Prioritization before risk assessment: The viability of uncertain data on food contact materials.

The shortage of data on non-intentionally added substances (NIAS) present in food contact material (FCM) limits the ability to ensure food safety. Recent strategies in analytical method development permit NIAS investigation by using chemical exploration, but this has not been sufficiently investigated in risk assessment context. Here, exploration is utilized and followed by risk prioritization on chemical compounds that can potentially migrate to food from two paperboard FCM samples. Concentration estimates from exploration are converted to tentative exposure assessment, while predicted chemical structures are assessed using quantitative structure-activity relationships (QSAR) models for carcinogenicity, mutagenicity, and reproductive toxicity. A selection of 60 chemical compounds from two FCMs is assessed by four risk assessors to classify compounds based on probable risk. For almost 60% of cases, the assessors classified compounds as either high priority or low priority. Unclassified compounds are due to disagreements between experts (18%) or due to a perceived lack of data (23%). Among the high priority substances are high-concentration compounds, benzophenone derivatives, and dyes. The low priority compounds contained e.g. oligomers from plasticizers and linear alkane amides. The classification scheme provides valuable information based on tentative data and is able to prioritize discovered chemical compounds for pending risk assessment.

A combined metabolomic and phylogenetic study reveals putatively prebiotic effects of high molecular weight arabino-oligosaccharides when assessed by in vitro fermentation in bacterial communities derived from humans.

Prebiotic oligosaccharides are defined by their selective stimulation of growth and/or activity of bacteria in the digestive system in ways claimed to be beneficial for health. However, apart from the short chain fatty acids, little is known about bacterial metabolites created by fermentation of prebiotics, and the significance of the size of the oligosaccharides remains largely unstudied. By in vitro fermentations in human fecal microbial communities (derived from six different individuals), we studied the effects of high-mass (HA, >1 kDa), low-mass (LA, <1 kDa) and mixed (BA) sugar beet arabino-oligosaccharides (AOS) as carbohydrate sources. Fructo-oligosaccharides (FOS) were included as reference. The changes in bacterial communities and the metabolites produced in response to incubation with the different carbohydrates were analyzed by quantitative PCR (qPCR) and Liquid Chromatography-Mass Spectrometry (LC-MS), respectively. All tested carbohydrate sources resulted in a significant increase of Bifidobacterium spp. between 1.79 fold (HA) and 1.64 fold (FOS) in the microbial populations after fermentation, and LC-MS analysis suggested that the bifidobacteria contributed to decomposition of the arabino-oligosaccharide structures, most pronounced in the HA fraction, resulting in release of the essential amino acid phenylalanine. Abundance of Lactobacillus spp. correlated with the presence of a compound, most likely a flavonoid, indicating that lactobacilli contribute to release of such health-promoting substances from plant structures. Additionally, the combination of qPCR and LC-MS revealed a number of other putative interactions between intestinal microbes and the oligosaccharides, which contributes to the understanding of the mechanisms behind prebiotic impact on human health.

Antimicrobial effect of nisin in processed cheese - Quantification of residual nisin by LC-MS/MS and development of new growth and growth boundary model for Listeria monocytogenes.

This study tested the hypothesis that growth of Listeria monocytogenes in processed cheese with added nisin can be predicted from residual nisin A concentrations in the final product after processing. A LC-MS/MS method and a bioassay were studied to quantify residual nisin A concentrations and a growth and growth boundary model was developed to predict the antilisterial effect in processed cheese. 278 growth rates were determined in broth for 11 L. monocytogenes isolates and used to determine 13 minimum inhibitory concentration (MIC) values for nisin between pH 5.5 and 6.5. To supplement these data, 67 MIC-values at different pH-values were collected from the scientific literature. A MIC-term was developed to describe the effect of pH on nisin MIC-values. An available growth and growth boundary model (doi: https://doi.org/10.1016/j.fm.2019.103255) was expanded with the new MIC-term for nisin to predict growth in processed cheese. To generate data for model evaluation and further model development, challenge tests with a total of 45 growth curves, were performed using processed cheese. Cheeses were formulated with 11.2 or 12.0 ppm of nisin A and heat treated to obtain residual nisin A concentrations ranging from 0.56 to 5.28 ppm. Below 15 °C, nisin resulted in extended lag times. A global regression approach was used to fit all growth curves determined in challenge tests. This was obtained by combining the secondary growth and growth boundary model including the new term for the inhibiting effect of nisin on µmax with the primary logistic growth model with delay. This model appropriately described the growth inhibiting effect of residual nisin A and showed that relative lag times depended on storage temperatures. With residual nisin A concentrations, other product characteristics and storage temperature as input the new model correctly predicted all observed growth and no-growth responses for L. monocytogenes. This model can support development of nisin A containing recipes for processed cheese that prevent growth of L. monocytogenes. Residual nisin A concentrations in processed cheese were accurately quantified by the developed LC-MS/MS method with recoveries of 83 to 110% and limits of detection and quantification being 0.04 and 0.13 ppm, respectively. The tested bioassay was less precise and nisin A recoveries varied for 53% to 94%.

Standard substances free quantification makes LC/ESI/MS non-targeted screening of pesticides in cereals comparable between labs.

LC/ESI/MS is the technique of choice for qualitative and quantitative food monitoring; however, analysis of a large number of compounds is challenged by the availability of standard substances. The impediment of detection of food contaminants has been overcome by the suspect and non-targeted screening. Still, the results from one laboratory cannot be compared with the results of another laboratory as quantitative results are required for this purpose. Here we show that the results of the suspect and non-targeted screening for pesticides can be made quantitative with the aid of in silico predicted electrospray ionization efficiencies and this allows direct comparison of the results obtained in two different laboratories. For this purpose, six cereal matrices were spiked with 134 pesticides and analysed in two independent labs; a high correlation for the results with the R2 of 0.85.

Quantification for non-targeted LC/MS screening without standard substances.

Non-targeted and suspect analyses with liquid chromatography/electrospray/high-resolution mass spectrometry (LC/ESI/HRMS) are gaining importance as they enable identification of hundreds or even thousands of compounds in a single sample. Here, we present an approach to address the challenge to quantify compounds identified from LC/HRMS data without authentic standards. The approach uses random forest regression to predict the response of the compounds in ESI/HRMS with a mean error of 2.2 and 2.0 times for ESI positive and negative mode, respectively. We observe that the predicted responses can be transferred between different instruments via a regression approach. Furthermore, we applied the predicted responses to estimate the concentration of the compounds without the standard substances. The approach was validated by quantifying pesticides and mycotoxins in six different cereal samples. For applicability, the accuracy of the concentration prediction needs to be compatible with the effect (e.g. toxicology) predictions. We achieved the average quantification error of 5.4 times, which is well compatible with the accuracy of the toxicology predictions.

Exploring the chemistry of complex samples by tentative identification and semiquantification: A food contact material case.

In fields such as food safety and environmental chemistry, ensuring safety is greatly challenged by large numbers of unknown substances occurring. Even with current state-of-the-art mass spectrometers, dealing with nonidentified substances is a very laborious process as it includes structure elucidation of a vast number of unknowns, of which only a fraction may be relevant. Here, we present an exploration and prioritization approach based on high-resolution mass spectrometry. The method uses algorithm-based precursor/product-ion correlations on quadrupole time-of-flight tandem mass spectrometry data to retrieve the most likely chemical match from a structure database. In addition, time-of-flight-only data are used to estimate analyte concentration via semiquantification. The method is demonstrated in recycled paper food contact material. Here, 585 chromatographic peaks were discovered, of which 117 were unique to the sample and could be tentatively elucidated via accurate mass, isotopic pattern, and precursor/product-ion correlations. Nearly 85% of these 117 peaks were matched with database entries, which provided varying certainty of information about the analyte structure. Semiquantitative concentration ranges of investigated compounds were between 0.7 and 1600 µg dm-2 . With these data, a subgroup of chemicals was risk-categorized and prioritized by using the most likely candidate structure(s) obtained. Prioritization based on expected health impact was possible by using the tentatively assigned data. Overall, the described method not only is a valuable chemical exploration tool for nonidentified substances but may also be used as a preliminary prioritization tool for substances expected to have the highest health impact, for example, in food contact materials.

Elucidating the mode-of-action of compounds from metabolite profiling studies.

Metabolite profiling has been carried out for decades and is as such not a new research area. However, the field has attracted increasing attention in the last couple of years, and the term metabolome is now often used to describe the complete pool of metabolites associated with an organism at any given time. Mass spectrometry (MS) and nuclear magnetic resonance (NMR) spectroscopy are the best candidates for comprehensive analysis of the metabolome and the application of these technologies is presented in this chapter. In this relation, the importance of efficient metabolite screening for discovery of novel drugs is discussed. Related to metabolite profiling, the principals underlying the application of labeled substrates to quantify in vivo metabolic fluxes are introduced, and the chapter is concluded by discussing the perspectives of metabolite measurements in systems biology.

A proposed framework for the description of plant metabolomics experiments and their results.

The study of the metabolite complement of biological samples, known as metabolomics, is creating large amounts of data, and support for handling these data sets is required to facilitate meaningful analyses that will answer biological questions. We present a data model for plant metabolomics known as ArMet (architecture for metabolomics). It encompasses the entire experimental time line from experiment definition and description of biological source material, through sample growth and preparation to the results of chemical analysis. Such formal data descriptions, which specify the full experimental context, enable principled comparison of data sets, allow proper interpretation of experimental results, permit the repetition of experiments and provide a basis for the design of systems for data storage and transmission. The current design and example implementations are freely available (http://www.armet.org/). We seek to advance discussion and community adoption of a standard for metabolomics, which would promote principled collection, storage and transmission of experiment data.

Fumonisin B2 production by Aspergillus niger.

The carcinogenic mycotoxin fumonisin B2 was detected for the first time in the industrially important Aspergillus niger. Fumonisin B2, known from Fusarium verticillioides and other Fusaria, was detected in cultures of three full genome sequenced strains of A. niger, in the ex type culture and in a culture of F. verticillioides by electrospray LC-MS analysis of methanolic extracts from agar plugs of cultures grown on several substrates. Whereas F. verticillioides produced fumonisins B1, B2, and B3 on agar media based on plant extracts, such as barley malt, oat, rice, potatoes, and carrots, A. niger produced fumonisin B2 best on agar media with a low water activity, including Czapek yeast autolysate agar with 5% NaCl. Of the media tested, only rice corn steep agar supported fumonisin production by both F. verticillioides and A. niger. However, A. niger had a different regulation of fumonisin production and a different quantitative profile of fumonisins, producing only B2 as compared to F. verticillioides. Fumonisin production by A. niger, which is a widely occurring species and an extremely important industrial organism, will have very important implications for biotechnology and especially food safety. A. niger is used for the production of citric acid and as producer of extracellular enzymes, and also as a transformation host for the expression of heterologous proteins. Certain strains of A. niger produce both ochratoxin A and fumonisins, so some foods and feeds may potentially contain two types of carcinogenic mycotoxins from this species.

A framework to estimate concentrations of potentially unknown substances by semi-quantification in liquid chromatography electrospray ionization mass spectrometry.

Risk assessment of exposure to chemicals from food and other sources rely on quantitative information of the occurrence of these chemicals. As screening analysis is increasingly used, a strategy to semi-quantify unknown or untargeted analytes is required. A proof of concept strategy to semi-quantifying unknown substances in LC-MS was investigated by studying the responses of a chemically diverse marker set of 17 analytes using an experimental design study. Optimal conditions were established using two optimization parameters related to weak-responding compounds and to the overall response. All the 17 selected analytes were semi-quantified using a different analyte to assess the quantification performance under various conditions. It was found that source conditions had strong effects on the responses, with the range of low-response signals varying from -80% to over +300% compared to centerpoints. Positive electrospray (ESI+) was found to have more complex source interactions than negative electrospray (ESI-). Choice of quantification marker resulted in better quantification if the retention time difference was minimized (12 out of 12 cases error factor < 4.0) rather than if the accurate mass difference was minimized (7 out of 12 cases error factor < 4.0). Using optimal conditions and retention time selection, semi-quantification in ESI+ (70% quantified, average prediction error factor 2.08) and ESI- (100% quantified, average prediction error factor 1.74) yielded acceptable results for untargeted screening. The method was successfully applied to an extract of food contact material containing over 300 unknown substances. Without identification and authentic standards, the method was able to estimate the concentration of a virtually unlimited number of compounds thereby providing valuable data to prioritize compounds in risk assessment studies.

Real-time PCR quantification of the AM-toxin gene and HPLC qualification of toxigenic metabolites from Alternaria species from apples.

Some Alternaria species are able to produce plant pathogenic as well as toxic metabolites. In both agriculture and the food industry it is important know if toxigenic Alternaria are present to rapidly employ the correct corrective actions. The purpose of this work was to establish a real-time PCR method, which can detect and quantify apple pathogenic and toxigenic Alternaria. An AM-toxin I primer set, which could recognize Alternaria DNA only, was designed by using primers complementary to the AM-toxin I gene. The method could detect small amounts of DNA (4 pg) and still obtain a large dynamic range (4 decades) without interference from apple material. Eight Alternaria isolates were analyzed for the presence of AM-toxin I gene and their production of secondary metabolites. Then analyses showed that all eight isolates contained the AM toxin gene and were able to produce the plant pathogenic tentoxin in addition to AM toxin I. The analyses also showed the production of tenuazonic acid, alternariols, Altenuene, altenusin and/or altertoxin I in pure culture. Analyses of inoculated apples showed that both the AM-toxin gene and alternariol monomethyl ether could be detected. Morphological analyses suggested that the eight Alternaria strains, though they all carried the AM toxin genes, probably belong to different but closely related un-described Alternaria taxa in the A. tenuissima species-group based on morphological and chemical differences.

Soyasaponins resist extrusion cooking and are not degraded during gut passage in Atlantic salmon (Salmo salar L.).

The stability of soyasaponins in fish feed formulations was investigated. The level of soyasaponin Ab, Bb, Bc, Ba-2,3-dihydro-2,5-dihydroxy-6-methyl-4H-pyran-4-one (Ba-DDMP), Bb-DDMP, and Bc-DDMP was quantified in 15 samples of defatted soybean meal, two full fat soybean meals, and two soybean protein concentrates by reverse phase high-performance liquid chromatography. The total level of saponins in the 15 samples of commercial defatted soybean meal ranged from 4.8-6.8 micromol/g (5.1-7.0 g/kg). The two full fat meals contained 4.4 and 4.7 micromol/g whereas no saponins could be detected in the alcohol-extracted soybean protein concentrates. Fifteen batches of fish feed containing 20% defatted soybean meal were produced by twin-screw extrusion from the 15 different samples of defatted soybean meal. Extrusion did not reduce the total level of group B saponins, but the ratio between DDMP-conjugated group B saponins and non-DDMP-conjugated group B saponins was slightly reduced. A soybean-containing diet was fed to seawater adapted Atlantic salmon for 9 weeks. Yttrium oxide was included in the feed as an inert marker in order to estimate the disappearance of saponins during gut passage. High levels of intact non-DDMP-conjugated group B soyasaponins were found in feces whereas only low levels of DDMP-conjugated saponins could be detected. The overall disappearance of saponins was close to zero, and the concentration of intact saponins in dry feces reached levels several fold higher than dietary levels. The present work demonstrates that non-DDMP-conjugated group B soyasaponins resist extrusion cooking and remain intact during gut passage in Atlantic salmon. The latter is contrary to earlier findings in endothermic animals.

Automated and unbiased classification of chemical profiles from fungi using high performance liquid chromatography.

In this paper we present a method for unbiased/unsupervised classification and identification of closely related fungi, using chemical analysis of secondary metabolite profiles created by HPLC with UV diode array detection. For two chromatographic data matrices a vector of locally aligned full spectral similarities is calculated along the retention time axis. The vector depicts the evaluating of the alikeness between two fungal extracts based upon eluted compounds and corresponding UV-absorbance spectra. For assessment of the chemotaxonomic grouping the vector is condensed to one similarity describing the overall degree of similarity between the profiles. Two sets of data were used in this study: One set was used in the method development and a second dataset used for method validation. First we developed a method for evaluating the secondary metabolite production from closely related Penicillium species. Then the algorithm was validated on fungal isolates belonging to the genus Alternaria. The results showed that the species may be segregated into taxa in full accordance with published taxonomy.

Automated and Unbiased Image Analyses as Tools in Phenotypic Classification of Small-Spored Alternaria spp.

ABSTRACT For more than 25 years, controversy has surrounded the characterization and differentiation of small-spored Alternaria spp. And, therefore, the application of names of several species that are involved in the pathology of diseases related to host-specific toxin production. The name A. alternata often has been broadly applied to various morphologically and chemically distinct groups of isolates from different hosts. The purpose of this study was to develop and evaluate automated and unbiased image analysis systems that will analyze different phenotypic characters and facilitate testing and application of the morphological species concept in Alternaria taxonomy. Host-specific toxin-producing Alternaria isolates assigned to five morpho-species were compared with representative isolates of morphologically distinct A. alternata. Combined results of growth rates at different temperatures, colony morphology, and metabolite profiles were found to be useful in characterization and differentiation of small-spored Alternaria spp. when standardized conditions are applied and representative isolates employed for comparison.

Andrastins A-D, Penicillium roqueforti Metabolites consistently produced in blue-mold-ripened cheese.

This is the first finding of andrastins in blue cheese as well as any other sample type. Here, they were produced by the secondary starter culture Penicillium roqueforti. After purification by normal-phase chromatography followed by combined reverse-phase ion-exchange chromatography, the andrastins A-D were detected by liquid chromatography combined with UV and high-resolution mass spectrometry. In 23 representative samples of European blue cheeses, andrastin A was consistently found in quantities between 0.1 and 3.7 microg/g of cheese (median 2.4 microg/g). Assuming the same molar response factors as for andrastin A, the B, C, and D analogues were present in approximately 5-, 3-, and 5-20-fold lower amounts than andrastin A, respectively. The andrastins are protein farnesyltransferase inhibitors and are capable of inhibiting the efflux of anticancer drugs from multidrug-resistant cancer cells. Thus, their presence in common blue cheese suggests a potential for a positive or negative impact on human health.

LC-MS analysis of the plasma metabolome--a novel sample preparation strategy.

Blood plasma is a well-known body fluid often analyzed in studies on the effects of toxic compounds as physiological or chemical induced changes in the mammalian body are reflected in the plasma metabolome. Sample preparation prior to LC-MS based analysis of the plasma metabolome is a challenge as plasma contains compounds with very different properties. Besides, proteins, which usually are precipitated with organic solvent, phospholipids, are known to cause ion suppression in electrospray mass spectrometry. We have compared two different sample preparation techniques prior to LC-qTOF analysis of plasma samples: the first is protein precipitation; the second is protein precipitation followed by solid phase extraction with sub-fractionation into three sub-samples: a phospholipid, a lipid and a polar sub-fraction. Molecular feature extraction of the data files from LC-qTOF analysis of the samples revealed 1792 molecular features from the protein precipitation procedure. The protein precipitation followed by solid phase extraction procedure with three sub-samples gave a total of 4234 molecular features. This suggests that sub-sampling into polar, lipid and phospholipid fractions enables extraction of more metabolomic information as compared to protein precipitation alone. Chromatography showed good separation of the metabolites with little retention time drift (<1s) and a mass accuracy below 3 ppm was observed. The performance of the method was investigated using plasma samples from rats administered the environmental pollutant perfluorononanoic acid.

A new matching algorithm for high resolution mass spectra.

We present a new matching algorithm designed to compare high-resolution spectra. Whereas existing methods are bound to compare fixed intervals of ion masses, the accurate mass spectrum (AMS) distance method presented here is independent of any alignment. Based on the Jeffreys-Matusitas (JM) distance, a difference between observed peaks across pairs of spectra can be calculated, and used to find a unique correspondence between the peaks. The method takes into account that there may be differences in resolution of the spectra. The algorithm is used for indexing in a database containing 80 accurate mass spectra from an analysis of extracts of 80 isolates representing the nine closely related species in the Penicillium series Viridicata. Using this algorithm we can obtain a retrieval performance of approximately 97-98% that is comparable with the best of the existing methods (e.g., the dot-product distance). Furthermore, the presented method is independent of any variable alignment procedures or binning.

Fungal metabolite screening: database of 474 mycotoxins and fungal metabolites for dereplication by standardised liquid chromatography-UV-mass spectrometry methodology.

A standardised LC-UV-MS micro-scale method for screening of fungal metabolites and mycotoxins in culture extracts is presented. The paper includes data for detection and dereplication of > 400 fungal metabolites to facilitate detection and identification when standards are not available. The data also shows the types of components that can be analysed by positive electrospray (ESI+) mass spectrometry (MS) along with common fragments and adducts of these, as well as giving suggestions on whether UV or ESI+-MS methods should be used. Examples of dereplication of penitrems and macro-cyclic ichothecenes, and detection of several novel compounds are shown. This was done by UV spectroscopy combined with accurate mass determination of adduct and fragment ions obtained by high-resolution orthogonal time-of-flight MS.

Penicillium expansum: consistent production of patulin, chaetoglobosins, and other secondary metabolites in culture and their natural occurrence in fruit products.

Penicillium expansum is known for its destructive rot and patulin production in apple juice. According to the literature, P. expansum can, among other compounds, produce citrinin, ochratoxin A, patulin, penitrem A, and rubratoxin B. In this study the qualitative production of metabolites was examined using TLC (260 isolates), HPLC (85 isolates), and MS (22 isolates). The results showed that none of the 260 isolates produced ochratoxin A, penitrem A, or rubratoxin B. However, chaetoglobosin A and communesin B were produced consistently by all 260 isolates. Patulin and roquefortine C were produced by 98% of the isolates. Expansolides A/B and citrinin were detected in 91 and 85% of the isolates, respectively. Chaetoglobosins and communesins were detected in naturally infected juices and potato pulp, whereas neither patulin nor citrinin was found. Because most P. expansum isolates produce patulin, citrinin, chaetoglobosins, communesins, roquefortine C, and expansolides A and B, foods contaminated with this fungus should ideally be examined for chaetoglobosin A as well as patulin.

Dynamic Metabolic Footprinting Reveals the Key Components of Metabolic Network in Yeast Saccharomyces cerevisiae.

Metabolic footprinting offers a relatively easy approach to exploit the potentials of metabolomics for phenotypic characterization of microbial cells. To capture the highly dynamic nature of metabolites, we propose the use of dynamic metabolic footprinting instead of the traditional method which relies on analysis at a single time point. Using direct infusion-mass spectrometry (DI-MS), we could observe the dynamic metabolic footprinting in yeast S. cerevisiae BY4709 (wild type) cultured on 3 different C-sources (glucose, glycerol, and ethanol) and sampled along 10 time points with 5 biological replicates. In order to analyze the dynamic mass spectrometry data, we developed the novel analysis methods that allow us to perform correlation analysis to identify metabolites that significantly correlate over time during growth on the different carbon sources. Both positive and negative electrospray ionization (ESI) modes were performed to obtain the complete information about the metabolite content. Using sparse principal component analysis (Sparse PCA), we further identified those pairs of metabolites that significantly contribute to the separation. From the list of significant metabolite pairs, we reconstructed an interaction map that provides information of how different metabolic pathways have correlated patterns during growth on the different carbon sources.