Labinfo NATIONAL REFERENCE LABORATORIES NRL Newsletter for the approved food safety laboratories SEMIANNUAL NEWSLETTER - N°14 DECEMBER 2015 FASFC AC-Kruidtuin - Food Safety Center, Kruidtuinlaan 55, 1000 Brussels Responsible editor : Herman Diricks 4 Detection of shiga-like toxin producing Escherichia coli, STEC, in food : methodology and bottlenecks 9 High resolution mass spectrometry 15 The usefulness of whole genome sequencing for outbreak investigation of food pathogens, Salmonella Enteritidis as a case study 19 Halal food: analysis methods to detect traces of pork in meat and meat products 22 Analysis of food allergens by LC-MS/MS 29 Emerging and Novel Brominated Flame Retardants (BFRs) in food: Current status of the European legislation 32 UGM characterization using DNA walking strategy 36 Workshops & Symposia
Transcript
Labinfo
LNRN A T I O N A L ER E F E R E N T I ELABORATORIA
L A B O R ATO I R E SN A T I O N A U XD E R E F E R E N C ENRLN A T I O N A L
R E F E R E N C ELABORATORIESNRL
Newsletter for the approved food safety laboratories
SEMIANNUAL NEWSLETTER - N°14 DECEMBER 2015
FASFCAC-Kruidtuin - Food Safety Center, Kruidtuinlaan 55, 1000 Brussels
Resp
onsi
ble
edito
r : H
erm
an D
irick
s
4 Detection of shiga-like toxin producing Escherichia coli, STEC, in food : methodology and bottlenecks
9 High resolution mass spectrometry
15 The usefulness of whole genome sequencing for outbreak investigation of food pathogens, Salmonella Enteritidis as a case study
19 Halal food: analysis methods to detect traces of pork in meat and meat products
22 Analysis of food allergens by LC-MS/MS
29 Emerging and Novel Brominated Flame Retardants (BFRs) in food: Current status of the European legislation
32 UGM characterization using DNA walking strategy
36 Workshops & Symposia
2
LabInfoNewsletter for the approved food safety laboratories
Editors’ groupDirk Courtheyn, Marnix De Gruyter, Conny De Schepper, Alain Dubois, Marc Evrard, Geert Janssens, Alain Laure, Kathleen Martens, Eva Wevers and Marie-Christine Wilem
Authors of this issueBert Matthijs, Sarah Denayer, Katelijne Dierick, Nadine Botteldoorn, Els Van Pamel, Els Daeseleire, Christof Van Poucke, Véro-nique Wuyts, Nancy Roosens, Wesley Mattheus, Sophie Bertrand, Kathleen Marchal, Sigrid De Keersmaecker, Rob Margry, Nathalie Gillard, Otto Gaëtan, Philippe Delahaut, Gauthier Eppe, Georges Scholl, Edwin De Pauw, Jean-François Focant, Ma-rie-Alice Fraiture, Herman Philippe and Nina Papazova
TranslationTranslation Service of the AgencyEditors’ group
Photographs and illustrationsSupplied by the laboratories
LNRN A T I O N A L ER E F E R E N T I ELABORATORIA
L A B O R ATO I R E SN A T I O N A U XD E R E F E R E N C ENRLN A T I O N A L
R E F E R E N C ELABORATORIESNRL
32
EditorialDear reader,
The year 2015 is drawing to a close. The FASFC has had quite a busy year. Our CEO’s business plan was approved by the Minister in charge of the FASFC, Minister Willy Borsus. Following this green light, each of the Agency’s administrations started working on a management plan to materialize the strategic objectives of the business plan. This time we were faced with a daunting challenge, considering the heavy budget cuts we had to implement.
Together with its middle management, the laboratories administration also contributed to this effort. The resulting management plan will serve as a guideline to improve our functioning for the years to come.
This management plan can basically be subdivided into two large parts. The first part per-tains to the functioning of the 5 FASFC laboratories, while the second part is concerned with the management of the National reference laboratories and the external laboratories.
In line with one of these objectives, we want to reassess our collaboration with the NRLs. To this effect, we have to consider the following questions: What do we expect from an NRL? Which tasks are essential? Which are nice-to-have? What are the NRLs’ expectations regarding the FASFC? In what ways can we improve our collaboration? In what ways can the NRLs offer more added value? How do we divide the budget? Are we setting the right priorities? Which indicators allow us to better follow up on and assess the work activities of the NRLs? Are there domains which currently have not been assigned to an NRL yet, even though this might be advisable? ...In short, a number of questions remain to be answered. The answers to these questions will determine the course to follow for the years to come regarding the NRLs.
From the articles in this issue of Labinfo, it will become clear that the NRLs have not been sitting idly by either. Among other things, they have been working on new methods and techniques to respond to new challenges.
I wish you a lot of reading pleasure with this fourteenth issue of Labinfo and wish you peaceful holidays and a happy New Year.
Bert MatthijsDirector-general Laboratories
4
Detection of shiga-like toxin producing Escherichia coli, STEC, in food : methodology and bottlenecksSarah Denayer, Katelijne Dierick and Nadine BotteldoornNRL Food microbiology, Rue Juliette Wytsmanstraat 14, 1050 Brussels
Introduction
Generally, E. coli is a harmless bacterium in the intestines of humans and animals. However, certain types can be pathogenic in humans and cause intestinal disorders. Depending on their virulence factors and on the syndrome these bacteria can cause, different groups are to be distinguished within the pathogenic E. coli. An important group consists of the shiga-like toxin producing Escherichia coli, also named STEC. The group of the enterohaem-orrhagic E. coli (EHEC) is part of these STEC, causing mild diarrhoea in humans, but which can progress to bloody diarrhoea. In 2-10% of the cases, it can lead to complications, such as haemolytic uraemic syndrome (HUS), requir-ing sometimes haemodialysis.The severeness of an infection caused by STEC depends on the presence of virulence factors, such as shiga-like toxins (encoded by the stx1 and stx2 genes), and other virulence factors playing a role in intestinal adhesion and invasion, such as for instance intimin (encoded by eae). In addition to E. coli O157:H7, which has already been demonstrated in different outbreaks and which has specific biochemical characteristics (sorbitol -), there are also other serotypes which can cause serious illnesses in humans.
With regard to the detection of this pathogen in food, the method (ISO/TS 13136:2012) (1) for the detection of the most important STEC serotypes (O157, O111, O26, O103 and O145) for humans, has been published. Shortly afterwards, a European Regulation concerning STEC in sprouts was published (Regulation (EU) 209/2013) (2), and also an EFSA opinion (3) in which a classification of the risks for STEC has been dressed on the basis of virulence genes and serotypes.
Detection of pathogenic E. coli ISO/TS 13136:2012
The detection of pathogenic E. coli using ISO/TS 13136:2012 is based on a PCR screening of the most important virulence genes, stx and eae, and the genes associated with the top 5 serotypes. In the next step, after obtaining a positive PCR result for a sample, an attempt will be made to isolate the germ. This can be realised with or without serogroup-specific accumulation technologies, such as immunomagnetic separation technology. The isolation of STEC strains is required so as to confirm the presence of the virulence genes within a single living bacteria and thus to consider a sample to be positive for STEC.
The method includes the following steps :A. Enrichment of STECB. Screening: DNA extraction and (real-time) PCR detection of virulence genes and serogroup-associated genesC. Isolation and confirmation of the germ from presumptive positive samples for STEC.
54
A. Enrichment
The choice of the enrichment medium depends on the condition of the bacteria and the food matrix to be analyzed. Buffered peptone water (BPW) is recommended for samples in which stressed bacteria (e.g. in deep-fro-zen products) and a low number of natural background flora are expected, so as to resuscitate STEC. In case of samples in which a higher level of natural background flora is expected, an mTSB medium supplemented with novobiocin is recommended, while an mTSB medium supplemented with acriflavin must be used in case of milk and dairy products. Both components inhibit the growth of Gram-positive bacteria, and this in favour of Gram-negative germs, among which STEC. However, concentrations of novobiocin that are too high, inhibit STEC, especially the non O157 E. coli strains. On the other hand, using BPW for these samples could lead to a limited growth of STEC which is due to a competition with the natural background flora which impedes isolation at a later stage. This was confirmed in a scientific study for sprouts and raw vegetables (lettuce, carrots), but has been less observed for raw milk, beef and swabs of bovine carcasses (IDESTEC RT12/12) (4) (Figure 1).It is certainly not easy for a laboratory to select the appropriate enrichment medium, since this strongly depends on the
food matrix to be analyzed.
a Grated carrots (E. coli O145)
Figure 1: (a) Presence of natural background flora (blue colony appearance) on CHROMagar STEC after isolation of STEC
(e.g. E. coli O145) (mauve colony appearance) from grated carrots, but (b) absence of natural background flora after
isolation of STEC (e.g. E. coli O26) from raw milk.
b Raw milk (E. coli O26)
B. Screening: DNA extraction and (real-time) PCR detection of virulence genes and serogroup-associated genes
DNA is directly prepared as from the enriched culture, using an appropriate method for Gram-negative bacteria. The subsequent screening for STEC in the enriched cultures is based on the detection of specific virulence genes stx1 -stx2 and eae using (real-time) PCR. Determination of the serogroup-associated genes provides information on the presumptive seropathotype and can be used to carry out a serogroup-specific isolation of STEC (see point C.).Both conventional PCR as well as real-time PCR can be used to screen enriched cultures. Annex E of ISO/TS 13136:2012 describes primer- and probe sequences, but other primers described in literature or in commercial kits can also be used if the same performance can be proven. This screening step probably doesn’t represent a problem within the method (IDESTEC RT 12/12). (4)
6
The results of the screening for serogroup-associated genes might simplify and orientate the choice of the isolation medium. For samples that are positive in PCR screening for serotype O157, it is recommended to use the reference method for E. coli O157 (ISO16654 or an alternative method) for isolation. For serotype O26 positive samples, a commercial MacConkey agar medium with rhamnose instead of lactose (RMAC) exists. On the other hand, isolation media, as described by Possé and co-workers (2008) (5), which are based on the difference in use of sugars by the top 5 serotypes, can also be applied for the isolation of the germ.
Plating of the enriched culture on an isolation medium can be preceded or not by a serogroup specific accumu-lation by means of for instance an immunomagnetic separation (with magnetic particles coated with O-group specific antibodies). However, this optional phase is not sufficiently specific for some serogroups, as a result of which lots of natural background flora remain, impeding isolation (Figure 3).
C. Isolation and confirmation of the germ from presumptive positive samples for STEC.
So as to confirm the presence of virulence genes within a single living E. coli cell, isolation of the STEC strain will be attempted for stx positive enriched cultures. For this purpose, the enriched culture is spread on a tryp-tone-bile-glucuronide agar medium (TBX), a chromogenic agar specific for isolating E. coli. However, this isolation medium does not allow distinguishing a commensal E. coli strain and STEC, hampering the isolation. As an alter-native, another specific (chromogenic) isolation medium can be used, if available, in order to simplify the isolation of STEC from the background bacteria (e.g. CHROMagarTM STEC, ChromIDTM EHEC) (Figure 2).
Figure 2: E. coli (STEC) isolation on TBX, ChromagarTMSTEC, CT-SMAC, RMAC and agar as described by Possé et al., 2008.
(5)
76
After plating on an isolation medium, up to 50 colonies with E. coli specific properties according to the the isola-tion medium used are examined. By way of point inoculation, these colonies are inoculated on nutrient agar (NA) plates and at the same time transferred in a water solution (pool per 10 colonies) for PCR detection of stx and eae genes. When the (real-time) PCR detection of the stx and/or eae genes is positive in a pool, all individual colonies are tested so as to identify the positive colony. For this purpose, the NA plate can be used to confirm the isolate. Finally, the serotype of the positive colony is determined by (real-time) PCR or agglutination.
Conclusion :
Especially the isolation from and not the (real-time) PCR screening of STEC in an enriched culture is a bottle neck of the ISO/TS 13136:2012 detection method for STEC. When this method is to be implemented in a laboratory, it has to take into account some important factors in order to obtain an optimal detection efficiency of STEC. A validation study, during which the enrichment medium as well as the isolation medium have to be selected, is very important in this process. The choice depends on the nature of the food matrix the lab will receive for analy-sis, as the food matrix and the present natural background flora have a large impact on the growth and isolation possibilities of STEC. Since the ISO/TS 13136:2012 method consists of two large parts, that is the detection of the pathogen by means of (RealTime) PCR and - if positive for the pre-established criteria - the subsequent isolation of the germ, a modular approach of the validation study is required.
Figure 3: Isolation of E. coli O145 from carrots (a) with IMS or (b) by direct plating on TBX
8
References :
1. ISO/TS 13136 :2012. Microbiology of food and animal feed -- Real-time polymerase chain reaction (PCR)-based method for the detection of food-borne pathogens -- Horizontal method for the detection of Shiga toxin-producing Escherichia coli (STEC) and the determination of O157, O111, O26, O103 and O145 serogroups
2. EFSA Panel on Biological Hazards (BIOHAZ); Scientific Opinion on VTEC-seropathotype and scientific criteria regarding pathogenicity assessment. EFSA Journal 2013;11(4):3138. [106 pp.]
doi:10.2903/j.efsa.2013.3138. Available online: www.efsa.europa.eu/efsajournal
3. Commission regulation (EU) No 209/2013 amending Regulation (EC) No 2073/2005 as regards microbiological criteria for sprouts and the sampling rules for poultry carcases and fresh poultry meat.
(2013) Official Journal of the European Union L68/19
4. Intensive identification of human pathogenic Shigatoxin producing Escherchia coli (STEC) in Belgium. (IDESTEC RT 12/12)
5. Possé B, De Zutter L, Heyndrickx M, Herman L. (2008). Novel differential and confirmation plating media for Shiga toxin-producing Escherichia coli serotypes O26, O103, O111, O145 and sorbitol-positive and -negative O157. FEMS Microbiol Lett. 282(1):124-31
Els Van Pamel, Els Daeseleire and Christof Van PouckeInstituut voor Landbouw- en Visserijonderzoek / Institute for Agricultural and Fisheries ResearchEenheid Technologie en Voeding - Voedselveiligheid / Technology and Food Science Unit - Food SafetyBrusselsesteenweg 370, 9090 Melle, Belgium
Introduction of high resolution mass spectrometers (HRMS)
Analytical techniques that can provide qualitative (what?) and/or quantitative (how much?) information on the composition of biological samples are crucial in a number of sectors, such as the food industry. The success of these techniques depends, among other things, on parameters such as sensitivity, selectivity, robustness, accu-racy, precision, speed and cost per analysis. Chromatography- mass spectrometry- based techniques such as gas chromatography-mass spectrometry (GC-MS) and liquid chromatography-mass spectrometry (LC-MS) play an important part in this respect. In what follows, the emphasis will be on the evolution in MS towards high resolu-tion mass spectrometers (HRMS).
These past few years, the world of analysis has seen a shift from low resolution mass spectrometers (such as single and tandem/triple quadrupoles) to high resolution appliances that are characterized by a high level of distinc-tiveness [i.e. high resolution; the ability of the detector to distinguish between two ions that only differ slightly in mass (mass-to-charge-ratio)], mass-accuracy and dynamic range. Magnetic sector, Time-of-Flight (TOF), Fourier transform ion cyclotron resonance (FTICR) and Fourier transform Orbitrap mass spectrometers are different types of commercially available “high resolution” appliances.
For a more detailed description of each of these mass spectrometers, please consult Labinfo N° 12 (July 2014 “Mycotoxin screening” by Bart Huybrechts). The review article by Xian et al. (2012) deals in greater detail with these types of mass spectrometers and their recent developments. In addition to these basic types, hybrid types are also available on the market, like for example quadrupole-TOF (QTOF) and linear trap quadrupole (LTQ) orbitrap configurations. In such hybrid appliances, multiple mass spectrometers are connected to each other. This conside-rably increases the possibilities for analysis, since they provide more diverse mass spectrometric data. These hybrid mass spectrometers, for example, make it possible not only to obtain the accurate mass of the target analyte (the precursor ion or the mother ion) but also to fragment this precursor ion and obtain accurate masses for each of the fragment ions (or daughter ions) (MS/MS or even MSn). The article by Liang et al. (2011) provides an overview of the basic types and hybrid types of these high resolution appliances, in terms of their performance capacity. This overview compares, among other things, the mass accuracy, the resolution, the detection sensitivity, the speed of data recording, the possibility of MS/MS and multi-stage MS (MSn) analysis.
10
Characteristics of HRMS
Even though the name suggests that high resolution or the capacity to make a precise distinction between closely spaced spectral peaks (corresponding to a different mass-charge-ratio) is the main feature of these ap-pliances, the mass accuracy or the ability to determine the components’ mass as precisely as possible, is at least as important. The performance of HRMS appliances is expressed in terms of “resolving power” and is situated somewhere in the range of >10,000 – 100,000 – depending on the type of high resolution appliance – or even higher for the ultra-high resolution appliances. This is considerably higher in comparison to values of 1,000 - 2,000 obtained with quadrupole appliances. Mass accuracy is usually expressed in parts per million (ppm) and depen-ding on the configuration, the mass accuracy is situated in the range of <1 – 5 ppm. A sufficient mass-accuracy contributes to a reduction of the number of possible elementary compositions (structure formulas) for a molecule within a certain mass tolerance, which simplifies the interpretation of data (Holčapek et al. 2010) and allows for instance for the identification of new and/or unknown components.
“Targeted or untargeted”
An important advantage of high resolution mass spectrometers is the possibility to work targeted as well as untargeted. Working targeted implies that screening for the presence/absence of known components in a sample (exhaustive list) is targeted and that the concentration thereof may be determined by means of a calibration curve using its available standard, whether or not in the matrix. Although tandem mass spectrometry (MS/MS) is the most used method for the targeted approach (because of the cost-effective high sensitivity and selectivity), high resolution mass spectrometers also have their merits in this field. However, characteristic of HRMS is rather the “untargeted approach”. Using this approach, a wide mass range view can continuously be used during (part of ) the analysis of a sample, making it possible to obtain a total profile (i.e. “full scan” analysis). Such an approach requires no prior (or a priori) knowledge on the components of interest (e.g. interest in certain classes of subs-tances). The approach thus makes it possible to detect a virtually unlimited amount of different components from a certain class of components (e.g. steroids), but also allows for multiple different components to be analysed simultaneously in a single analytical run (e.g. steroids and coccidiostats).When trying to determine the identity of «unknown» molecules, the mass spectra or peak patterns obtained using the general «untargeted» approach will be analysed by means of software (see figure 1). This will yield a list of all possible elementary compositions or structure formulas. Subsequently, a database search can be started to get a reduced list of possible identifications («hits») or, ideally, even one possible identification. Databases such as PubChem, Chemspider, KEGG... can be used for this purpose. Then it is up to the analyst to retrieve the most likely candidate from the possible component propositions based on extra information. This extra information may be present in fragmentation data (MS/MS or MSn). More precisely, as mentioned earlier, hybrid appliances can also generate MSn data in addition to full scan data. Concretely, this means that in addition to information on mother molecules also mass spectrometric data on the fragment ions (daughter ions) of these precursors (mother ions) can be obtained. This considerably enhances the identification ability or the chances of retrieving the right com-ponent. In addition, chemical properties such as the polarity of the proposed structures can provide an indication on the likelihood of a candidate based on its elution behaviour (i.e. the time needed to run through the chroma-tographic column). The «Seven Golden Rules», as described by Kind and Fiehn (2007), may also be useful here.
1110
Figure 1: Overview of the “untargeted” analysis work flow for the identification of unknown substances
For the «untargeted approach», it is also possible that initially there is no immediate interest in retrieving the identification of the different components, but in comparing the profiles (or peak patterns) of different groups of samples obtained during «full scan analyses». For instance, it is possible that researchers want to study the influence of certain circumstances/conditions or treatments on the composition of their biological material (e.g. changes in certain classes of components) or that they want to verify if the authenticity of food products has been respected (e.g. adulterated olive oil). For such research questions, the profiles generated will be analysed statistically using software. Multivariate classification methods such as «Principal Component Analysis» (PCA) and «Partial Least Squares Analysis” (PLS) or cluster analysis can be used for this purpose. Without going too much into detail on the theoretical aspects of such metabolomic approaches, these techniques can be helpful in identifying those markers/components that contribute to the largest variation between data sets (treatments/conditions). In other words, which components change (e.g. disappear or change in quantity) under which circumstances. These markers can possibly be identified using database searches. One should bear in mind that – even though the high mass-accuracy of HRMS appliances for the «untargeted approach» is unmistakeably present and is indispensable – uniformly identifying «unknown elements» remains a challenging process. And even though fragmentation information obtained during HRMS is very useful, «Nuclear Magnetic Resonance» (NMR) remains indispensable for structure elucidation of completely unknown components.
Samplepreparation
Identification Databasesearch
Data acquisitionFull-scan
Accurate MassSpectra
List Elementalcompositions
MSn fragmentationdata
12
Inject now, analyse later?
An additional aspect adding to the advantages of HRMS is that retrospective data analysis becomes an option owing to the «untargeted/full scan» analysis. This allows for the re-analysis (a posteriori) of previously recorded data by means of software, which makes it possible to verify whether or not (at that point unknown) components were present in the sample at the time of data recording. This basically comes down to the principle of «inject now, analyse later», which may prove to be very interesting for the world of doping and veterinary residues for instance.
«Ion mobility», a third dimension ...
Nowadays, the aspect of separation based on ion mobility («ion mobility separation», IMS) receives an increasing amount of attention in the field of HRMS. The review article by Kanu et al. (2008) provides a summary of the diffe-rent types and advantages of IMS and describes the link of IMS with IMS for TOF, quadrupole, ion trap and FTICR mass spectrometers. The link between IMS and MS makes it possible to separate gas phase ions based on their size-load-ratio and their interactions with a buffer gas on the one hand and on the other hand to separate them in the MS based on their mass-load-ratio. IMS gives rise to an extra dimension translated into drift time, which in its turn can be converted to the «collision cross section» (CCS) of the molecule, basically the 3D dimension based on the size of the molecule. This is also referred to as multidimensional separation which offers the possibility to separate isomers, isobars and conformers and also to reduce chemical noise. IMS-MS is therefore a powerful ana-lytical technique which makes it possible to separate very similar components in complex samples and reduces the risk of false-negative and false-positive results. This technique has applications in different domains such as proteomics, glycomics and metabolomics (Kanu et al. 2008).
«-OMICS»
The wide spectrum of applications of high resolution mass spectrometers, ranging from component identifi-cation and structure elucidation to the analysis of target components and also the wide diversity of biological components which can be analysed together (i.e. the multi-analyte) (medicines, metabolites, organic compo-nents, proteins (and modifications), peptides, lipids, glycoconjugates and other biological components) have contributed to their success in the world of «-OMICS» (proteomics, metabolomics, lipidomics,...). Examples of HRMS applications in the field of food safety can be found in Volume 28 of Food Additives & Contaminants Part A: Chemistry of High Resolution Mass Spectrometry to Food Safety. The evolution towards HRMS in the world of veterinary residue analyses will be briefly discussed in the following paragraph.
1312
HRMS in the world of veterinary residues.
Since a few decades, a European food safety law applies to chemical components which fall under the heading of veterinary medicines. Antimicrobial medicines such as antibiotics and medicines with growth-enhancing proper-ties, such as steroids and β-agonists for example, fall under this category of medicines. Over the years, investments have been made to develop and validate methods for analysis that make it possible to detect residues and/or me-tabolites of these medicines in different types of feed-/ food matrices and biological matrices. In addition to rapid screening methods to make a distinction between suspect and negative (i.e. compliant) samples, the emphasis today is on mass spectrometry as a confirmatory method for an incontestable identification and/or quantification (in other words: which quantity of each (metabolite) residue is present in a sample). For the «targeted» analysis of different veterinary medicine residues in a single analysis (i.e. multi residue methods) low resolution tandem/triple quadrupoles (QqQ) are still the most used mass spectrometers. However, the high resolution and mass accuracy, the «non a priori» approach, the retrospective aspect and the option of MSn have caused a recent trend of increased use of high resolution mass spectrometers, for the «targeted» as well as for the «untargeted» ap-proach for the analysis of often complex samples (insofar as the high price allows these appliances to be used in routine laboratories). HRMS also makes it possible to conduct analyses on substances which previously were not included in the «targeted» methods or to look for new bio markers (Le Bizec et al. 2009). But how does one verify the effectiveness of a method in detecting residues of veterinary medicines or their metabolites? The technical instructions in terms of performance characteristics, limits and criteria which these methods have to meet, are described in Decision 2002/657/EC. This Decision also introduces the concept of identification points (IPs) and ion ratios for mass spectrometric confirmation methods. This guarantees an incontestable identification (and keeps the chance of false results to a minimum). For substances of group A (prohibited substances and substances with anabolic effects) a minimum of 4 IPs is required, whereas for those of group B (veterinary medicines and conta-minants) a minimum score of 3 IPs is required. The section «Performance criteria and other requirements for mass spectrometric detection» describes how many IPs per ion are attributed to methods that use HRMS (whether or not with HRMSn). For HRMS the section also mentions that the resolution normally has to be larger than 10,000 for the entire mass range (at 10% valley).
However, with the recent evolution in technology (and certainly when it comes to HRMS), the appropriateness of the current guidelines in the above-mentioned Decision has to be reviewed. Aspects such as resolution and mass accuracy, the lack of measurable noise, the extra dimension of IMS and other aspects have to be considered in this review. The question arises whether the analysis methods in Decision 2002/657/EC has to be reviewed in the framework of the most recent technological developments. In this framework, the RIKILT Wageningen UR is currently coordinating a project aiming at proposing appropriate performance criteria. A validation study is being carried out which includes different techniques and different laboratories worldwide to collect scientifically based data to check the appropriateness of pre-established performance criteria.
14
A great step forward, but still no light at the end of the tunnel
In conclusion, however much the most recent mass spectrometers available on the market contribute to an im-mense high-quality analysis potential and however much in theory it is possible to apply the «dilute-and-shoot» principle, in practice correct sampling, sample preparation, and sample purification and a sound chromatographic separation remain indispensable in order to be beneficial to the outcome of the analyses. Thus, not only the new developments in the world of HRMS are important, but also those in the field of chromatography (UHPLC, two-di-mensional LC), software for data analysis,... What is more, the advantage of HRMS is also often a disadvantage, more precisely because generating a large amount of data also leads to complex and often time-consuming data analysis. In other words: you can see almost everything, but what exactly do you see? The price tag is also a reason why this technique is not automatically introduced in routine laboratories. There is still much to be said about the use of HRMS in the world of residues. Topics such as method validation and identification criteria still require in-depth discussions and studies.
References:
1 Commission Decision of 12 August 2002 implementing Council Directive 96/23/EC concerning the perfor-mance of analytical methods and the interpretation of results (2002/657/EG)
2 Holčapek M., Jirsko R., Lsa M. (2010) Basic rules for the interpretation of atmospheric pressure ionization mass spectra of small molecules. J. Chrom. A, 1217, 3908-3921
3 Kanu A.B., Dwivedi P., Tam M., Matz L., and Hill H.H. (2008) Ion mobility-mass spectrometry. J. Mass Spectrom., 43, 1-22
4 Kind T., and Fiehn O. (2007) Seven Golden Rules for heuristic filtering of molecular formulas obtained by ac-curate mass spectrometry. BMC Bioinformatics, 8, 105-124
5 Le Bizec B., Pinel G., and Antignac J.-P. (2009) Options for veterinary drug analysis using mass spectrometry. J. Chrom. A, 1216, 8016-8034
6 Liang Y., Wang G., Xie L., and Sheng L. (2011) Recent development in Liquid Chromatography/Mass Spectro-metry and Emerging Technologies for Metabolite Identification. Curr. Drug Metab., 12, 329-344
7 Xian F., Hendrickson C.L., and Marshall A.G. (2012) High Resolution Mass Spectrometry. Anal. Chem., 84, 708-719
The usefulness of whole genome sequencing for outbreak investigation of food pathogens, Salmonella Enteritidis as a case studyVéronique Wuyts1,2,3, Sarah Denayer4, Nancy H.C. Roosens1, Wesley Mattheus5, Sophie Bertrand5, Kathleen Marchal3,6,
Katelijne Dierick4, Sigrid C.J. De Keersmaecker1
1 Platform Biotechnology and Molecular Biology, Scientific Institute of Public Health (WIV-ISP), Brussels, Belgium2 Department of Microbial and Molecular Systems, Centre of Microbial and Plant Genetics, KU Leuven, Leuven,
Belgium3 Department of Plant Biotechnology and Bioinformatics, Ghent University, Ghent, Belgium4 National Reference Laboratory of Food-borne Outbreaks (NRL-FBO) and NRL Salmonella in food, Food-borne
Pathogens, Scientific Institute of Public Health (WIV-ISP), Brussels, Belgium5 National Reference Centre for Salmonella and Shigella (NRCSS), Bacterial Diseases, Scientific Institute of Public
Health (WIV-ISP), Brussels, Belgium6 Department of Information Technology, Ghent University, IMinds, Ghent, Belgium
Classical food-borne outbreak investigations consist of collecting samples from human cases, in many cases via stool, interrogating the human cases to identify a common food source through a standard questionnaire and sampling the suspected food. The collected epidemiological information and food and/or human samples are sent to the National Reference Laboratory of Food-borne Outbreaks (NRL-FBO) for detection, isolation and identification of the food-borne pathogen to define the source of the outbreak and subsequently to be able to control the outbreak as soon as possible. Generally, further characterisation, or fingerprinting, of the pathogen allows to make a strong link between the isolate from the human cases and that from the suspected food or to identify other human cases linked to the consumption of the same food. The latter is important for outbreaks with a dispersed geographical distribution of human cases. This characterisation usually consists of (sub)typing, often through different techniques, and antimicrobial susceptibility testing.
However, the classical characterisation methods do not always show a complete fingerprinting profile of an outbreak strain, as was clearly illustrated by the German Escherichia coli outbreak in 2011. Traditional multi-locus sequence typing (MLST) based on 7 housekeeping genes showed a close relationship between the 2011 out-break strain and an historical German haemolytic uraemic syndrome (HUS) linked E. coli strain. Whole genome sequencing (WGS) however, exposed that the outbreak strain and the historical pathogenic E. coli strain differed in chromosomal gene and plasmid content, thereby revealing a distinct virulence potential [1]. This demonstrated that WGS can provide important information on new emerging pathogens for their treatment or specific detec-tion method requirements.
16
As WGS is postulated as the universal, ultimate resolution (sub)typing method, it is assumed to be a convenient technique for outbreak investigation. For whole genome sequencing of bacterial pathogens, different next gen-eration sequencing (NGS) platforms can be applied to generate the data. Currently, a more pressing issue is the analysis of the WGS data, consisting of millions of short sequencing reads, for which substantial computer power and bioinformaticians are needed. These resources are not typically found in an average NRL/NRC. Similar to tra-ditional characterisation methods, pathogens can be characterised with WGS based on their gene content, e.g. to find antimicrobial resistance or virulence genes. They can then be further fingerprinted to identify the relationship between different isolates. Hereto, two WGS workflows are suggested. One workflow starts with de novo assembly of these sequencing reads, after which a sequence type is assigned through an allele-based analysis, also called whole genome MLST. For this kind of analysis, typing schemes and international databases to which the alleles of an outbreak isolate can be compared need to be established for each pathogenic species. As whole genome MLST schemes and databases are, thus far, not available for most pathogens, another workflow is currently most used. In this workflow single nucleotide polymorphisms (SNPs) are identified after mapping of the reads against a reference genome. For any of the two workflows, well-trained bioinformaticians are needed for the development of user-friendly WGS analysis pipelines.At the moment, retrospective WGS investigations are being conducted to estimate the genetic diversity within defined outbreaks and between outbreak isolates and circulating background strains. This may assist in data interpretation of future outbreaks. In this context, in April and May 2014, the NRL-FBO (Scientific Institute of Public Health (WIV-ISP)) investigated two geographically separated outbreaks of Salmonella, one in Flanders and one in Wallonia. As for both outbreaks both food and human isolates were available, these outbreaks were taken as a case study for retrospective outbreak investigation by WGS, in collaboration with the Platform Biotechnology and Molecular Biology (WIV-ISP).
The classical characterisations methods for Salmonella, namely serotyping, phage typing and multiple-locus varia-ble-number of tandem repeats analysis (MLVA), revealed that the outbreaks were caused by Salmonella Enteritidis with the same phage type and the same MLVA profile, with the exception of one human isolate that showed a variant phage type. Based on these results and the epidemiological investigations, the Flemish outbreak could be linked to chocolate mousse prepared with raw, non-commercial eggs and also the Walloon outbreak could be linked to raw, non-commercial eggs, which were consumed via tiramisu. However, the outbreaks could only be distinguished based on their separate locations. WGS was performed by NGS on six food and human isolates. The WGS data analysis was performed with software and tools that run under Windows 7, in view of the applicability and feasibility of WGS within a non-bioinformatics expert environment. The WGS reads were uploaded to the CSI Phylogeny server [2] of the Center for Genomic Epidemiology for SNP analysis. This resulted in about 52 SNPs difference between the two outbreaks and 0 to 2 SNPs difference within each outbreak. A phylogenetic tree is shown in Figure 1, clearly making a distinction be-tween the isolates of each outbreak, while showing a clear link between the food and human isolate within each outbreak. Additional typing was done by uploading de novo assembled contigs (constructed with easy-to-use, commercial NGS data analysis software, i.e. CLC Genomics Workbench) to the MLST server [3], ResFinder [4] and PlasmidFinder [5]. MLST server, based on the classical MLST scheme of 7 housekeeping genes, assigned the same sequence type to all 6 isolates. With ResFinder no resistance genes could be discovered, except for a bla
TEM gene
in the human isolate with the variant phage type. This isolate also presented a second plasmid through Plasmid-Finder, whereas for all other isolates only one plasmid could be found. Antimicrobial susceptibility testing using minimal inhibitory concentrations phenotypically confirmed the ResFinder results. Further, visual analysis with BRIG [6] revealed that the bla
TEM gene was located on the additional plasmid.
1716
Figure 1. Phylogenetic tree of the outbreak isolates created with FigTree [7]. The reference is S. Enteritidis P125109 (black).
Isolates of the Flemish outbreak are shown in blue, those of the Walloon outbreak in purple.
This case study shows that more detailed fi ngerprinting of outbreak isolates is possible with WGS, as the S. Enteritidis isolates of each outbreak could be distinguished based on SNP analysis, which was not possible with classical phage typing and MLVA. It also shows that WGS analyses based on gene content can further fi ne-tune the characterisation of outbreak isolates. Although the NGS technology might in some cases still be expensive, with the further decrease of the NGS prices, current separate typing assays, each signifi cantly adding to the total cost of an outbreak investigation, might in the future be replaced by this single, universally applicable assay. Further, it highlighted that WGS analysis is not strictly reserved for bioinformaticians. Therefore, this study clearly demonstrates the potential of the use of WGS in the NRL-FBO for future outbreak investigations. These outbreak investigations will even more benefi t from a future routine WGS analysis of each isolate sent to the NRL-FBO or specifi c NRC to better discriminate outbreak strains from circulating background isolates. Detailed results of this study have been submitted for peer-review publication.
Flan
dre_
nour
ritur
e-1
reference
Wallonie_nourriture
Flandre_humain-1
Wall
onie_
humain
Wallonie_nourriture
Wall
onie_
humain
Wallonie_nourriture
Flandre_humain-2
Flandre_nourriture-2
Flan
dre_
nour
ritur
e-1
Flandre_nourriture-2
Flan
dre_
nour
ritur
e-1
Flandre_humain-1
Flandre_nourriture-2
Flandre_humain-1
18
Acknowledgements
The authors wish to thank the Federal Agency for the Safety of the Food Chain (FASFC) and the Flemish and French speaking Communities for the excellent collaboration. This study was supported by grant P4044.0103 (Sal-MolType) from the Scientific Institute of Public Health (WIV-ISP - RP/PJ). The NRL-FBO is co-funded by the Federal Agency for the Safety of the Food Chain and the Federal Public Service Health, Food Chain Safety and Environ-ment. The NRCSS is partially supported by the Belgian Ministry of Social Affairs through a fund within the Health Insurance System.
References
1. Mellmann A, Harmsen D, Cummings CA, Zentz EB, Leopold SR, Rico A, Prior K, Szczepanowski R, Ji Y, Zhang W, McLaughlin SF, Henkhaus JK, Leopold B, Bielaszewska M, Prager R, Brzoska PM, Moore RL, Guenther S, Rothberg JM, Karch H (2011) Prospective genomic characterization of the German enterohemorrhagic Escherichia coli O104:H4 outbreak by rapid next generation sequencing technology. PLoS ONE 6: e22751.
2. Kaas RS, Leekitcharoenphon P, Aarestrup FM, Lund O (2014) Solving the problem of comparing whole bacterial genomes across different sequencing platforms. PLoS ONE 9: e104984.
4. Zankari E, Hasman H, Cosentino S, Vestergaard M, Rasmussen S, Lund O, Aarestrup FM, Larsen MV (2012) Identification of acquired antimicrobial resistance genes. J Antimicrob Chemother 67: 2640-2644.
5. Carattoli A, Zankari E, García-Fernández A, Larsen MV, Lund O, Villa L, Aarestrup FM, Hasman H (2014) In silico detection and typing of plasmids using PlasmidFinder and plasmid multilocus sequence typing. Antimicrob Agents Chemother 58: 3895-3903.
6. Alikhan N-F, Petty NK, Ben Zakour NL, Beatson SA (2011) BLAST Ring Image Generator (BRIG): simple prokaryote genome comparisons. BMC Genomics 12: 402.
7. Rambaut, Andrew (2014) FigTree. Available: http://tree.bio.ed.ac.uk/software/figtree/. Accessed: 15-4-2015.
Halal food: analysis methods to detect traces of pork in meat and meat productsRob Margry
President AOAC Lowlands Section
Integrity, authenticity and «fair trade practices» are becoming increasingly important concepts in the production and marketing of food. The recent horse meat scandal is a good example of this. In this respect, the question arises as to whether the actual composition of the product corresponds to the ingredients listed on the label. This question does not concern food safety, but is inspired by the consumer’s demands, which are mostly based on traditional, religious or emotional grounds. For instance, the demand for fresh halal meat and halal meat products free from traces of pork has been on the rise in recent years. Europe currently has a Muslim population of 16 mil-lion, 72 million Muslims live in Turkey and there are about 1.5 billion Muslims worldwide. Consumers demand transparency. Like fair trade and biological products, halal products are in need of a quality mark. The existing halal quality marks are however not uniform.In the Qur’an, many more animal species and animal products are considered haram (not allowed for consump-tion), but so far the halal practice has remained limited to requests for analysis aimed at detecting pork.
In practice, a zero tolerance policy is applied to pork. Needless to say, when diff erent analytical methods are used, each with their own particular performance characteristics, the results will often contradict each other. These methods are based on diff erent principles (for example based on the detection of DNA or proteins). Consequent-ly, there will be diff erences in detection limits and in the percentage of false negative and false positive results. More precisely, in the case of very low contamination levels of meat with pork, an analysis repeated by the same laboratory could also yield an opposite result. Manufacturers who want to have their products checked, often mis-takenly believe that laboratories that have the ISO 17025 accreditation will all produce the same analytical results.
20 21
There is thus a clear need for a reference method against which the other methods can be assessed. A year and a half ago, the European Committee for Standardization (CEN) has founded a working group (CEN/TC 425) that will develop the standard for a production management system for Halal Food. The production process described in this system will be based on HACCP principles. In addition to traceability, conformity assessments will also be incorporated, including verification by means of analyses. In this respect, not only the final product, but also the contamination of the production line will be checked, for example by means of swabs.
In 2006, a European reference laboratory was founded in order to develop methods for the detection and species identification of animal material in feed: the EURL-AP (CRA-W, Gembloux). Since the consumption of farm animal meat approved for human consumption is not considered a health risk, there is no official EURL for animal species specifications in food. Moreover, in 2013, the DG Health instructed the EURL-AP to issue a recommendation for the detection of horse meat in meat and meat products.
Common analytical methods
K. Nakyinsige et al (2012) provide an overview of analytical methods that could be used to detect the presence of pork tissue in meat (products). In addition to PCR (DNA detection), immuno-chemical and chromatographic methods (protein detection), LC-MS/MS biosensor methods, FTIR (fat analysis) and Differential Scanning Calorim-etry (DSC) are also mentioned. In practice, mainly PCR and immuno-chemistry are applied routinely. The other techniques are still being developed and are not expected to be validated for halal monitoring in the years to come.
In the meantime, many analytical methods for species identification are available on the market. According to the suppliers, some analysis kits can only be used for fresh meat. Other kits could also be used for the analysis of processed (heated) meat products. Considering the zero tolerance policy, it is striking that different suppliers apply different detection limits to their products. These detection limits seem to vary between 5% and 0.0001%. The question here is which type of reference material has been used to determine this limit. By for example spiking fresh beef with pork, a calibration curve can be set up, which yields a (semi-) quantitative result based on weight. After processing (more precisely heating) the meat, it is however no longer clear to what extent the DNA or proteins are still intact. Both molecules can be denatured to such an extent that an unknown fraction can no longer be detected. After processing, the methods applied can only be used for qualitative purposes: it can only be reported that pork is «detectable» or «not detectable». Should this be the case, the choice can be made to base the results on fresh meat, but this will produce an underestimation of the actual contents.
PCR methods
Nowadays, real-time PCR methods are preferably used in halal checks for the routine examination of fresh meat or processed meat products. Sometimes, Restriction Fragment length polymorphism PCR with electrophoretic sep-aration of the multiplied DNA fragments is still used. However, this approach is laborious and more susceptible to false positive results. The different real-time PCR methods usually use different targets (DNA fragments), for which a target-specific matching set of primers and a probe are used. This results into different performance character-istics, more precisely in terms of the percentage of false positive and false negative results, but also in terms of different detection limits. Of course, the method used to prepare (grinding, homogenizing and extracting) the
20 21
sample also plays an important part in this respect. The detection limit of a real-time PCR method not only de-pends on the type of equipment used, but also on the individual appliance. EURL-AP (cfr. Planchon et al, 2010) has developed a protocol to determine a decision limit to make a distinction between «detectable» and «not detecta-ble» using plasmid calibrants. This way, the differences between results will become significantly smaller.
In their overview, K. Nakyinsige et al (2012) mention, among other things, the real-time PCR method developed by Fumière et al (2006). The advantage of this method is that a short (68 base pairs long) amplicon is used. Ampli-cons shorter than 100 base pairs are less sensitive to (heat) denaturation of DNA. A new improved version of this method was recently developed, which is currently being tested by means of an inter-laboratory validation study. This method could also be used for the analysis of meat. NutriControl has conducted preliminary experiments on meat which confirm this hypothesis.
Immunochemical methods
Enzyme Linked Immuno Sorbant Assays (ELISA) as well as lateral flow assays (dip sticks) are used These methods are generally less sensitive than real-time PCR methods and are able to detect pork proteins. A target protein that is often used is the heat-resistant skeleton muscle protein Troponine I. A disadvantage of protein detection is that the target protein is not present in all types of tissue. DNA is, however, present in every cell that contains a core.NutriControl is currently comparing the practical usability of a number of ELISA methods.
Standardization of analytical methods
Since an agreement on a tolerance limit for pork components is not to be expected any time soon, there is a great need for a standardized analysis method against which laboratories can check their own method. In addition, it is also important that laboratories participate in proficiency tests. It is possible that multiple adequate methods exist, but the new method developed by EURL-AP could definitely be a suitable candidate.
Literature - Nakyinsige K., Che Man Y.B. and Sazili A.Q. (2012): Halal authenticity issues in meat and meat products; Meat
Science 91, 207 – 214. - Fumière O., Dubois M., Baeten V., von Holst C. and Berben G.(2006): Effective PCR detection of animal species
in highly processed animal byproducts and compound feeds; Anal. Bioanal. Chem. 385, 1045 – 1054.- Planchon V., Oger R., Marien A., Berben G. and Fumière O.; AGROSTAT, February 23 -26 2010, Benevento, Italy.
Analysis of food allergens by LC-MS/MSGillard Nathalie, Otto Gaëtan and Delahaut Philippe
CER Groupe - Département Santé - rue du point du jour, 8 - 6900 Marloie
Introduction
Food allergies constitute a growing problem which allergic consumers, food producers and regulatory agen-cies have to cope with. More and more allergic consumers are diagnosed (2 to 3 % among adults and up to 8% among children) and only the total elimination of allergens in the diet helps to avoid an allergic reaction. The non identified and non declared presence of allergens in foodstuffs can sometimes lead to anaphylaxis-type reactions. The European regulation 1169/2011 makes it compulsory to label 14 allergenic ingredients (cereals with gluten, crustacean, eggs, fish, groundnut, soya, milk, nuts, celery, mustard, sesame, sulphites, molluscs, lupin and all derivatives of these products) for pre-packaged products.For most of the food industries and especially for the small ones, it is economically impossible to dedicate pro-duction lines exclusively to one type of products and cross-contaminations are by consequence possible. This incidental presence of allergens is not covered by the legislation and various companies currently prefer precau-tionary labelling to the setting up of a risk management programme of cross contamination in their production chain.
In order to evaluate the possible presence of allergens in their products and label consequently, agro-food indus-tries need trustworthy and specific analytical tools.
Currently, the detection of allergens in foodstuffs is mainly carried out by ELISA and PCR methods (or RealTime PCR). However, these two types of methods pose limitations at the moment. Indeed, the PCR analysis only detects the presence of specific DNA of allergy-causing ingredients. The immunological methods on the other hand, represent a high variability on quantification, due to the presence of matrix interferences and the characteristics of the different antibodies of which the commercial ELISA kits are made up; the results of these analyses are highly influenced by the transformation of foodstuffs and false negative results are generated in this way.
Liquid chromatography coupled to the mass spectrometry (LC-MS/MS) turns out to be a preferred technique to identify, confirm and routinely dose allergens. Indeed, this technique is very specific (it allows to identify and confirm the presence of a compound based on its retention time and profile of fragmentation in mass spectrom-etry), robust and sensitive. It represents also the advantage to dose several allergens during one single analysis, as opposed to ELISA Tests, which reduces the costs of analyses and is economically interesting. Moreover, its specific-ity and robustness can reduce the matrix interferences.
2322
Principle of analyses of allergens by mass spectrometry
The basic principle of the analysis of allergens by mass spectrometry is described in the fi gure 1.
Figure 1: Principle of analysis of food allergens by liquid chromatography coupled with mass spectrometry in tandem.
The used approach to analyse allergens represents certain peculiarities, compared to the one applied to analyse residues and contaminants of non protein-rich type (drug residues, toxins,...). These particularities are linked to the size of allergens and also to their structure. These strategies are detailed below.
1. Pre-instrumental stage: digestion of allergens
After extraction/purifi cation, the extract undergoes an enzymatic digestion before its analysis with MS. Indeed, the allergens are proteins of which the molecular mass goes from a few dozen kDa (lysozyme, casein) to more than 150 kDa (ovomucin). Even if the analysis of these whole proteins is possible using mass spectrometers type ‘’Time of Flight’’ (ToF), these proteins are too large to be analysed by spectrometers of type triple quadrupole, of which the range of mass exceeds rarely 2000 Da. The enzymatic digestion allows to generate peptides of which the size will be compatible with their analysis of LC-MS. Among the commercially available enzymes, trypsin represents an enzyme of choice because it is very specifi c - it cleaves after the amino-acids lysine and arginine -, generates peptides of between 7 and 20 amino-acids - which are compatible with the analysis of LC-MS - and is relatively cheap compared to other endoproteases. For certain proteins, such as gliadins, trypsin won’t be an enzyme of choice because its primary sequence counts too few lysines and arginines (see fi gure 2); in this case, another enzyme, a combination of enzymes (trypsin/ chemotrypsin) or a chemical digestion are preferred.
24
Figure 2: Sequence of wheat gliadin. The sites of cleavage of trypsin are represented in red.
2. Development of instrumental methods for dosage of allergens by LC-MS
After digestion, the samples are analysed by LC-MS. A major challenge of analysis of food allergens by mass spec-trometry is to identify proteins and/or peptides which are specific of the aimed allergen, robust to the transforma-tion of foodstuffs and allow reaching a good sensitivity.
For an allergenic ingredient, several proteins can be chosen. The selection criteria are the availability of its se-quence in the databases, its abundance in the allergen and its digestibility by the chosen enzyme. Table 1 illustrates proteins specific to groundnuts. In the case of groundnuts, the proteins Ara h1 and Ara h2 represent 20% and 10% of the total protein content, respectively, and seem to be good target proteins for MS analysis by consequence. It is also important to underline that a target protein in mass spectrometry must not necessarily be allergenic, the aim is to prove the presence of an allergenic ingredient and not to prove the allergenicity of the tested foodstuff.
Table 1: Proteins present in groundnut
Protein Family Molecular mass (kDA)
Ara h1 vicillines 64.5
Ara h2 conglutines 17.5
Ara h3 glycinines 14
Ara h4 glycinines 35.9
Ara h5 profilines 14
Ara h6 conglutines 14.5
Ara h7 conglutines 15.8
For each selected protein, several peptides are generated after an enzymatic digestion. Figure 3 shows the exam-ple of proteins Ara h1, Ara h2 and Ara h3-4 of groundnuts; trypsin cleaves after the residues arginine and lysine, which allows to identify several peptides which will be generated after a tryptic digestion.
geen uitgelezen enzym, aangezien hun primaire sequentie te weinig lysine en arginine bevat (zie Figuur 2). In dat geval verkiest men voor de digestie een ander enzym, een combinatie van enzymen (trypsine/chemotrypsine) of een chemische digestie.
Figuur 2. Sequentie van tarwegliadine. De plaatsen waar de proteïne door trypsine geknipt wordt staan in het rood.
2. Ontwikkeling van een instrumentele methode voor dosering van allergenen door LC-MS/MS Na enzymatische vertering worden de monsters met behulp van LC-MS/MS geanalyseerd. Een grote uitdaging in de analyse van voedingsallergenen door middel van massaspectrometrie is het identificeren van proteïnen en/of peptiden die specifiek zijn voor het doelallergeen, die robuust zijn, met andere woorden ook te detecteren in verwerkte levensmiddelen, en die toelaten een goede gevoeligheid te bekomen. Verschillende proteïnen kunnen geselecteerd worden voor een allergeen ingrediënt. Selectiecriteria zijn de beschikbaarheid van zijn sequentie in de databanken, de hoeveelheid van het proteïne in het allergene ingrediënt en de verteerbaarheid van het gekozen enzym. In Tabel 1 worden de specifieke proteïnen van pinda's weergegeven. In het geval van pinda's vertegenwoordigen de proteïnen Ara h1 en Ara h2 respectievelijk 20% en 10% van de totale proteïnen en lijken dus goede doelproteïnen voor de MS analyse te zijn. Het is eveneens belangrijk te benadrukken dat een doelproteïne voor massaspectrometrie niet noodzakelijk allergeenopwekkend hoeft te zijn, het is hier de bedoeling de aanwezigheid van een allergeen ingrediënt te bewijzen en niet om de allergeniciteit van het geteste levensmiddel aan te tonen.
Tabel 1. In pinda's aanwezige proteïnen
Proteïne Familie Moleculaire massa (kDa)
Ara h1 vicillinen 64.5 Ara h2 conglutinen 17.5 Ara h3 glycininen 14 Ara h4 glycininen 35.9 Ara h5 profilinen 14 Ara h6 conglutinen 14.5 Ara h7 conglutinen 15.8
Voor elke geselecteerde proteïne zullen tal van peptiden na enzymatische vertering gegenereerd worden. In figuur 3 is het voorbeeld van de proteïnen Ara h1, Ara h2 en Ara h3-4 van pinda's terug te vinden; trypsine knipt na de residuen arginine en lysine, wat het mogelijk maakt de vele peptiden te onderscheiden die na vertering door trypsine gegenereerd worden.
Figuur 3. Aminozuursequenties van de proteïnen Ara h1, Ara h2 en Arah 3-4 van pinda's. De plaatsen waar de proteïne door trypsine geknipt wordt staan in het rood (lysine K en arginine R). De peptiden die voor de LC-MS/MS methode geselecteerd zijn, zijn in blauw onderstreept.
Er moet opnieuw een selectie worden uitgevoerd, gebaseerd op de lengte van de peptiden (de analyse met chromatografie zal optimaal zijn voor peptiden met 7 tot 20 aminozuren; de specificiteit zelf van een peptide kan betrouwbaar geverifieerd worden voor peptiden met minstens 8 aminozuren), op hun specificiteit voor het doelallergeen (de specificiteit wordt geverifieerd door de sequentie van het peptide in een UNIPROT databank te zoeken, waarbij wordt nagegaan of deze peptidesequentie niet in andere proteïnen kan worden gedetecteerd) en op hun robuustheid bij de verwerking van de levensmiddelen. Wat de robuustheid bij de verwerking van de levensmiddelen betreft, worden bepaalde proteïnen gevormd en/of bepaalde aminozuren van die proteïnen gewijzigd (glycosylering, Maillard reactie,...). Dit is geen probleem als de doelpeptiden in MS/MS zelf nog steeds intact zijn na verwerking en het extractie/zuiveringsprotocol het mogelijk maakt deze te recupereren. De eerste twee selectiecriteria van peptiden kunnen gemakkelijk worden toegepast bij de ontwikkeling van de LC-MS/MS methode. Wanneer geen informatie beschikbaar is over de robuustheid van peptiden bij de verwerking van levensmiddelen, wordt aanbevolen om een maximum aan peptiden in de ontwikkelde methode in te brengen en later een selectie uit te voeren.
Figure 3: Sequences of amino-acids of proteins Ara h1, Ara h2 and Arah 3-4 of peanuts. The sites of cleavage of trypsine are
represented in red (lysine K and arginine R). The selected peptides for the LC-MS method are underlined in blue.
Again, a selection must be made, based upon the length of peptides (chromatographic analysis will be optimal for peptides having between 7 and 20 amino acids; the specificity of a peptide can be verified for peptides of at least 8 amino-acids), their specificity for the target allergen (the specificity must be verified by researching the sequence of the peptides in databases of the type UNIPROT and by verifying that this peptide sequence cannot be detected in other proteins) and their robustness to transformation of foodstuffs. With regard to the robustness during the transformation of foodstuffs, the confirmation and/or some of the amino-acids of certain proteins can be modified (glycosylation, Maillard reaction,...); this is no problem if the target peptides for MS are still intact after transformation and the protocol of extraction/purification still allows to retrieve them.
26
The fi rst two selection criteria of peptides can easily be applied during the development of the LC-MS method. If there is no other information available regarding the robustness of peptides during transformation of foodstuff s, it is advised to introduce a maximum of peptides in the developed method and to do a subsequent selection.
Figure 4 shows the obtained chromatogram while analysing a sample of groundnut. The retained peptides in this method are those that match the criteria of length and specifi city and present a good sensitivity during the analysis of a sample of groundnut.
Figure 4: Chromatogram which represents the sought peptides for the allergen ‘’groundnut’’. Peptides coming from the
proteins rah1 (blue), Arah2 (green) and Arah3-4 (red) are represented by the following code: (1) NNPFYFPSR, (2) DQS-
The analysis of peptides in routine will be carried out in MRM mode; this analysis represents meanwhile some particularities in comparison with the analysis of drug residues. Indeed, the parent peptides are often present under several stages of charge (1+, 2+ and 3+), just as the daughter ions. With regard to the type of daughter ions that are observed during the MS-analysis of peptides, the nomen-clature of Biemann represented in Figure 5 allows to understand the fragmentation of peptides and the type of produced ions. While for each peptide, a multitude of peptide fragments can theoretically be detected, it appears that in practice, the ions a, b and y are most frequently observed in low energetic collisions.
2726
Figure 5: Fragmentation of peptides according to the nomenclature of Biemann.
For higher specifi city of the MS method, selection criteria can also be applied to MRM transitions, for example to target transitions for which the relation m/z of the parent ion is lower than for the daughter ion (this is for instance possible if you target a 2+ peptide as a parent ion and a 1+ fragment as daughter ion) and to avoid daughter ions of small size (a fragment of 2-3 amino-acids has more chance to come from another peptide than a fragment of 6-7 amino-acids).
After having applied all the previously described selections, it is fi nally possible to obtain an LC-MS method that is capable to detect and identify several allergens; fi gure 6 shows a chromatogram obtained after the analysis of a bread sample with an LC-MS method allowing detection of 7 allergens (Heick et al., 2011).
Figure 6: Followed MRM transitions to detect 7 allergens (walnut, hazelnut, soy, milk, groundnuts, almonds, eggs) present
at 1000 ppm in a sample of bread (Heick et al., 2011).
28
References
- Heick J., Fischer M., Popping B. (2011) First screening method for the simultaneous detection o 1 f seven al-lergens by liquid chromatography mass spectrometry. Journal of Chromatography A. 1218, p. 938–943
- Biemann K. (1988) Contributions of mass spectrometry to peptide and protein structure. Biomed Environne-mental Mass Spectrometry. (1-12), p.99-111
Emerging and Novel Brominated Flame Retardants (BFRs) in food: Current status of the European legislation
Gauthier Eppe, Georges Scholl, Edwin de Pauw and Jean-François Focant
CART University of Liège, Allée de la Chimie 3, B-6c Sart-Tilman, B-4000 Liège, Belgium
Established BFRs
Brominated fl ame retardants (BFRs) are anthropogenic chemicals that are used with the aim of increasing the fi re resistance of materials. These brominated organic compounds started to be industrially produced in the begin-ning of the 1970s and their cumulative current production volume exceeds 400,000 tons/year [1]. The main usage concerns the electronic industry, principally in the electronic goods in printed circuit boards, connectors and cables or components such as plastic covers (e.g. television, computers) but they are also used in carpets, uphols-tery, furnishing and paints.
1
Opkomende en nieuwe gebromeerde vlamvertragers (BFR's) in voedsel: Huidige status van de Europese wetgeving
Gauthier Eppe, Georges Scholl, Edwin de Pauw en Jean-François Focant CART Universiteit Luik, Allée de la Chimie 3, B-6c Sart-Tilman, B-4000 Luik, België
30
BFRs are incorporated as either additive (blended with polymer), such as polybrominated diphenyl ethers (PBDEs), polybrominated biphenyls (PBBs) and hexabromocyclododecanes (HBCDDs) or reactive ingredients (covalently bonded to polymer), such as tetrabromobisphenol-A (TBBPA). PBBs, PBDEs, HBCDDs, TBBPA and their derivatives are the most commonly used. These BFRs can leach out or evaporate from the products in which they are incor-porated. They have been found to be ubiquitously present in remote areas in either abiotic and biota samples providing evidence that these substances are persistent in the environment, including long-range environmental transport, bioaccumulation in aquatic and terrestrial food and human biota. In 2009, HexaBBs, BDE congeners 47, 99, 153, 154, 175 and 183 were classified as new Persistent Organic Pollutants (POPs) under the Stockholm Convention; in 2013, HBCDDs were also added to the new POPs list. This has led to ban the production and use of certain formulations of these BFRs.
In order to assess the need for regulatory measures (or not), the European Commission asked the European Food Safety Authority (EFSA) to prepare a scientific opinion on the risks to human health related to the presence of BFRs in food. The Scientific panel on contaminants in food adopted several scientific opinions on the different classes of BFRs between 2010 and 2012 [2,3,4,5,6]. EFSA concluded that the most important BFRs (BDE congeners 28, 47, 100, 153, 154, 183, and 209; BB congener 153; HBCDD α,β,γ isomers; TBBP-A) have to be monitored based on the analytical feasibility to measure their occurrence in food and feed in accredited laboratories. The European Commission has adopted a recommendation (2014/118/EU) indicating that member states should perform the monitoring of BFRs during the year 2014 and 2015 for a wide variety of foodstuffs reflecting consumption habits [7]. Analytical methods have to reach a limit of quantification (LOQ) of 0.01 ng/g wet weight or lower for PBDEs and HBCDDs while 0.1 ng/g wet weight or lower is accepted as LOQ for TBBP-A and its derivatives.
Emerging and novel BFRs
Beside these ‘established BFRs’ a series of less well-known and studied BFRs were classified as ‘emerging’ and ‘novel’ BFRs [8]. According to the EFSA report on these new classes of BFRs and also based on the scientific publication of Bergman and co-workers [9], emerging BFRs are defined as chemicals which are applied as flame retardants that have been identified as anthropogenic chemicals in any environmental compartment, in wildlife, in food or in humans. Novel BFRs are defined as chemicals applied as flame retardants, and with confirmed presence in materials and/or goods in concentrations above 0.1% but not identified in environmental samples, wildlife, food or humans. These two groups of BFRs encompass 17 and 10 individual compounds, respectively. The complete list is available in the EFSA report [8]. It is rather difficult to estimate accurately the production of these new BFRs. The report from Harju and coworkers estimates the total volume of production around 180000 tons/year [10]. Regarding the analytical methodologies, the EFSA report pointed out the lack of specific analytical methods for many of them. However, amongst the list, Commission Recommendation 2014/118/EU asked to carry out analysis of tris(2,3-dibromopropyl) phosphate (TDBPP); N,N’-ethylenebis(tetrabromophthalimide) (EBTEBPI); hexabromocy-clodecane (HBCYD); bis(2-ethylhexyl) tetrabromophthalate (BEH-TEBP); 2- ethylhexyl 2,3,4,5-tetrabromobenzoate (EH-TBB) and dibromoneopentyl glycol (DBNPG) in fish and other seafood, meat and meat products, animal and vegetable fats and oils, milk and dairy products, eggs and egg products and food for infants and small children [7. A limit of quantification of 1 ng/g wet weight or lower is requested for those BFRs. One should note that the analytical challenges to develop accurate methods here is much more complicated compared to the methods developed a few years ago for established BFRs like PBDEs. A wide variety of analytical approaches for sample extraction, purification and instrumental analysis are needed [11]. In addition, a limited number of standards are available for emerging and novel BFRs, including some 13C labeled standards for isotope dilution quantification by LC or GC-mass spectrometry (MS) based techniques, but not sufficiently. A large range of standards and reference materials need to be developed.
3130
Scientific reports, opinions and publications revealed and identified a number of research gaps covering analytical aspects, environmental issues, levels in food, physico-chemical characteristics, toxicological hazards and human exposure to these emerging and novel BFRs. Studies and research projects need to be undertaken to gather ad-ditional experimental data. In that context, a recent publication reported possible concerns regarding the present of Dechloranes (Dechloranes Plus, Dechloranes 602, Dechloranes 603, Dechloranes 604, and Chlordane Plus) in human serum samples from Western Europe [12]. Despite the fact that these chlorinated and mix chloro-bromo FRs were never produced in Europe, they were measured at levels higher than the most common PBDEs. Dechlo-ranes were further found in Belgian foodstuffs at the pg/g fat level, corresponding to an estimated daily intake of more than 100 pg [13]. As very little is known about their toxicity, these reports do not yet demonstrate the need for regular food-feed control, but at least highlight Dechloranes as amongst the possible next targets.
References: (1) Eljarrat E, Barcelo, Brominated Flame Retardants, New York : Springer 2011(2) Scientific Opinion on Polybrominated Biphenyls (PBBs) in Food. EFSA Journal 2010; 8(10):1789. [151 pp.].
doi:10.2903/ j.efsa.2010.1789. (3) Scientific Opinion on Polybrominated Diphenyl Ethers (PBDEs) in Food. EFSA Journal 2011; 9(5):2156. [274
pp.] doi:10.2903/ j.efsa.2011.2156. (4) Scientific Opinion on Hexabromocyclododecanes (HBCDDs) in Food. EFSA Journal 2011; 9(7):2296. [118 pp.]
doi:10.2903/ j.efsa.2011.2296. (5) Scientific Opinion on Tetrabromobisphenol A (TBBPA) and its derivatives in food. EFSA Journal 2011;
9(12):2477. [61 pp.] doi:10.2903/j.efsa.2011.2477. (6) Scientific Opinion on Brominated Flame Retardants (BFRs) in Food: Brominated Phenols and their Deriva-
tives. EFSA Journal 2012; 10(4):2634. [42 pp.] doi:10.2903/j.efsa.2012.2634. (7) Commission Recommendation 2014/118/EU of 3 March 2014, Official Journal of the European Union(8) Scientific Opinion on Emerging and Novel Brominated Flame Retardants (BFRs) in Food. EFSA Journal 2012;
10(10):2908. [125pp.] doi:10.2903/j.efsa.2012.2908.(9) Bergman A, RydenA, Law R. J, de Boer J., Covaci A., Alaee M., Birnbaum L., Petreas M., Rose M., Sakai S., Van
den Eede N., van der Veen I., Environment International 49 (2012) 57-82(10) Harju M, Heimstad ES, Herzke D, Sandanger T, Posner S, Wania F. Report 2462. Oslo, Norway: Norwegian
Pollution Control Authority; 2009. 113.(11) Covaci A., Harrad S., Abdallah M.A., Ali N., Law R.J., Herzke D., de Wit C. A., Environment International, 37
(2011), 532-556.(12) Brasseur C., Pirard C., Scholl G., De Pauw E., Viel J-F., Shen L., Reiner E.J., Focant J-F., Environment Internatio-
UGM characterization using DNA walking strategyFraiture Marie-Alice1,2,3, Herman Philippe1, Papazova Nina1 and Roosens Nancy1
1 Scientific Institute of Public Health (WIV-ISP), Platform of Biotechnology and Molecular Biology (PBB) and Biosafety and
Biotechnology Unit (SBB), J. Wytsmanstraat 14, 1050 Brussels, Belgium2 Instituut voor Landbouw- en Visserijonderzoek (ILVO), Eenheid Technologie & Voeding (T&V), Burg. Van Gansberghelaan
115, 9820 Merelbeke, Belgium 3 University of Gent, Faculty of Pharmaceutical Sciences, Laboratory of Pharmaceutical Biotechnology, Ottergemsesteen-
weg 460, 9000 Ghent, Belgium
To increase the productivity in agriculture, strategies of biotechnology allowing to produce genetically modi-fied (GM) crops have been developed. As rice is one of the most predominant staple food, in the context of the UGMONITOR project (convention RF 11/6242), more than 1000 peer-reviewed publications were collected, covering a period from 1991 to 2015, to provide an overview of the transgenic rice lines described at the R&D step (IRRI; Scopus). The majority of these studies were done in Asia (77.8%), especially in China (47.4%) and Japan (20.9%). Even if these GM rice are mainly tested only in laboratories (70.6%), 23.8% of them have been subjected to field trials. Concerning the genetic information, a high variety of traits with their corresponding genes were observed (e.g. insect resistance, herbicide tolerance, biotic stress resistance, abiotic stress resistance, improved grain yield, healthier nutritional composition). Among the transformation vectors, the family of pCAMBIA vector has been identified as frequently used (544 GM rice (34.6%) in 359 peer-reviewed publications) (Cambia, Canberra, Australia). Besides, 30% of transgenic plants have been reported as transformed with this vector family (Komori et al., 2007). In addition, more than 70% of the collected GM rice possess the p35S and/or tNOS elements in their transgenic cassette.
Based on this overview and the information available in several databases (Biosafety Cleaning-House; CERA’s database; GMDD; GMO Compass), the key elements p35S and tNOS, present in both EU-authorized genetically modified organisms (GMOs) and EU-unauthorized GMOs (UGM), as well as the t35S element from the pCAMBIA family vector, not present in any EU-authorized GMO but found in around 30% of EU-unauthorized GM crop, were selected. These elements were targeted to develop an innovative and integrated strategy, based on DNA walking, in order to prove the presence of a broad spectrum of GMOs in the food/feed chain.
In this strategy, the presence of GMOs is first assessed by qPCR screening targeting the selected key elements (p35S, tNOS and t35S pCAMBIA). In case of a positive signal, the suspected presence of GMOs can be subsequent-ly confirmed by the DNA fragments containing the transgene flanking regions and/or the unnatural associations of elements obtained via the corresponding DNA walking methods using the same primers employed in the screening step (Fraiture et al., 2014, 2015a and 2015b) (Figure 1).
3332
Figure 1: DNA walking strategy applied on genetically modifi ed Bt rice (100%).
For each DNA walking method, a schematic representation of the potential start position and direction, applied on
the transgenic cassette of the Bt rice, is illustrated by the yellow (t35S pCAMBIA), green (p35S-R), blue (p35S-F), purple
These DNA walking methods were initially developed on pure material (100% Bt rice grains). Their performance were then successfully assessed in terms of sensitivity (going from 100% to 0.005% of the target) as well as applicability to a range of model food (rice fl our, rice noodle, rice-maize mixture, maize grain and maize powder) commonly encountered in GMO routine analysis (Fraiture et al., 2014, 2015a and 2015b). The proposed DNA walking strategy is thus a crucial molecular tool, which provides rapidly the results (in three working days) and is easily implementable by enforcement laboratories, in order to demonstrate the presence of GMOs in any given food/feed matrix. This strategy is now implemented in the food and feed samples analysed for GMO in the Belgian NRL-GMO (Laboratory PBB, Scientifi c Institute of Public health).
Acknowledgement
The research that yielded these results, was funded by the Belgian Federal Public Service of Health, Food Chain Safety and Environment through the contract UGMMONITOR (convention RF 11/6242). The authors would like also to thank Emmanuel Guiderdoni (CIRAD, UMR AGAP, Biological Systems department, Montpellier, France) for his kindness to provide rice grains.
Karakterisering van UGM met behulp van DNA walking strategie
Fraiture Marie-Alice1,2,3, Herman Philippe1, Papazova Nina1 en Roosens Nancy1
1 Wetenschappelijk Instituut Volksgezondheid (WIV-ISP), Platform Biotechnologie en moleculaire Biologie (PBB) en Dienst Bioveiligheid en Biotechnologie (SBB), J. Wytsmanstraat 14, 1050 Brussel, België2 Instituut voor Landbouw- en Visserijonderzoek (ILVO), Eenheid Technologie & Voeding (T&V), Burg. Van Gansberghelaan 115, 9820 Merelbeke, Belgium 3 University of Gent, Faculty of Pharmaceutical Sciences, Laboratory of Pharmaceutical Biotechnology, Ottergemsesteenweg 460, 9000 Ghent, Belgium
Om de productiviteit in de landbouw te verhogen werden biotechnische strategieën ontwikkeld die de productie toelaten van genetisch gemodificeerde (GG) gewassen. Aangezien rijst een van de meest overheersende gewassen is in de voedselketen, zijn in het kader van het project UGMONITOR(conventie RF 11/6242) meer dan 1000 peer-reviewde publicaties verzameld, van de periode van 1991 tot 2015, met als doel een overzicht te maken van alle transgene rijst lijnen ontwikkeld tot R&Dfase (IRRI, Scopus). De meeste van deze studies werden uitgevoerd in Azië (77,8%), vooral in China(47,4%) en Japan (20,9%). Hoewel deze GG rijst lijnen hoofdzakelijk in laboratoria getest werden(70,6%), ondergingen 23,8% daarvan ook veldproeven. Met betrekking tot de genetische informatiewerd een grote variatie van kenmerken met hun overeenkomstige genen waargenomen (bv resistentie tegen insecten, herbicidetolerantie, biotische stress-weerstand, abiotische stress-weerstand, verbeterde graanopbrengst, gezonder nutritionele samenstelling). De familie van pCambiavector werd geïdentificeerd als veel gebruikte familie onder de transformatievectoren (544 GM rijst (34,6%) in 359 peer-reviewde publicaties) (Cambia, Canberra, Australië). Overigens zijn 30% van detransgene planten gerapporteerd als getransformeerd met vectoren van deze familie (Komori et al.,2007). Bovendien bezitten meer dan 70% van de GG- rijst lijnen p35S en / of tNOS elementen in huntransgene cassette.
Op basis van dit overzicht en de beschikbare informatie in verschillende databases (Biosafety Cleaning-Huis; databank CERA's; GMDD; GMO Compass), werden de belangrijkste transgene elementen - p35S, tNOS en t35S_pCAMBIA geselecteerd. p35S en tNOS zijn aanwezig in zowel de EU geautoriseerde genetisch gewijzigde organismen (GGO’s) als in GGO’s die niet-geautoriseerd zijn in de EU. Het t35S element uit de pCambia vector familie is niet aanwezig in EU geautoriseerdeGGO's, maar kan wel gevonden worden in ongeveer 30% van de niet-geautoriseerde GG-gewassen. Op deze elementen werd gemikt om een innovatieve en geïntegreerde strategie te ontwikkelen,gebaseerd op DNA walking, met het oog op het detecteren van de aanwezigheid van een breed spectrum GGO’s in de voedsel- en voederketen.
In deze strategie wordt de aanwezigheid van GGO's eerst beoordeeld door het uitvoeren van qPCR screening op de geselecteerde transgene elementen (p35S, tNOS en t35S_pCAMBIA). Bij een positief signaal kan de vermoedelijke aanwezigheid van een GGO vervolgens bevestigd worden doorde DNA fragmenten die de transgene flankerende gebieden en/of onnatuurlijke combinaties van elementen bevatten, verkregen via de DNA walking strategie, gebruikmakend van dezelfde primers als voor de screening (Fraiture et al. , 2014, 2015a en 2015b) (Figuur 1).
Figuur 1: DNA walking toegepast op genetisch gewijzigde Bt rijst (100%).
Voor elke DNA walking methode werd een schematische voorstelling van de mogelijkestartpositie en richting aangebracht op de transgene cassette van de Bt rijst, in het geel (t35S pCAMBIA), in groen (p35S-R), in blauw (p35S-F), in paars (tNOS-R) en in roze (tNOS-F). LB(linker ‘border’ gebied); t35S (CaMV 35S terminator); hpt (hygromycine fosfotransferase gen); p35S (CaMV 35S promoter); lacZ (LacZ-alfa-fragment); Pubi (maïs ubiquitine promotor);
34
References
1. Biosafety Cleaning House. URL <http://bch.cbd.int/database/organisms/>.2. Breitler, J. C., Vassal, J. M., del Mar Catala, M., Meynard, D., Marfa, V., Melé, E., Royer, M., Murillo, I., San Segundo, B., Guiderdoni, E. & Messeguer, J. (2004). Bt rice harbouring cry genes controlled by a constitutive or wound-inducible promoter: protection and transgene expression under Mediterranean field conditions. Plant Biotechnology Journal, 2, 417-430.3. Cambia, Canberra, Australia. URL <http://www.cambia.org/daisy/bioforge_legacy/3724.html>.4. Center for Environmental Risk Assessment (CERA). URL <http://www.cera-gmc.org/?action=gm_crop_database>.5. ENGL ad hoc working group on “unauthorised GMOs” (2011). Overview on the detection, interpretation and reporting on the presence of unauthorised genetically modified materials. URL < http://gmo-crl.jrc.ec.europa.eu/doc/2011-12-12%20ENGL%20UGM%20WG%20Publication.pdf>.6. Fraiture, M. A., Herman, P., Taverniers, I., De Loose, M., Deforce, D., Roosens, N. H. (2014). An innovative and integrated approach based on DNA walking to identify unauthorised GMOs. Food Chemistry, 147:60-69.7. Fraiture, M. A., Herman, P., Taverniers, I., De Loose, M., Van Nieuwerburgh, F., Deforce, D., Roosens, N. H. (2015a). Validation of a sensitive DNA walking strategy to characterize unauthorised GMOs using model food matrices mimicking common rice products. Food Chemistry, 173:1259-1265.8. Fraiture, M. A., Herman, P., Lefèvre, L., Taverniers, I., De Loose, M., Deforce, D., Roosens, N. H. (2015b). High coverage and integrated DNA walking system to characterize GMOs in food/feed matrices. (submitted).9. GMDD. URL<http://gmdd.shgmo.org/index/search>10. GMO Compass. URL <http://www.gmo-compass.org/eng/gmo/db/>.11. IRRI. <http://www.irri.org/our-work/research>12. Komori, T., Imayama, T., Kato, N., Ishida, Y., Ueki, J. & Komari, T. (2007). Current Status of Binary Vectors and Superbinary Vectors. Plant Physiology, 145, 1155-1160.13. Scopus (transgenic rice). URL <http://www.scopus.com/home.url>.
The trainings for the approved laboratories organized by the FASFC in co-operation with the National Reference Laboratories are available on the website of the FASFC
(www.favv.be > Business Sectors > Laboratories > Trainings).
The schedule is updated regularly, it is therefore recommended to check the website from time to time.
Other interesting workshops and symposia are mentioned below.
3736
20-21.06.2016
18th International Fresenius AGRO ConferenceBehaviour of Pesticides in Air, Soil and Water
