mg dust samples using the
DNeasy Plant Mini Kit (QIAGEN Inc., Valencia, Calif.). Dustsamples were transferred to 1.5-mL microcentrifuge tubes,and an equal volume of sterile sea sand (Merck, Darmstadt,Germany) was added. The dust was ground manually withthe sea sand and 400 µL AP1 lysis buffer from the kit for
1–2 min with a Teflon pestle. Additional steps in the DNAextraction were performed according to the manufacturers’recommendations. The resulting DNA was eluted in a totalvolume of 50 µL of AE buffer from the kit.
PCR AnalysesDNA (2 µL) was used as the template in the PCR
reaction. The following primers were used: Tr5F/R(33) to detectall trichothecene-producing Fusarium spp., JiAf/r to detectF. avenaceum,(29) UBC85F/R to detect F. graminearum,(27)
PfusF/FspoR to detect F. sporotrichioides, PfusF/FlanR todetect F. langsethia, and FpoF/ITS4 to detect F. poae.(31) Forthe specific detection of F. culmorum, a nested PCR in onetube was performed using the primer pairs FculAF/R for thefirst cycles and FculBF/R for the last cycles of PCR.(30) Am-plification of the plant and fungal ITS2 region with primer pairITS3 (5′-GCATCGATGAAGAACGCAGC-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′)(38) was used as a generalpositive control to discriminate between uninfected samplesand possible PCR inhibition. The Fusarium-specific primersequences are given in Table I.
PCR was performed in a final volume of 25 µL. Except forthe detection of F. culmorum, the reaction mixtures contained25 pmol of each primer, 0.6 U Taq polymerase (Roche, Basel,Switzerland), 0.2 mM dNTP, 0.1 mg/mL BSA, and 3 mMMgCl2. Amplification was performed in a GeneAmp PCRSystem 9700 thermal cycler (Applied Biosystems, Foster City,Calif.) programmed for initial denaturation at 94◦C for 2 minfollowed by 30 cycles of 30 sec at 94◦C, 30 sec at the annealingtemperature, and 30 sec at 72◦C. For ITS3/ITS4, JiAF/R, andUBC85F/R, the annealing temperatures were 55◦C, 58◦C, and68◦C, respectively. For the other primer pairs the annealingtemperature was 66◦C. For the detection of F. culmorum theprimer pair FculAF/R was diluted 10,000 times (2.5 fmol),whereas the other components in the PCR reaction mixturesremained unchanged.
The thermal cycler was programmed for initial denaturationat 94◦C for 2 min followed by 25 cycles of 30 sec at 94◦C,30 sec at 70◦C, and 30 sec at 72◦C. The thermal cycler was fur-ther programmed to continue with 40 cycles of 30 sec at 94◦C,30 sec at 58◦C, and 30 sec at 72◦C. Ten µL of the amplificationreaction mixture was run on a 1.2% agarose gel, and theresulting DNA bands were evaluated by densitometry (Gel DokEQ; Bio-Rad Laboratories Inc., Hercules, Calif.). The DNAquantities were determined by calculating the pixel volumeof the bands (area times intensity), using Quantity One 1-DAnalysis Software, version 4.4 (Bio-Rad). The negative controlwith no Fusarium DNA was set to a quantity of zero, andthe quantity of the PCR fragment resulting from 50 ng of theFusarium spp. used as the positive control was regarded as100%.
The software calculated the quantities of the PCR productsresulting from the unknown samples as percentages of thepositive control sample. The potential inhibition of the PCRreactions was tested using the general positive control primerpair ITS3/ITS4. Only samples that amplified 100% of the
Journal of Occupational and Environmental Hygiene December 2006 653
Dow
nloa
ded
by [
Uni
vers
ity o
f T
enne
ssee
At M
artin
] at
10:
53 0
8 O
ctob
er 2
014
TABLE I. PCR-primers Used to Identify Fusarium spp. and tri5
Fusarium species/gene Primer Sequence 5′ → 3′
F. graminearum UBC85F GCAGGGTTTGAATCCGAGACUBC85R AGAATGGAGCTACCAACGGC
F. avenaceum JiAf GCTAATTCTTAACTTACTAGGGGCCJiAr CTGTAATAGGTTATTTACATGGGCG
F. culmorum FculAF TGCCAGACCAAGACGAAGTGAGAGFculAR TGAACTGCCACTCCGTTGCAAGTGFculBF TTGATCAAACCATCATCATCFculBR AGAAAGGGTTAGAATCATGC
F. sporotrichioides PfusF CCGCGCCCCGTAAAACGFspoR ACTGTGTTTGCACACAGATC
F. langsethiae PfusF CCGCGCCCCGTAAAACGFlanR CTGTCGGTAAGGACAGATC
F. poae FpoF CGGATCAGCCCGTCCTTCITS4 TCCTCCGCTTATTGATATGC
tri5 tr5F AGCGACTACAGGCTTCCCTCtr5R AAACCATCCAGTTCTCCATCTG
positive control, indicating no PCR inhibition, were included inthe analyses. All PCR reactions were run twice, and the averagevalues of the DNA quantities were used in statistical analyses.
Statistical AnalysesThe percentage of samples above the DL and the median
and maximum values were determined for each mycotoxin.The percentage of positive samples and median and maximumvalues were determined for the cultivated Fusarium spp. andfor the Fusarium DNA. Associations between detected moldsand mycotoxins were evaluated with nonparametric Spearmanrank correlation coefficients if more than 10% of the samplescontained measurable amounts of the mold parameter that wasto be included. All samples were included in the correlationanalysis by replacing values that were less than the DL withthe lowest value determined divided by the square root of 2if not quantifiable.(39) Statistical analysis was performed withthe software package SPSS version 11.5 for Windows (SPSS
TABLE II. Trichothecene Concentration in Settled Grain Dust (n = 109)
MycotoxinDetection
Limits (µg/kg)
Samples AboveDetection Limits,
% (n)Samples Above
Zero, % (n)
Median ofAll Samples
(µg/kg)
Median of SamplesAbove DetectionLimits (µg/kg)
HT-2 30 58 (64) 71 (77) 52 (0–2400) 114 (32–2400)DON 20 44 (48) 60 (65) 17 (0–2200) 46 (21–2200)T-2 50 23 (26) 27 (29) 0 (0–1200) 94 (54–1200)DAS 20 5 (6) 12 (13) 0 (0–37) 26 (23–37)NIV 50 2 (2) 18 (19) 0 (0–67) 59 (51–67)MAS 10 0 11 (20) 0 (0–5) 03-Acetyl-DON 20 0 0 0 0Fusarenon-X 40 0 0 0 0
Inc., Chicago, Ill.). Graphic presentations were prepared withthe software package SigmaPlot 2001 (Systat Software Inc.,Point Richmond, Calif.).
RESULTS
Trichothecenes in Settled Grain DustThe dominant trichothecenes in settled grain dust were
HT-2, DON, and T-2, whereas NIV, DAS, and MAS wereseldom found at levels above the DL (Table II). No 3-acetyl-DON or Fusarenon-X was found. The total amount of detectedtrichothecenes was median 69 µg/kg (range 0–6000 µg/kg).Samples containing mycotoxins above the DL had median114 µg/kg (range 33–2400, n = 63) of HT-2; median 94 µg/kg(range 54–1200, n = 26) of T-2; average 59 µg/kg (range 51–67, n = 2) of NIV; median 46 µg/kg (range 21–2200, n = 42)of DON; and median 26 µg/kg (range 23–37, n = 5) of DAS(Table II).
654 Journal of Occupational and Environmental Hygiene December 2006
Dow
nloa
ded
by [
Uni
vers
ity o
f T
enne
ssee
At M
artin
] at
10:
53 0
8 O
ctob
er 2
014
TABLE III. Fusarium Species and tri5 in Settled Grain Dust Measured by PCR and Cultivation
Cultivation (n = 109) PCR (n = 103)CFU/mg Dust Relative DNA Content (%)
PositiveSamples (%) Median Maximum
PositiveSamples (%) Median Maximum
F. avenaceum 39 0 124 94 28 99F. poae 5.5 0 24 82 25 96F. langsethiae 0 0 0 80 52 100F. sporotrichoides 0 0 0 5 0 20F. culmorum 3.7 0 12 40 0 92F. graminearum 0.9 0 4 9 0 9tri5 — — — 86 25 100
Cultivated Fusarium spp. from Grain DustF. avenaceum was the dominant Fusarium species in grain
dust and was cultivated from 39% of the samples (Table III).The prevalence of F. culmorum, F. poae, and F. graminearumwas less than 6%, and no colonies of F. sporotrichioides andF. langsethiae were identified.
Detection of Species-Specific Fusarium DNAin Grain Dust
With the species-specific Fusarium PCR, F. avenaceum wasidentified in 94% of the grain dust samples, whereas F. poae,F. langsethiae, and F. culmorum were detected in 84%, 80%,and 40% of the grain dust samples, respectively (Table III). F.sporotrichioides and F. graminearum were detected in less than10% of the samples. The median F. langsethiae DNA contentwas 53% of the positive control, whereas the DNA content ofthe other species was lower (Table III). The variability betweenthe samples of F. graminearum DNA ranged from 0% to 9%and for F. langsethiae DNA ranged from 0% to 100% , relativeto the positive control. The tri5 was detected in 86% of thesamples, with a median of 25% relative DNA content (range
TABLE IV. Associations Between PCR-AmplifiedFusarium DNA and Trichothecenes Determined inSettled Grain Dust (n = 103)
HT-2 T-2 DON
F. avenaceum −0.01 −0.13 0.02F. poae 0.41∗∗∗ 0.35∗∗∗ 0.19F. langsethiae 0.77∗∗∗ 0.59∗∗∗ 0.15F. culmorum 0.25∗ 0.11 0.15F. graminearum na na natri5 0.68∗∗∗ 0.50∗∗∗ 0.16
Note: na, not analyzed (not subjected to correlation analysis due to few samplescontaining F. graminearum).∗p < 0.05; ∗∗∗p≤ 0.001.
of 0–100). Higher percentages of samples containing Fusariumwere detected by PCR than by cultivation.
Associations Between Trichothecenesand Fusarium DNA
F. poae, F . langsethiae, and tri5 correlated strongly withHT-2 and T-2 (Table IV and Figure 1). Tri5 also correlatedstrongly with the combination of HT-2, T-2, and DON (rs =0.63, p < 0.01; results not shown in table). The combination ofF. graminearum and F. culmorum correlated moderately, butsignificantly, with DON (rs = 0.24, p = 0.02). However, eachone of these species did not show significant correlations withDON (Table IV). On the contrary, F. culmorum was moderatelyand significantly correlated with HT-2. F. avenaceum does notproduce either of the dominant trichothecenes found in thisstudy and, as expected, no correlations with the trichotheceneswere found (Table IV). Correlation analyses that includedcultivated dust were omitted because of the low percentageof positive samples.
DISCUSSION
H T-2, DON, and T-2 were the dominant trichothecenemycotoxins in settled grain dust (Table II). These toxins
often are found in Norwegian grain,(4,10,40) although in lowerconcentrations than those we found in the grain dust in thisstudy. Grain dust is probably enriched with particles from theouter shell layer of the grain, as other studies have reportedhigher trichothecene concentrations in the shell than in wholegrain samples.(41) These findings are further supported bySorensen et al.,(42) who reported a higher concentration ofaflatoxin in the <7 µm particle fraction of corn dust thanin larger particles. The DON concentration in our study(median 17 µg/kg, range 0–2,200 µg/kg), with 44% of samplescontaining DON above the DL of 20 µg/kg (Table II), wassomewhat higher than reported by Krysinska-Traczyk et al.,(43)
who found 15–460 µg/kg DON in 4 of 10 grain dust samples.However, in the latter study, the NIV concentration was higher(30–880 µg/kg in 4 of 10 grain dust samples) than in our
Journal of Occupational and Environmental Hygiene December 2006 655
Dow
nloa
ded
by [
Uni
vers
ity o
f T
enne
ssee
At M
artin
] at
10:
53 0
8 O
ctob
er 2
014
FIGURE 1. Scatter plots of HT-2 and T-2 vs. PCR-amplified DNA of F. poae, F. langsethiae, or tri5 in grain dust
656 Journal of Occupational and Environmental Hygiene December 2006
Dow
nloa
ded
by [
Uni
vers
ity o
f T
enne
ssee
At M
artin
] at
10:
53 0
8 O
ctob
er 2
014
samples (median 0 µg/kg, range 0–67 µg/kg, with only 2% ofsamples above the DL of 50 µg/kg).(43) This possibly reflectsdifferent profiles of mycotoxin-producing Fusarium spp. dueto different growth conditions in the two countries.
Mycotoxins other than trichothecenes also have been de-tected in dust generated from grain handling. Median 4 µg/kgochratoxin A (range 2–130 µg/kg) was found in the same graindust samples as found in the present study.(44) Zearalenone wasmeasured at concentrations ranging from 25 to 100 µg/kg insettled dust from grain elevators in the New Orleans, Louisiana,area,(45) whereas up to 6.2 µg/kg of aflatoxin B1 in settleddust was measured during corn harvesting.(46) We found 69 µgtrichothecene mycotoxins per kg settled grain dust (range 0–5900) when summarizing all the detected trichothecenes in thesettled grain dust and 73 µg/kg (range 2–6000) when includingochtatoxin A. Grain work that generates mycotoxin-containingdust clearly represents health risks for the farmer. Adding tothis risk are other mycotoxins that were not detected in thisstudy—bacteria, endotoxins, glucans, and allergens—that alsoare grain dust components that may affect farmers’ health.
Molds in dust samples generally are difficult to cultivate dueto few viable propagules. It is probable that because of this, cul-tivation was not the optimal method to use for the detection ofFusarium species in grain dust in the present study (Table III).Additionally, although F. langsethiae grows well on the non-selective PDA, it is an inferior competitor when grown on CZIDand CZPD media compared with the other Fusarium spp. inthis study, and, therefore, it may have been underestimated bycultivation.
As live and intact organisms are not needed for thedetection of specific fragments of DNA, PCR can detect bothlive molds and molds that are dead at the time of sampling, aslong as the DNA has not disintegrated. In the present study, weused PCR to detect the gene coding for trichodiene synthase(tri5) and species-specific Fusarium DNA in grain dust.Indeed, the various Fusarium spp. were detected successfullyby PCR in the present study, and even F. langsethiae andF. sporotrichioides, which were not found by cultivation, weredetected. F. avenaceum, F. poae, and F. langsethiae were thedominant Fusarium species in settled grain dust (Table III).Tri5 also was detected by PCR and revealed that molds capableof producing trichothecene mycotoxins were present in 86% ofthe samples. However, the sensitivity of the PCR method mayvary depending on each specific primer pair, which sometimesgives results that quantitatively are not directly comparablebetween species/assays. PCR results, therefore, can becompared only within each specific PCR assay. The PCRdetection of F. graminearum was 100 times less sensitive thanthe species-specific detection of the other Fusarium species(results not shown). Thus, the PCR technique was expectedto underestimate the amount of F. graminearum DNA relativeto the other Fusarium spp., and this may partially explain thelow levels of F. graminearum detected by PCR (Table III).
Of all Fusaria detected, F. langsethiae had the strongestcorrelation with HT-2 and T-2, consistent with the hypothesisthat F. langsethiae is the main producer of T-2 and HT-2 in
Norwegian grain.(8) The relatively strong correlation observedbetween F. poae and HT-2 and T-2 is not in agreementwith Thrane et al.,(6) who recently reported the mycotoxinprofile of 49 F. poae strains of various origins. Only five ofthese cultivated strains produced HT-2 or T-2.(8) Althoughmycotoxin production in the environment may differ fromproduction under laboratory conditions, this can suggest thatF. poae may be a poor indicator for HT-2 and T-2, and theobserved correlation between F. poae and HT-2 in our studycould be due to co-variation with F. langsethiae, as suggestedby a strong correlation between F. poae and F. langsethiae(rs = 0.63, p < 0.01). We found only a moderate correlationbetween DON and the combination of F. graminearum andF. culmorum, whereas others have reported strong correlationsbetween DON and DNA of Fusarium spp. [r = 0.96, n =300;(47) r2= 0.8, n = 40(48)], but these studies examinedwhole wheat and not grain dust. Fusarium spp. producing T-2and HT-2 are the dominant trichothecene-producing Fusariumspecies in the Nordic countries, whereas the DON producingF. graminearum and F. culmorum are found more frequentlyin the other European countries and in the United States.(49)
Furthermore, different strains of both F. culmorum andF. graminearum may have different chemotypes that do notall necessarily produce DON.(50,51) This could not explain theweak correlation though, since DON was one of the dominanttrichothecenes found in the grain dust and only a few or nosamples were positive for NIV and 3-acetyl-DON (Table II).However, mycotoxins may be present in the dust long afterthe producers have died and disintegrated, which may be theexplanation for this mismatch between the biology and thechemistry. Relatively high tri5 levels and strong correlationswith HT-2 and T-2 were found. However, there was nocorrelation with DON, although this was expected as alltrichothecene-producing molds have this gene. This suggeststhat the majority of the trichothecene-producing molds itwas possible to detect in the dust were not DON-producingspecies, and it supports the hypothesis that the DON-producersdisintegrated before the time of analysis.
Any comparison of cultivated mold and trichothecenes wasprecluded by the low incidence of morphologically identifiedFusarium spp. in the dust. As personal measurements ofexposure to mycotoxins are not feasible at present, personalsampling of airborne dust for gravimetric analysis and thesimultaneous collection of settled dust for mycotoxin anal-ysis may be used to estimate the inhalant exposure underthe assumption that the mycotoxin content in settled andairborne dust is similar. Stationary sampling with high volumepumps is another alternative enabling mycotoxin detectionin airborne dust.(52) However, due to the spatial variation indust concentrations and the sedimentation of larger particles,stationary sampling is not a good surrogate for personalexposure measurements. The detection of airborne mycotoxin-producing molds by microbiological methods has been usedas mycotoxin exposure indicators,(24,53) and indirect qualitativeexposure indicators, such as late blight fungal forecasts, havebeen proposed for epidemiologic studies of grain farmers.(54)
Journal of Occupational and Environmental Hygiene December 2006 657
Dow
nloa
ded
by [
Uni
vers
ity o
f T
enne
ssee
At M
artin
] at
10:
53 0
8 O
ctob
er 2
014
The search for mycotoxin determinants in this study wasapproached by the parallel examination of Fusarium toxins andFusarium DNA in settled dust. The strong positive correlationbetween Fusarium DNA and trichothecenes in settled dust thatwas demonstrated in our present study suggests that a similarassociation can be expected in airborne dust. Although themycotoxin production is dependent on varying environmentalconditions, and the detection of the tri5 gene or the DNAof a potentially toxigenic mold species in general does notnecessarily mean that any mycotoxin is present, the detection ofgroup-specific or species-specific Fusarium DNA in airbornedust samples may possibly provide an approach for the assess-ment of mycotoxin exposure and should be subjected to furtherinvestigation. Various strategies and primers for detecting theDNA of airborne fungi are published,(55−58) and two have beenapplied on samples from the work environment.(56,58)
CONCLUSION
T his study shows that PCR is superior compared withcultivation in detecting Fusarium species in settled grain
dust under these specific study conditions. The trichodienesynthase gene (tri5) enabled the detection of all potentialtrichothecene-producing molds, whereas the various Fusariumspecies were detected by species-specific assays, regardless ofviability or cultivability. Both tri5 and F. langsethiae-specificDNA strongly correlated with HT-2 and T-2 in grain dust,and possibly may indicate the presence of HT-2 and T-2.However, not all expected associations between mold DNA andtrichothecenes were present; thus, the use of molecular markersfor trichothecenes should be used cautiously. Nevertheless, thisstudy shows that PCR improves the estimates of Fusariumin settled grain dust, both in relation to identity, possibletoxicity, and the total quantity of cultivable and non-cultivabletoxigenic Fusaria as possible sources of farmers’ health risk.We will test this application on aerosol samples in additionalstudies to obtain more sensitive and more specific exposuremeasurements of molds and also to obtain better exposureestimates of trichothecenes.
ACKNOWLEDGMENTS
W e thank Lene Madsø for valuable participation in col-lecting samples, and Grete Lund and Jafar Razzaghian
for excellent technical assistance. We acknowledge the lateWenche Langseth for contributing her great knowledge ofmolds and mycotoxins in the early stages of this work. Thisstudy was supported by the Norwegian Research Council.
REFERENCES
1. Council for Agricultural Science and Technology: Mycotoxins—Risksin Plant, Animal, and Human Systems. Task Force Report no. 139,ISSN0194-4088, Ames, Iowa (2004).
2. Creppy, E.E.: Update of survey, regulation, and toxic effects of myco-toxins in Europe (Review). Toxicol. Lett. 127:19–28 (2002).
3. Desjardins, A.E., T.M. Hohn, and S.P. McCormick: Trichothecenebiosynthesis in Fusarium species—Chemistry, genetics, and significance.Microbiol. Rev. 57(3):595–604 (1993).
4. Langseth, W., and T. Rundberget: The occurrence of HT-2 toxin andother trichothecenes in Norwegian cereals. Mycopathologia 147:157–165(1999).
5. Sweeney, M.J., and A.D.W. Dobson: Mycotoxin production by As-pergillus, Fusarium, and Penicillium species (Review). Int. J. FoodMicrobiol. 43:141–158 (1998).
6. Thrane, U., A. Adler, P.-E. Clasen, F. Galvano, W. Langseth, H. Lewet al.: Diversity in metabolite production by Fusarium langsethiae, F.poae, and F. sporotrichioides. Int. J. Food Microbiol. 95:257–266 (2004).
7. Logrieco, A., C. Altomare, A. Moretti, and A. Bottalico: Cultural andtoxigenic variability in Fusarium acuminatum. Mycol. Res. 96:518–523(1992).
8. Torp, M., and W. Langseth: Production of T-2 toxin by a Fusariumresembling Fusarium poae. Mycopathologia 147:89–96 (1999).
9. Desjardins, A.E., A.M. Jarosz, R.D. Plattner, N.J. Alexander, D.W.Brown, and J.E. Jurgenson: Patterns of trichothecene production,genetic variability, and virulence to wheat of Fusarium graminearumfrom smallholder farms in Nepal. J. Agric. Food Chem. 52(20):6341–6346 (2004).
10. Langseth, W., A. Bernhoft, T. Rundberget, B. Kosiak, and M. Gareis:Mycotoxin production and cytotoxicity of Fusarium strains isolated fromNorwegian cereals. Mycopathologia 144:103–113 (1999).
11. Liu, W.Z., L. Sundheim, and W. Langseth: Trichothecene productionand the relationship to vegetative compatibility groups in Fusarium poae.Mycopathologia 140(2):105–114 (1998).
12. Kosiak, B.E., A. Holst-Jensen, T. Rundberget, M.T.G. Jaen, andM. Torp: Morphological, chemical, and molecular differentiation ofFusarium equiseti isolated from Norwegain cereals. Int. J. Food Microbiol.99:195–206 (2005).
13. Sorensen, W.G.: Fungal spores: Hazardous to health? Environ. HealthPerspect. 107(Suppl. 3):469–472 (1999).
14. Austwick, P.K.C.: Human mycotoxicosis—past, present, and future.Chem. Ind. 6:47–551 (1984).
15. Flannigan, B.: Mycotoxins in the air. Int. Biodeter. 23:73–78 (1987).16. Hendry, K.M., and E.C. Cole: A review of mycotoxins in indoor air. J.
Toxicol. Environ. Health 38:183–198 (1993).17. Creasia, D.A., J.D. Thurman, L.J.D. Jones III, et al.: Acute inhalation
toxicity of T-2 mycotoxin in mice. Fundam. Appl. Toxicol. 8:230–235(1987).
18. Creasia, D.A., J.D. Thurman, R.W. Wannermacher Jr., and D.L.Bunner: Acute inhalation toxicity of T-2 mycotoxin in the rat and guineapig. Fundam. Appl. Toxicol. 14:54–59 (1990).
19. Schiefer, H.B., and D.B. Hancock: Systemic effects of topical ap-plication of T-2 toxin in mice. Toxicol. Appl. Pharmacol. 76:464–472(1984).
20. Skogstad, A., L. Madsø, and W. Eduard: Classification of particlesfrom the farm environment by automated sizing, counting, and chemicalcharacterization with scanning electron microscopy-energy dispersivespectroscopy. J. Environ. Monit. 1:379–382 (1999).
21. Yoshida, K. and J. Maybank: Physical and environmental characteristicsof grain dust. In Occupational Pulmonary Disease: Focus on Grain Dustand Health, J.A. Dosman and D.J. Cotton (eds.). New York: AcademicPress Inc., 1980. pp. 441–461.
22. Lacey, J., P. Auger, W. Eduard, S. Norn, M.S. Rohrbach, andP.S. Thorne: Tannins and mycotoxins. Am. J. Ind. Med. 25:141–144(1994).
23. Bunger, J., G. Westphal, A. Monnich, B. Hinnendahl, E. Hallier, andM. Muller: Cytotoxicity of occupational and environmentally relevantmycotoxins. Toxicology 202:199–211 (2004).
24. Di Paolo, N., A. Guarnieri, G. Garosi, G. Sacchi, A.M. Mangiarotti,et al.: Inhaled mycotoxins lead to acute renal failure. Nephrol. Dial.Transplant. 9(4):116–120 (1994).
658 Journal of Occupational and Environmental Hygiene December 2006
Dow
nloa
ded
by [
Uni
vers
ity o
f T
enne
ssee
At M
artin
] at
10:
53 0
8 O
ctob
er 2
014
25. Kristensen, P., L.M. Irgens, A. Andersen, A.S. Bye, and L. Sundheim:Gestation age, birth weight, and perinatal death among births to Norwe-gian farmers, 1967–1991. Am. J. Epidemiol. 146(4):329–338 (1997).
26. Kristensen, P., A. Andersen, and L.M. Irgens: Hormone-dependentcancer and adverse reproductive outcomes in Norwegian farmers’families—Effect of climatic conditions favoring fungal growth in grain.Scand. J. Work Environ Health. 26:331–337 (2000).
27. Schilling, A.G., E.M. Moller, and H.H. Geiger: Polymerase chainreaction-based species-specific detection of Fusarium culmorum, F.graminearum, and F. avenaceum. Phytopathology 86:515–522 (1996).
28. Parry, D.W. and P. Nicholson: Development of a PCR assay to detectFusarium poae in wheat. Plant Pathol. 45(2):383–391 (1996).
29. Turner, A.S., A.K. Lees, H.N. Rezanoor, and P. Nicholson: Refinementof PCR-detection of Fusarium avenaceum and evidence from DNAmarker studies for phenetic relatedness to Fusarium tricinctum. PlantPathol. 47:278–288 (1998).
30. Klemsdal, S.S., and O. Elen: Development of a highly sensitivenested-PCR method using a single closed tube for detection of Fusar-ium culmorum in cereal samples. Lett. Appl. Microbiol. 42:544–548(2006).
31. Yli-Mattila, T., S. Paavanaen-Huhtala, P. Parikka, P. Konstantinovaand T.Y. Gagkaeva: Molecular and morphological diversity of Fusariumspecies in Finland and Northwestern Russia. Eur. J. Plant Pathol.110:573–585 (2004).
32. Niessen, M.L., and R.F. Vogel: Group-specific PCR detection of potentialtrichothecene-producing Fusarium species in pure cultures and cerealsamples. Syst. Appl. Microbiol. 21:618–621 (1998).
33. Doohan, F.M., G. Weston, H.N. Rezanoor, D.W. Parry, and P.Nicholson: Development and use of a reverse transcription-PCR assayto study expression of tri5 by Fusarium species in vitro and in planta.Appl. Environ. Microbiol. 65:3850–3854 (1999).
34. Central Bureau of Statistics: Census of Agriculture and Forestry 1989.Vol. IV. Agriculture Area Utilization. NOS (Norwegian Official Statistics)C24, Oslo/Kongsvinger, (1992).
35. Abildgren, M.P., F. Lund, U. Thrane, and S. Elmholt: Czapek-Dox agarcontaining iprodione and dicloran as a selective medium for the isolationof Fusarium species. Lett. Appl. Microbiol. 5:83–86 (1987).
36. Nirenberg, H.I.A.: A simplified method for identifying Fusarium spp.occurring on wheat. Can. J. Bot. 59:1599–1609 (1981).
37. Nelson, P.E., T.A. Toussoun, and W.F.O. Marasas: Fusarium Species:An Illustrated Manual for Identification. State College, Pa: The Pennsyl-vania State University Press, 1983.
38. White, T.J., T. Bruns, S.Lee, and J.Taylor: Amplification and directsequencing of fungal ribosomal RNA genes for phylogenetics. In PCRProtocols—A Guide to Methods and Application, M.A. Innis, D.H.Gelfland, J.J. Sninsky, and T.J. White (eds.). San Diego, Calif.: AmericanPress, 1990. pp. 315–322.
39. Eduard W.: Estimation of mean and standard deviation. Letter to theeditor. AIHA J. 63:4 (2002).
40. Langseth, W., and O. Elen: The occurrence of deoxynivalenol inNorwegian cereals—Differences between years and districts, 1988–1996.Acta Agric. Scand. [B] Soil Plant Sci. 47:176–184 (1997).
41. Trigo-Stockli, D.M., C.W. Deyoe, R.F. Satumbaga, and J.R. Pedersen:Distribution of deoxynivalenol and zearalenone in milled fractions ofwheat. Cereal Chem. 73:388–391 (1996).
42. Sorensen, W.G., J.P. Simpson, M.J. Peach III, T.D. Thedell, and S.A.Olenchock: Aflatoxin in respirable corn dust particles. J. Toxicol. Environ.Health. 7(2–4):669–672 (1981).
43. KrysiÒska-Traczyk, E., I. Liecana, J. Perkowski, and J. Dutkiewicz:Levels of fungi and mycotoxins in samples of grain and grain dustcollected on farms in Eastern Poland. Ann. Agric. Environ. Med. 8:269–274 (2001).
44. Halstensen, A.S., K.C. Nordby, O. Elen, and W. Eduard: Ochratoxin Ain grain dust—Estimated exposure and relations to agricultural practicesin grain production. Ann. Agric. Environ. Med. 11:245–254 (2004).
45. Palmgren, M.S., L.S. Lee, A.J. Delucca II, and A. Ciegler: Preliminarystudy of mycoflora and mycotoxins in grain dust from New Orleans areagrain elevators. Am. Ind. Hyg. Assoc. J. 44(7):485–488 (1983).
46. Selim, M.I., A.M. Juchems, and W. Popendorf: Assessing airborneAflatoxin B1 during on-farm grain handling activities. AIHA J. 59(4):252–256 (1998).
47. Schnerr, H., R.F. Vogel, and L. Niessen: Correlation between DNA oftrichothecene-producing Fusarium species and deoxynivalenol concen-trations in wheat-samples. Lett. Appl. Microbiol. 34:121–125 (2002).
48. Waalwijk, C., R. van der Heide, I. de Vries, et al.: Quantitativedetection of Fusarium species in wheat using TaqMan. Eur. J. PlantPathol. 110:481–494 (2004).
49. Kosiak, B., M. Torp, E. Skjerve, and U. Thrane: The prevalence anddistribution of Fusarium species in Norwegian cereals: A survey. ActaAgric. Scand. [B] Soil and Plant Sci. 53:168–176 (2003).
50. Quarta, A., G. Mita, M Haidukowski, A. Santino, G. Mule, andA. Visconti: Assessment of trichothecene chemotypes of Fusariumculmorum occurring in Europe. Food Addit Contamin. 22(4):309–315(2005).
51. Li, H.P., A.B. Wu, C.S. Zhao, O. Scholten, H. Loffler, and Y.C. Liao:Development of a generic PCR detection of deoxynivalenol and nivalenolchemotypes of Fusarium graminearum. FEMS Microbiol. Lett. 243:505–511 (2005).
52. Lappalainen, S., M. Nikulin, S. Berg, E.L. Hintikka, and A.L.Pasanen: Fusarium toxins and fungi associated with handling of grainon eight Finnish farms. Atmos. Environ. 30:3059–3065 (1996).
53. Fischer, G., T. Muller, R. Schwalbe, R. Ostrowski, and W. Dott:Exposure to airborne fungi, MVOC, and mycotoxins in biowaste handlingfacilities. Int. J. Hyg. Environ. Health. 203:97–104 (2000).
54. Nordby, K.C., A.S. Halstensen, O. Elen, et al.: Trichothecene myco-toxins and their determinants in settled dust related to grain production.Ann. Agric. Environ. Med. 11:75–83 (2004).
55. Williams, R.H., E. Ward, and H.A. McCartney: Methods for integratedair sampling and DNA analysis for detection of airborne fungal spores.Appl. Environ. Microbiol. 67:2453–2459 (2001).
56. Wu, Z., G. Blomquist, S.O. Westermark, and X.R. Wang: Applicationof PCR and probe hybridization techniques in detection of airborne fungalspores in environmental samples. J. Environ. Monit. 5:673–678 (2002).
57. Zeng, Q.Y., A. Rasmuson-Lestander, and X.R. Wang: Extensive set ofmitochondrial LSU rDNA-based oligonucleotide probes for the detectionof common airborne fungi. FEMS Microbiol Lett. 237:79–87 (2004).
58. Zeng, Q.Y., S.O. Westermark, A. Rasmuson-Lestander, and X.R.Wang: Detection and quantification of Wallemia sebi in aerosols by real-time PCR, conventional PCR, and cultivation. Appl. Environ. Microbiol.70:7295–7302 (2004).
Journal of Occupational and Environmental Hygiene December 2006 659
Dow
nloa
ded
by [
Uni
vers
ity o
f T
enne
ssee
At M
artin
] at
10:
53 0
8 O
ctob
er 2
014
