Alternaria toxins and conjugates in selected foods in the Netherlands

Accepted Manuscript Alternaria toxins and conjugates in selected foods in the Netherlands Patricia López, Dini Venema, Hans Mol, Martien Spanjer, Joyce de Stoppelaar, Erika Pfeiffer, Monique de Nijs PII:

S0956-7135(16)30170-0

DOI:

10.1016/j.foodcont.2016.04.001

Reference:

JFCO 4960

To appear in:

Food Control

Received Date: 21 February 2016 Revised Date:

27 March 2016

Accepted Date: 1 April 2016

Please cite this article as: López P., Venema D., Mol H., Spanjer M., de Stoppelaar J., Pfeiffer E. & de Nijs M., Alternaria toxins and conjugates in selected foods in the Netherlands, Food Control (2016), doi: 10.1016/j.foodcont.2016.04.001. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

1

Alternaria toxins and conjugates in selected foods in the Netherlands.

2

Patricia Lópeza*, Dini Venemaa, Hans Mola, Martien Spanjerb, Joyce de Stoppelaarb, Erika

3

Pfeifferc, Monique de Nijsa

4

6

RI PT

5 a

RIKILT – Wageningen UR, Akkermaalsbos 2, 6708 WB, Wageningen, The Netherlands

b

NVWA – Netherlands Food and Consumer Product Safety Authority, Catherijnesingel 59,

7 8 9

3511 GG Utrecht, The Netherlands

12

c

Institute of Applied Biosciences, Department of Food Chemistry and Phytochemistry,

Karlsruhe Institute of Technology (KIT), Adenauerring 20a, 76131 Karlsruhe, Germany

M AN U

11

SC

10

13 14

*

Corresponding author: [email protected]

AC C

EP

TE D

15

1

ACCEPTED MANUSCRIPT

ABSTRACT

17

A survey on Alternaria toxins in the food categories dried figs (n=14), sunflower products

18

(n=24) and tomato products (n=43) was carried out in the Netherlands on samples collected in

19

retail stores in autumn 2014.

20

The occurrence data from this survey confirmed the previously reported data of a pilot survey

21

in 2013, with tenuazonic acid being an ubiquitous Alternaria toxin in dried figs (100% of the

22

samples), sunflower seeds (80% of the samples) and tomato products (60% of samples) at

23

relatively high concentrations, up to 1,728 µg/kg in dried figs and 1,350 µg/kg in sunflower

24

seeds. Despite the occurring of high concentrations of tenuazonic acid in sunflower seeds and

25

figs, it is unlikely that the population in the Netherlands is exposed to levels of concern.

26

Alternariol was detected in 27% of the tomato products ranging from LOQ (2 µg/kg) up to 26

27

µg/kg, while alternariol monomethyl ether was present in 7% of the tomato products. In the

28

worst case situation, consumption of these products may result in exposure at levels above the

29

threshold of toxicological concern of 2.5 ng/kg body weight/day for alternariol. None of seven

30

conjugates of Alternaria toxins was detected above the limit of quantification (LOQ) of 2.5

31

µg/kg.

M AN U

SC

RI PT

16

32 33

TE D

36

EP

35

Keywords: Alternaria toxins, conjugates, dried figs, sunflower, tomato processed products.

AC C

34

2

ACCEPTED MANUSCRIPT

1.

INTRODUCTION

38

Alternaria toxins are mycotoxins produced by Alternaria species, pathogenic and saprophytic

39

fungi widely distributed in soil. Alternaria species have been reported to occur in cereals

40

(mainly in wheat, sorghum and barley), oilseeds, such as sunflower and rapeseed, tomatoes,

41

apples, citrus fruits, olives and several other fruits and vegetables (EFSA, 2011). They grow

42

at low temperature; hence they are generally associated with extensive spoilage during

43

refrigerated transport and storage (Ostry, 2008).

44

Alternaria species can produce more than 70 secondary metabolites, but only a small

45

proportion of them have been chemically characterized and reported as toxic to humans and

46

animals (Barkai-Golan, 2008). However, there are currently no harmonised legal limits for

47

Alternaria toxins in food products in the EU. The European Food Safety Authority (EFSA)

48

published their opinion in 2011 on the risks for animal and public health related to the

49

presence of Alternaria toxins in feed and food (EFSA, 2011). EFSA evidenced a lack of

50

robust occurrence data of Alternaria toxins in food and processed products, and recommended

51

the collection of representative data across Europe. Following this conclusion, the Scientific

52

Committee on the Food Chain and Animal Health identified the following five toxins of

53

concern for human health: alternariol (AOH), alternariol monomethyl ether (AME),

54

tenuazonic acid (TeA), tentoxin (TEN) and altenuene (ALT). Other groups of Alternaria

55

toxins, such as altertoxins (ATX) and Alternaria alternate f. sp. lycopersici toxins (AAL

56

toxins), were deemed to be of less relevance (Standard Committee on the Food Chain and

57

Animal Health, 2012).

58

In their opinion, EFSA stated that the risk for public health related to these mycotoxins is not

59

expected to result from acute toxicity (EFSA, 2011). However, although AOH and AME

60

showed no strong acute toxic effects, they could induce fetotoxic and teratogenic effects

61

(EFSA, 2011). AOH, AME and especially ATX II showed genotoxic effects in cultured

62

mammalian cells (Fleck, Burkhardt, Pfeiffer, & Metzler, 2012). As the toxicity data were not

63

sufficient for identification of reference points for different toxicological effects, EFSA

64

decided to use the threshold of toxicological concern (TTC) approach to assess the relative

65

level of concern for dietary exposure of humans. For AOH and AME the TTC value of 2.5

66

ng/kg b.w./day was applied, for TEN and TeA the TTC value of 1500 ng/kg b.w./day (EFSA,

67

2011). In their risk assessment, the estimated chronic dietary exposure to AOH and AME

68

exceeded the TTC value, which emphasized the need for more additional data on the toxicity

69

of Alternaria toxins (EFSA, 2011).

AC C

EP

TE D

M AN U

SC

RI PT

37

3

ACCEPTED MANUSCRIPT

In response to the EFSA recommendation on the need for more data on the occurrence of

71

Alternaria toxins, a survey was conducted in 2013 in the Netherlands, on 95 retail products

72

classified in 10 different food categories, to identify the food groups of concern (López, et al.,

73

2016). The results from that survey showed that TeA occurred in 27% of the samples and at

74

highest concentrations in sunflower seeds, tomato sauces and dried figs. The prevalence of the

75

other Alternaria toxins (AOH, AME, TEN and ALT) was low in all food groups. It was

76

concluded that more occurrence data on each specific type of product were necessary for a

77

proper evaluation of the occurrence of Alternaria toxins (e.g. annual variations) and for the

78

exposure of the Dutch consumers to TeA through the consumption of these products.

79

The possible occurrence of conjugated or “masked” Alternaria toxins, in which the mycotoxin

80

is bound to a more polar substance like glucose or sulfate, has been pointed out (Ostry, 2008).

81

Conjugates of Alternaria toxins may be of relevance to exposure, as they may release the

82

native mycotoxin after hydrolysis in the human digestive tract. Hildebrand et al. have recently

83

identified five β-D-glucopyranosides and a 6’-malonyl-β-D-glucopyranoside of AOH, as well

84

as a D-glucopyranoside and 6’- and 4’-malony-β-D-glucopyranosides of AME in tobacco

85

suspension cells (Hildebrand, et al., 2015). The same authors observed that in plant cells

86

AME glucosides underwent malonyl conjugation, whereas AOH glucosides underwent

87

glucosylation (Hildebrand, et al., 2015). AOH and AME sulfate conjugates were detected in

88

tomato products and glucoside conjugates in sesame seeds in a survey performed in Belgium

89

(Walravens, et al., 2014b).

90

This work describes the results from a survey carried out in 2014 in the Netherlands for the

91

occurrence of five Alternaria toxins (AOH, AME, TeA, TEN and ALT) in tomato products,

92

sunflower products and dried figs. These products were identified as vulnerable to

93

contamination with Alternaria toxins in a pilot study. The potential presence of seven

94

conjugate forms of AOH and AME, i.e. AOH-9-glucoside (AOH9G), AOH-3-glucoside

95

(AOH3G), AOH-9-diglucoside (AOH9DG), AOH-3-sulfate (AOH3S), AME-3-glucoside

96

(AME3G), AME-3-malonylglucoside (AME3MG) and AME-3-sulfate (AME3S) was also

97

monitored.

AC C

EP

TE D

M AN U

SC

RI PT

70

98 99

2.

MATERIALS AND METHODS

100 101

2.1. Chemicals and reagents

102

Acetonitrile and methanol, both UPLC grade, were purchased from Actu-all Chemicals (Oss,

103

the Netherlands), formic acid (98–100%) from Merck (Amsterdam, the Netherlands), and 4

ACCEPTED MANUSCRIPT

ammonium formate (99%) from Across Organics (Geel, Belgium). MilliQ water (>18MΩ)

105

was used (Millipore, Bedford, MA, USA). AOH and AME were purchased from Biosolve

106

(Greyhound Chromatography and Allied Chemicals, Merseyside, UK). TeA, TEN and ALT

107

were supplied by Sigma-Aldrich (Zwijndrecht, the Netherlands). All standards had a purity

108

higher than 99%. Individual stock solutions of each mycotoxin at 100 µg/mL in acetonitrile

109

and in methanol for TeA were prepared. A solution of ALT of 1 mg/mL was provided by the

110

Chemistry laboratory of the Netherlands Food and Consumer Product Safety Authority. A

111

mixture working solution containing AOH, AME, ALT and TEN at 1 µg/mL and TeA at 5

112

µg/mL in acetonitrile was prepared.

113

standard.

114

The AOH and AME conjugates (AOH9G, AOH3G, AOH9DG, AME3G, AME3MG) were

115

isolated and provided by the Institute of Applied Bioscience (Karlsruhe Institute of

116

Technology, Germany). Figure 1 shows the structure of these conjugates. A mix stock

117

solution containing the conjugates in concentrations ranging from 0.8 to 2 µg/mL was

118

prepared in methanol.

RI PT

104

13

M AN U

SC

C-Caffeine (Sigma-Aldrich) was used as internal

119

2.2. Samples

121

A total of 81 samples was purchased from retail stores in the Netherlands in the autumn of

122

2014. The samples covered the food categories dried figs (n=14), processed tomato products

123

(15 tomato juices, 13 tomato sauces, 5 tomato ketchups and 10 other processed tomato

124

samples) and sunflower products (11 sunflower oils, 3 sunflower pastes and 10 roasted

125

shelled sunflower seed samples). The samples were stored at the temperature recommended

126

on the label until processing. Sunflower paste and sunflower oil were stored as such at room

127

temperature. Processed tomato samples were homogenized and stored at -20 °C. Dried figs

128

and sunflower seeds were homogenized by milling and mixing with water to form a slurry.

129

The water content in ratio was 1:3 for dried figs and 1:4 for sunflower seeds.

EP

AC C

130

TE D

120

131

2.3. Method

132

An aliquot of 2.5 g of tomatoes, sunflower oil or sunflower pasta were weighed, spiked with

133

12.5 µL of internal standard (13C-caffeine at 10 µg/mL) and extracted with 10 mL of

134

acetonitrile/water (84/16, v/v) acidified with 1% of acetic acid. 10 g of the slurry of figs and

135

12.5 g of the slurry of sunflower seeds were weighed, spiked with 12.5 µL of internal

136

standard (13C-caffeine at 10 µg/mL) and extracted with 10 mL of acetonitrile containing 1%

137

of acetic acid. Samples were shaken for two hours at 250 rpm and then centrifuged for 10 min 5

ACCEPTED MANUSCRIPT

at 3,000 rpm. The extracts were filtered in mini-uniprep PTFE filter vials and analysed by

139

LC-MS/MS.

140

The LC-MS/MS instrument was a Shimadzu ‘Prominence’ HPLC system (Shimadzu Europa

141

GmbH, Duisburg, Germany), which consisted of a DGU-20A3 degasser, a SIL 20AC XR

142

autosampler, two LC-20 AD XR UFLC pumps, and a CTO-20AC column oven. This system

143

was coupled to an AB SCIEX QTRAP® 5500 mass spectrometer (AB Sciex Netherlands

144

B.V., Nieuwerkerk aan den Ijssel, The Netherlands).

145

HPLC separation was performed by injecting 5 µL of the extract into a Waters Atlantis HSS-

146

T3, (3µm), 3.0 x 100 mm column (Waters, Milford, MA, USA). The chromatographic

147

conditions were as follows: column temperature 35 °C; mobile phases consisting of water as

148

eluent A and 4% (v/v) water in methanol as eluent B. Both mobile phases contained 1 mM of

149

ammonium formate and 1% of formic acid. The flow rate was set at 0.4 mL/min. A gradient

150

elution was applied as follows: after an initial hold time of 1 min at 100% eluent A, 50%

151

eluent B was reached within 2 min, which was kept for 1 min, and 100% eluent B within the

152

next 5 min. This composition was kept for 2 min, after which 100% eluent A was reached

153

within the next 0.5 min and kept for 4.5 min for column re-equilibration. The gradient used

154

before (López, et al., 2016) was slightly modified to allow the determination of the

155

conjugates.

156

The analysis of the samples was carried out using electrospray ionisation (ESI) and multiple

157

reaction monitoring (MRM). The precursor and product ion transitions for the five Alternaria

158

toxins were the same as previously used. The determination of the precursor and parent ion

159

for the Alternaria toxin glucoside conjugates was carried out by infusion of a solution

160

containing the mixed conjugate standard. Optimization was conducted in both positive and

161

negative ionisation modes. The transitions of the Alternaria toxin sulfate-conjugates were also

162

monitored. Their transitions were taken from literature (Walravens, et al., 2014a). Table 1

163

summarizes the retention times (RTs) and the applied values for fragmentation voltages and

164

collision.

SC

M AN U

TE D

EP

AC C

165

RI PT

138

166

2.4. Sample analysis

167

The Alternaria toxins in the samples were quantified by standard addition. The addition was

168

10 µg/kg of AOH, AME, TEN and ALT and 50 µg/kg of TeA. If necessary, the addition was

169

repeated at higher levels (equivalent to 50 µg/kg for AOH, AME, TEN and ALT and 250

170

µg/kg for TeA) and/or the extracts diluted. Quality control samples for each matrix under

171

study were included in each analytical series. 6

ACCEPTED MANUSCRIPT

Recoveries for quality controls fell between 50-60% for each toxin in sunflower products, and

173

between 80-120% for each toxin in tomato products and dried figs. As the standard addition

174

procedure already takes into account the losses due to extraction and matrix effects in the LC-

175

MS/MS instrument, the results were not corrected for recovery.

176

The LOQs for each Alternaria toxin were estimated in previous work (López, et al., 2016) as

177

the concentration in a spiked sample, for which the signal-to-noise (S/N) ratio of the

178

quantifier ion was equal to ten, providing that the signal of the qualifier ion was equal to

179

three. The LOQs were 1.5 µg/kg for ALT, 2.0 µg/kg for AOH, 1.0 µg/kg for AME, 2.5 µg/kg

180

for TEN y 5.0 µg/kg for TeA.

181

The Alternaria toxins and conjugates were identified by similarity of retention times and ion

182

ratios between the compounds in the sample extract and the standards in the matrix. The

183

criteria used to ensure the correct identification of the target mycotoxins were according to

184

SANTE guidelines (SANTE, 2015): (a) the retention times of the analyte in the sample extract

185

matched those of spiked sample extract within a deviation of ±0.2 min; (b) the deviation of

186

the ion ratio of the sample extract fell within ±30% of the ion ratio for the spiked sample

187

extract.

M AN U

SC

RI PT

172

188

2.5. Data treatment

190

Non-detects for calculation purposes were treated with the substitution method (EFSA, 2010).

191

In this study, the upper bound was followed, which means that results below LOD were

192

equalled to LOD and results between LOD and LOQ were equalled to LOQ.

193 194 195

3.

EP

TE D

189

RESULTS AND DISCUSSION

3.1. Optimization of the instrumental method for Alternaria toxin conjugates

197

Seven Alternaria toxin conjugates were included in the LC-MS/MS method as published by

198

Lopez et al. (López, et al., 2016). Their behaviour as regards separation and fragmentation

199

was studied for the conditions described in the materials and method section. Under those

200

conditions, the elution order of the Alternaria toxin conjugates is shown in Figure 2. The

201

AOH and AME conjugates eluted according to their decreasing polarity: AOH9DG, AOH9G,

202

AOH3G, AME3G and AME3MG.

203

The protonated molecule of AOH glucosides ([M + H]+ m/z 421) lost the hexose group during

204

fragmentation to produce the fragment m/z 259 [AOH + H]+, which in turn lost formic acid to

AC C

196

7

ACCEPTED MANUSCRIPT

form the fragment m/z 213. The fragmentation of the diglucoside conjugate of AOH ([M +

206

H]+ m/z 583) resulted in the sequential loss of two hexose moieties (162) to yield the product

207

ions m/z 421 ([glucoside + H]+ and m/z 259 [AOH + H]+.

208

Two different parent ions were chosen to detect AME3G, the protonated molecule [M + H]+

209

m/z 435 and the ammonium adduct [M+NH4]+ m/z 452. Both transitions provided the same

210

product ion [AME + H]+ m/z 273. For AME3MG only one transition was detectable, which

211

consisted of the loss of the malonylglucoside group to give the fragment [AME + H]+ m/z

212

273.

213

The transitions detected in this work were in agreement with those found when these

214

glucosides conjugates were first elucidated and detected in tobacco cells (Hildebrand, et al.,

215

2015). Walravens et al. (Walravens, et al., 2014a) analysed the AOH and AME conjugates

216

with negative electrospray ionization. The parent ion was for both Alternaria toxins the

217

deprotonated molecule [M – H]- and the resulting product ions were the deprotonated toxin,

218

either [AME –H]- or [AOH – H]-, and the deprotonated toxin with a loss of formic acid

219

moiety. Negative ionization was also assessed, but sensitivity was lower compared to positive

220

mode.

221

The method was not validated for the analysis of Alternaria toxin conjugates, due to the

222

limited amounts of standards available. Therefore, an LOQ of 2.5 µg/kg for the Alternaria

223

toxin conjugates was derived from the signal-to-noise ratio of injected standards in solvent.

TE D

M AN U

SC

RI PT

205

224

3.2. Occurrence of Alternaria toxins and their conjugates in the selected products

226

The results of the survey of Alternaria toxins and their conjugates in the 81 samples are

227

summarized in table 2 and table 3. Detailed information on the occurrence of Alternaria

228

toxins and their conjugates in each sample is presented in Supplementary Information SS1.

229

As described previously (López, et al., 2016), TeA was the most prevalent toxin, being

230

present in this study in 100% of dried figs, 80% of sunflower seeds and 60% of tomato

231

products.

232

TeA was the main Alternaria toxin in sunflower seeds, ranging from LOQ to 1,350 µg/kg.

233

AOH and AME were detected in one sample. These findings differed from reports by other

234

authors in which AOH, AME and TEN were present in this food commodity (EFSA, 2011;

235

Hickert, Bergmann, Ersen, Cramer, & Humpf, 2015). The processing of sunflower seeds

236

might have resulted in a reduction of TeA, since it was detected neither in sunflower oil nor in

237

sunflower paste samples in the present study. Sunflower oil is produced by pressing of seeds

238

and/or solvent extraction. Solvent extraction with organic solvent was proven to enhance in

AC C

EP

225

8

ACCEPTED MANUSCRIPT

particular the transfer of AME to oil, whereas TeA was hardly transferred to the oil (Chulze,

240

et al., 1995). Sunflower paste is produced by hardening (hydrogenation) of sunflower oil.

241

The results obtained for dried figs in the present survey confirmed the previous findings: all

242

analysed samples contained TeA at relatively high levels ranging from 81 to 1,728 µg/kg.

243

Despite the high levels of Alternaria toxins detected in sunflower seeds or figs in this study,

244

their consumption does not pose a risk for the health of the population in the Netherlands. In

245

the worst case scenario, a 70-kg person would need to eat more than 77 g of sunflower seeds

246

contaminated with 1,350 µg/kg of TeA, or 60 g of dried figs contaminated with 1,728 µg/kg

247

of TeA to exceed the level of concern (TTC) of 1,500 ng/kg b.w./day for TeA on routine basis

248

(77 g x 1,350 ng/g / 70 kg b.w. = 1485 ng/kg b.w./day of TeA).

249

TeA was the main Alternaria toxin detected in tomato products ranging from 5.0 to 344 µg/kg

250

(60% of samples), followed by AOH (27% of samples) and AME (7%). AME only occurred

251

in tomato paste, tomato sauces and ketchup samples, but not in tomato juices or processed

252

tomato pieces. The highest concentrations of TeA were found in tomato pastes (60%, ranging

253

from LOQ to 344 µg/kg) followed by tomato juices and tomato ketchup. These results are in

254

agreement with the results in the study by Noser et al. in products on the market in

255

Switzerland (Noser, Schneider, Rother, & Schmutz, 2011), but differed from the survey by

256

Zhao in China (Zhao, Shao, Yang, & Li, 2015), where tomato ketchup samples contained the

257

highest levels of TeA. The levels of AOH and AME were similar to previous studies (Hickert,

258

et al., 2015; López, et al., 2016; Noser, et al., 2011; Zhao, et al., 2015). The fact that TeA was

259

found at lower levels in tomato products containing whole tomato pieces may indicate that the

260

quality of the ingredients used for producing tomato pastes, tomato purees and tomato juices

261

might be an issue. No significant correlations between AOH and AME (r2 = 0.42) or AOH

262

and TeA (r2 = 0.39) were found in tomato samples. This may be due to the contamination of

263

the raw material with different Alternaria strains with variations in their biosynthetic profile

264

or the selective degradation of certain compounds during food processing (Hickert, et al.,

265

2015).

266

The levels of AOH and AME found in tomato products indicated that high consumption of

267

these products may pose a health risk for the population in the Netherlands since the daily

268

intake may exceed the TTC level of 2.5 ng/kg b.w./day. In the worst case scenario, the intake

269

of AOH would be above the threshold if a 70-kg person consumed 7 g of tomato paste

270

contaminated with 26 µg/kg of AOH (7 x 26 / 70 = 2.6 ng/kg b.w./day) or 10 g of tomato

271

ketchup contaminated with 18 µg/kg o1f AOH (10 x 18 / 70 = 2.6 ng/kg b.w./day). These data

AC C

EP

TE D

M AN U

SC

RI PT

239

9

ACCEPTED MANUSCRIPT

are in agreement with the risk assessment carried out in the Netherlands in 2013 (Sprong, et

273

al., 2015), which revealed that tomato products were one of the main contributors to the

274

exposure to Alternaria toxins and concluded that health risk based on the TTC approach for

275

AME or AOH could not be ruled out.

276

Masked mycotoxins or mycotoxin conjugates can be formed during food processing. None of

277

the seven Alternaria toxin conjugates were detected in the 81 samples in the current survey.

278

The presence of other mycotoxin conjugates in foodstuffs have been reported for

279

deoxynivalenol (DON) (3-acetylDON, 15-acetylDON and DON-3-glucoside), zearalenone,

280

fumonisins, ochratoxin A (Berthiller, et al., 2013) and HT-2 and T-2 toxins (Lattanzio, et al.,

281

2015). In case of DON and DON-3-glucoside, the ratio of occurrence may be quite stable,

282

although it can vary depending on the year and the plant genotype (Berthiller, et al., 2013).

283

Alternaria toxins sulphate conjugates have been only reported to occur in tomato products and

284

sesame seeds (Walravens, et al., 2014b). These authors detected AOH3S and AME3S in 23%

285

and 41% of tomato juice samples, in 14% and 27% of tomato sauces and in 33% and 76% of

286

tomato pastes, respectively, and co-occurring with AOH and AME. However, the authors did

287

not provide information on the occurrence of Alternaria toxins and conjugates in the

288

individual samples, and therefore a ratio between conjugated and non-conjugated forms, as

289

carried out for DON-3-glucoside and DON (Janssen, Sprong, Wester, De Boevre, &

290

Mengelers, 2015), could not be estimated. Taking into account the occurrence of AOH and

291

AME in the present study and the maximum concentration of AOH, AME and conjugates,

292

reported by Walravens, it is unlikely that AME3S would have been detected in the tomato

293

products in this study. However, AOH3S might have been detected in tomato pastes and/or

294

sauces at values close to the LOQ (2.5 µg/kg). Nevertheless, only the availability of high

295

quality commercial standards would allow a more accurate quantification on the occurrence

296

of Alternaria toxin conjugates.

297

The results of this survey and the previous survey (López, et al., 2016) indicate that

298

contamination of sunflower seeds, sunflower seed products and tomato products available on

299

the retail market in the Netherlands with Alternaria toxins seem to be relatively constant over

300

the years 2013-2104.

AC C

EP

TE D

M AN U

SC

RI PT

272

301 302 303

CONCLUSIONS

304

The study reported in this paper showed that TeA was present at relatively high

305

concentrations in tomato products, in dried figs and sunflower oils available in retail market in 10

ACCEPTED MANUSCRIPT

the Netherlands in 2014. Due to its hydrophilic properties, TeA seems to be lost during the

307

processing of seeds to obtain oil and/or sunflower paste. The data from this survey confirm

308

the results from a previous study in 2013 and indicate that in the Netherlands contamination

309

of sunflower seeds, dried figs and tomato products with Alternaria toxins seems to be

310

relatively constant over the years 2013-2104. The seven conjugated forms of AOH and AME

311

(sulfates and glucosides) were not detected in any of the samples.

312

The results showed that a risk to the health of consumers of high amounts of tomato products

313

in the Netherlands cannot be excluded. It is, therefore, advisable to monitor occurrence of

314

Alternaria toxins at least in these mentioned food categories.

SC

315

RI PT

306

ACKNOWLEDGMENTS

317

The authors gratefully acknowledge Alwin Kruijt, Food Safety Laboratory of the Netherlands

318

Food and Consumer Product Safety Authority, for kindly providing the altenuene standards,

319

Manfred Metzler and Ron Hoogenboom for the fruitful discussions. This study was carried

320

out in WOT project-02-001-061 ‘Method development and surveys on mycotoxins in food’ at

321

RIKILT Wageningen UR, on behalf of and funded by the Netherlands Ministry of Economic

322

Affairs.

M AN U

316

323

REFERENCES

325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355

Barkai-Golan, R. (2008). Alternaria Mycotoxins. In R. B.-G. a. N. Paster (Ed.), Mycotoxins in fruits and vegetables. San Diego, CA, USA: Elsevier. Berthiller, F., Crews, C., Dall'Asta, C., De Saeger, S., Haesaert, G., Karlovsky, P., Oswald, I. P., Seefelder, W., Speijers, G., & Stroka, J. (2013). Masked mycotoxins: A review. Molecular Nutrition & Food Research, 57(1), 165-186. Chulze, S. N., Torres, A. M., Dalcero, A. M., Etcheverry, M. G., Ramirez, M. L., & Farnochi, M. C. (1995). Alternaria mycotoxins in sunflower seeds: Incidence and distribution of the toxins in oil and meal. Journal of Food Protection, 58(10), 1133-1135.

EP

TE D

324

AC C

Summary report of the Standing Committe on the food chain and animal health (section Toxicological Safety of the food chain), held in Brussels on 29 May 2012. Alternaria toxins. EFSA. (2011). Scientific Opinion on the risks for animal and public health related to the presence of Alternaria toxins in feed and food EFSA Journal, 9(10), 97. EFSA (2010). Management of left-censored data in dietary exposure assessment of chemical substances. EFSA Journal 8(3) 1557. Fleck, S. C., Burkhardt, B., Pfeiffer, E., & Metzler, M. (2012). Alternaria toxins: Altertoxin II is a much stronger mutagen and DNA strand breaking mycotoxin than alternariol and its methyl ether in cultured mammalian cells. Toxicology Letters, 214(1), 27-32. Hickert, S., Bergmann, M., Ersen, S., Cramer, B., & Humpf, H.-U. (2015). Survey of Alternaria toxin contamination in food from the German market, using a rapid HPLC-MS/MS approach. Mycotoxin research, 1-12. Hildebrand, A. A., Kohn, B. N., Pfeiffer, E., Wefers, D., Metzler, M., & Bunzel, M. (2015). Conjugation of the Mycotoxins Alternariol and Alternariol Monomethyl Ether in Tobacco Suspension Cells. Journal of Agricultural and Food Chemistry, 63(19), 4728-4736. Janssen, E. M., Sprong, R. C., Wester, P. W., De Boevre, M., & Mengelers, M. J. B. (2015). Risk assessment of chronic dietary exposure to the conjugated mycotoxin deoxynivalenol-3-β-glucoside in the Dutch population World Mycotoxin Journal, 8(5), 561-572. Lattanzio, V. M. T., Ciasca, B., Terzi, V., Ghizzoni, R., McCormick, S. P., & Pascale, M. (2015). Study of the natural occurrence of T-2 and HT-2 toxins and their glucosyl derivatives from field barley to malt by high-resolution Orbitrap mass spectrometry. Food Additives and Contaminants Part a-Chemistry Analysis Control Exposure & Risk Assessment, 32(10), 1647-1655.

11

ACCEPTED MANUSCRIPT

RI PT

López, P., Venema, D., de Rijk, T., de Kok, A., Scholten, J. M., Mol, H. G. J., & de Nijs, M. (2016). Occurrence of Alternaria toxins in food products in The Netherlands. Food Control, 60, 196-204. Noser, J., Schneider, P., Rother, M., & Schmutz, H. (2011). Determination of six Alternaria toxins with UPLCMS/MS and their occurrence in tomatoes and tomato products from the Swiss market. Mycotoxin research, 27(4), 265-271. Ostry, V. (2008). Alternaria mycotoxins: an overview of chemical characterization, producers, toxicity, analysis and occurrence in foodstuffs. World Mycotoxin Journal, 1(2), 175-188. SANTE (2015). Guidance document on analytical quality control and method validation procedures for pesticides residues analysis in food and feed. SANTE/11945/2015. last accessed: 25-03-2016 http://ec.europa.eu/food/plant/docs/plant_pesticides_mrl_guidelines_wrkdoc_11945_en.pdf Sprong, R. C., de Wit-Bos, L., te Biesebeek, J. D., Alewijn, M., Lopez, P., & Mengelers, M. J. B. (2015). A mycotoxin-dedicated total diet study in the Netherlands in 2013: Part III – exposure and risk assessment World Mycotoxin Journal. Walravens, J., Mikula, H., Rychlik, M., Asam, S., Ediage, E. N., Di Mavungu, J. D., Van Landschoot, A., Vanhaecke, L., & De Saeger, S. (2014a). Development and validation of an ultra-high-performance liquid chromatography tandem mass spectrometric method for the simultaneous determination of free and conjugated Alternaria toxins in cereal-based foodstuffs. Journal of Chromatography A, 1372, 91101. Walravens, J., Mikula, H., Rychlik, M., Asam, S., Ediage, E. N., di Mavungu, J. D., Van Landschoot, A., Vanhaecke, L., & de Saeger, S. (2014b). Occurrence of (masked) Alternaria toxins - A survey in foodstuffs commercially available on the Belgian market. In 8th Conference of the World Mycotoxin Forum. Vienna, Austria. Zhao, K., Shao, B., Yang, D., & Li, F. (2015). Natural occurrence of four alternaria mycotoxins in tomato- and citrus-based foods in China. Journal of Agricultural and Food Chemistry, 63(1), 343-348.

SC

356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379

M AN U

380

AC C

EP

TE D

381

12

ACCEPTED MANUSCRIPT

382

Table 1. Retention times, the ionisation mode, declustering potential (DP), collision

383

energy (CE), collision cell exit potential (CXP) and the precursor and product ions for

384

the targeted Alternaria toxins and conjugates. ESI mode

RT

Transition

(min) 6.8

TeA

Negative

7.4

AOH

Negative

8.1

TEN

Negative

8.1

AME

Negative

9.3

Positive

6.9

Positive

7.0

Positive

8.0

Positive

6.4

Positive

8.3

AOH-9glucoside

glucoside AME-3glucoside

AME-3-

AME-3-sulfate

(V)

293.1>257.0 56

20

10

293.1>239.1 60

30

10

196.0>112.0 -35

-20

-11

196.0>139.0 -35

-20

-11

257.0>215.0 -5

-30

-15

257.0>213.0 -5

-26

-15

413.0>141.0 -60

-26

-11

-22

-17

271.0>256.0 -90

-32

-13

271.0>227.0 -90

-50

-9

421.0>259.0 186

25

18

421.0>213.0 186

63

14

421.0>259.0 186

25

18

421.0>213.0 186

63

14

435.0>273.0 126

33

18

452.0>273.0 126

33

18

583.0>259.0 121

49

18

583.0>421.0 121

15

30

521.0>273.0 121

19

18

Negative

337.0>257.0 -190

-10

-34

Positive

339.0>259.0 120

33

18

Negative

351.0>271.0 -170

-10

-42

Positive

353.0>273.0 230

33

13

AC C

malonylglucoside AOH-3-sulfate

(V)

413.0>271.0 -60

EP

AOH-9diglucoside

(V)

TE D

AOH-3-

CXP

SC

Positive

CE

M AN U

ALT

DP

RI PT

Mycotoxin

13

ACCEPTED MANUSCRIPT

Table 2. Occurrence of Alternaria toxins (µg/kg) in sunflower seeds, sunflower paste, sunflower oil and figs in the current survey Product

Toxin

# above LOQ

Sunflower seeds

AOH

(N=10)

Concentration (µg kg-1) Range

a

1 (10%)

<2.0-36

5.4

AME

1 (10%)

<1.0-1.7

1.1

1.0

TeA

8 (80%)

<5.0-1,350 240

95

ALT

0

<1.5

TEN

0

<2.5

Sunflower paste

AOH

0

<2.0

(N=3)

AME

0

<1.0

TeA

0

<5.0

ALT

0

TEN

0

Sunflower oil

AOH

0

(N=11)

AME

1 (9%)

<1.0-17

TeA

0

<5.0

ALT

0

TEN

0

<2.5

Figs

AOH

1(7%)

<2.0-8.7

(N=14)

AME

0

<1.0

TeA

14 (100%)

41-1,728

ALT

0

<1.5

TEN

0

<2.5

Median

M AN U

SC

RI PT

2.0

<2.5

<2.0

2.5

1.0

2.5

2.0

473

316

<1.5

TE D

EP

a

<1.5

Based on upper bound conditions. If concentration was between LOD and LOQ, then for the

AC C

a

Mean

level was equalled to LOQ. If concentration
14

ACCEPTED MANUSCRIPT

Table 3. Occurrence of Alternaria toxins (µg/kg) in tomato products from the current survey Tomato product

Toxin

# above LOQ

All tomato products

AOH

(N=43)

Concentration (µg/kg) Range

a

16 (37%)

<2.0-26

4.8

AME

4 (9%)

<1.0-5.6

TeA

26 (60%)

<5.0-344

ALT

0

<1.5

TEN

0

<2.5

Tomato sauces

AOH

3 (33%)

<2.0-16

4.0

2.0

(N=9)

AME

2 (22%)

<1.0-1.6

1.1

1.0

TeA

7 (78%)

<5.0-111

41

37

Tomato pastes

AOH

3 (60%)

<2.0-26

8.1

3.8

(N=5)

AME

1 (20%)

<1.0-5.6

1.9

1.0

TeA

4 (80%)

<5.0-344

131

111

Tomato ketchup

AOH

3 (60%)

<2.0-18

7.2

6.7

(N=5)

AME

1 (20%)

<1.0-5.6

1.9

1.0

Tomato pieces (sieved, cubes) a

Median

RI PT

2.0 1.0

63

36

SC

1.2

2 (40%)

<5.0-138

44

5.0

AOH

4 (28%)

<2.0-11

3.3

2.0

TeA

7 (50%)

<5.0-220

77

50

AOH

3 (30%)

<2.0-15

4.7

2.0

6 (60%)

<5.0-100

38

15

chopped, TeA

AC C

(N=10)

TE D

(N=14)

a

TeA

EP

Tomato juices

M AN U

Per category

Mean

Based on upper bound conditions. If concentration was between LOD and LOQ, then for the

level was equalled to LOQ. If concentration
15

ACCEPTED MANUSCRIPT

FIGURE CAPTIONS Figure 1. Chemical structure of the glucosides conjugates of alternariol (a) and alternariol methyl ether (b). Figure 2. Chromatography of a standard containing the AOH and AME conjugates at 25

D-glucopyranoside;

3:

AOH-3-O-β-D-glucopyranoside;

RI PT

ng/mL (1: AOH-9-O-β-D-glucopyranosyl-(6’→1’’)-β-D-glucopyranoside; 2: AOH-9-O-β4:

AC C

EP

TE D

M AN U

SC

glucopyranoside; 5: AME-3-O-(6’-O-malonyl-β-D-glucopyranoside).

16

AME-3-O-β-D-

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

HIGHLIGHTS

RI PT

SC M AN U TE D EP

•

Tenuazonic acid (TeA) occurs at high levels in sunflower seeds and dried figs in 2014. Processing of sunflower seeds to obtain oil or paste may degrade TeA. Alternariol (AOH), alternariol methyl ether (AME) and TeA occur in tomato products. The consumption of high amounts of tomato products may pose a risk for health in the Netherlands. Conjugates of Alternaria toxins could not be found at levels higher than 2.5 µg/kg.

AC C

• • • •