Labeling Milk along Its Production Chain with DNA Encapsulated in

Oct 8, 2014 - Loss of miRNAs during Processing and Storage of Cow's (Bos taurus) Milk. Katherine M. Howard , Rio Jati Kusuma , Scott R. Baier , Taylor...
2 downloads 0 Views 2MB Size
Article pubs.acs.org/JAFC

Labeling Milk along Its Production Chain with DNA Encapsulated in Silica Madeleine S. Bloch, Daniela Paunescu, Philipp R. Stoessel, Carlos A. Mora, Wendelin J. Stark, and Robert N. Grass* Institute for Chemical and Bioengineering, Department of Chemistry and Applied Biosciences, ETH Zurich, Vladimir-Prelog-Weg 1, 8093 Zurich, Switzerland S Supporting Information *

ABSTRACT: The capability of tracing a food product along its production chain is important to ensure food safety and product authenticity. For this purpose and as an application example, recently developed Silica Particles with Encapsulated DNA (SPED) were added to milk at concentrations ranging from 0.1 to 100 ppb (μg per kg milk). Thereby the milk, as well as the milk-derived products yoghurt and cheese, could be uniquely labeled with a DNA tag. Procedures for the extraction of the DNA tags from the food matrixes were elaborated and allowed identification and quantification of previously marked products by quantitative polymerase chain reaction (qPCR) with detection limits below 1 ppb of added particles. The applicability of synthetic as well as naturally occurring DNA sequences was shown. The usage of approved food additives as DNA carrier (silica = E551) and the low cost of the technology (99.9%), toluene (>99.7%), and cupric chloride anhydrous (CuCl2, >97%) were used from Fluka. Ammonia Received: July 17, 2014 Revised: October 7, 2014 Accepted: October 8, 2014

A

dx.doi.org/10.1021/jf503413f | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

Figure 1. (a) Particle size distribution and a characteristic electron microscopy image of the applied Silica Particles with Encapsulated DNA (SPED). (b) Milk is complemented with SPED with a specific code and subsequently manufactured to yoghurt and cheese.

Figure 2. Flowchart for incorporation of SPED into milk, cheese, and yoghurt and subsequent extraction of the marker from the foodstuff by washing/centrifugation procedures for final analysis via qPCR. sample well was filled with 5 μL of sample volume, 10 μL of master mix (Lightcycler 480 SYBR Green I Master, Roche), 3 μL of PCRgrade water, 1 μL of forward primer, and 1 μL of reverse primer (2 μM each). The qPCR consisted of a three-step amplification with 45 cycles. The samples were first heated to 95 °C for 15 s, then remained at 60 °C for 30 s, and last were held at 72 °C for 30 s. Each sample was analyzed in replicates (technical triplicates). Food Products. Pasteurized milk (0.5 L, 3.9% fat, organic) and yoghurt (180 g, 4% fat, Naturaplan) were bought at Coop, ETH Hönggerberg. The rennet tablets were supplied by Swissmilk (Winkler AG, approximately 95% chymosin, 5% pepsin, salt as support material). DNA Codes. Sequence MB1 is a randomly generated DNA sequence, which was selected for good PCR amplification properties and checked for specificity using Basic Local Alignment Search Tool (BLAST):26 (5′−3′) ATT GCA CCC TTA CCA CGA AGA CAG GTT TGT CCA ATC CCA TAT TGC GAC CTT GGC AGG GGG TTC GCA AGT CCC ACC CGA AAC GTT GCT GAA GGC TCA GGT TTC TGA GCG ACA AAA GC. Sequence TOM1 is a short sequence taken from the tomato genome (species Solanum lycopersicum), Chromosome 1, NC_015438.1: (5′−3′) TGA AGG ACT GAA GCA ACA CGA AAC CTA TAC AAC TAA CTA GCA GAG CTC TCG ATT TTG CTA ATT CAT CAT AAC TCA TAC TAT GAG CAA AAA AGT AAA GGG AAA TTT TCG TGC ATG GTG TTG A.

solution (25 wt %), hydrogen peroxide (H2O2, 30 wt %), and ammonium hydrogen difluoride (NH4FHF, pure) were purchased from Merck. Isopropanol (>99.8%), ammonium fluoride (NH4F, puriss.), and ammonium hydroxide solution (25 wt %) were used as supplied from Sigma-Aldrich. Tetraethoxysilane (TEOS, ≥99.0%) was received from Aldrich, N-trimethoxysilylpropyl-N,N,N-trimethylammonium chloride (TMAPS, 50 wt % in methanol) from ABCR, L-ascorbic acid sodium salt (99%) from Acros Organics, and hexyltrimethoxysilane (>96%) from Tokyo Chemical Industry. Sodium dodecyl sulfate (SDS, >98%) was purchased from Sigma. Qubit dsDNA high-sensitivity (HS) assay was supplied by Invitrogen, QIAquick PCR purification kit from Quiagen, and EDTA disodium salt dihydrate (99%) from Biosolve. Ultrapure water (18.2 MΩ cm at 25 °C) was used. The double-stranded DNA solutions for the 116 base-pair (bp) amplicon MB1 and the 121 bp amplicon TOM1 (from tomato DNA sequence) as well as their corresponding primers were purchased from Microsynth. MB1 was used for the experiments with milk, yoghurt, cheese, and whey, while TOM1 was only used in one yoghurt experiment. Real-Time Quantitative PCR. The qPCR was executed with a Roche Lightcycler 96. Based on the cycle of threshold (Ct), conclusions on the initial DNA concentration in the sampleand with that, the SPED concentrationcan be drawn. The positive controls for the PCR were obtained by dissolving known SPED amounts directly in buffered oxide etch (BOE) solution. A PCR blank (no-template control) confirmed clean working conditions. Each B

dx.doi.org/10.1021/jf503413f | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

Statistical Analysis. The validity of the SPED labeling method was proven. Please find the discussion on the necessary variables in the Supporting Information. The experimental data were analyzed with Origin 8.6. The twosample t tests were based on a 5% significance level. One-way analysis of variance (ANOVA) was used to compare the sample means at different time points.

Particle Synthesis. To produce the SPED, the protocol of Paunescu et al.22 was applied. Particles containing either the DNA sequence MB1 or the DNA sequence TOM1 were produced (for sequences, see above). Following their production, the SPED were tested on their radical treatment stability as previously described.22 The particle structure was analyzed by scanning transmission electron microscopy (STEM, NovaNanoSEM 450, FEI) at 30 kV. The particle size was determined by volume-weighted sedimentation analysis (LUMiSizer, LUM GmbH). Labeling of Food. Milk. SPED concentrations of 10, 1, and 0.1 ppb were prepared in milk. For every SPED concentration, ten 0.5 mL samples of pasteurized milk were labeled by adding a 5 μL of SPED dispersion at prediluted concentrations to give the final SPED concentration. Yoghurt. The yoghurt was spiked with 100, 10, 1, and 0.1 ppb SPED containing the MB1 sequence. Additionally, yoghurt with 10 ppb SPED containing DNA code TOM1 was made. To produce 100 g of yoghurt, 87 g of pasteurized milk, 13 g of yoghurt, and 10 μL of prediluted SPED dispersion were used. After vortexing, the yoghurt was placed into an incubator at 42 °C for 5.5 h. Subsequently, the glasses were carefully transferred to a refrigerator (T = 4−6 °C) and left there overnight. Experiments were performed in triplicates (three yoghurts for every concentration). From every yoghurt, two samples were analyzed. To ensure a representative sample, the yoghurt was stirred before sampling. Cheese. For the cheese production, 5.8 g of yoghurt, 100 g of pasteurized milk, and a defined amount of SPED were mixed in a beaker. The mixture was added to 900 g of pasteurized milk. The detailed cheese recipe can be found in the Supporting Information. Label Extraction from Food. Milk. For the extraction, the 0.5 mL sample was mixed with 1 mL of SDS solution (10 wt %) in a 2 mL Eppendorf tube (Figure 2). The sample was sonicated for 30 s using an ultrasonicator (Ultraschall-Prozessor UP50H, Huberlab) and then centrifuged for 3 min at 21500g. The supernatant was discarded, while 50 μL of ultrapure water was added to the pellet. The sample was vortexed for 1 min before 70 μL of the BOE solution was added. After the final purification with a QIAquick PCR Purification Kit, a qPCR was performed with a Roche Lightcycler 96. See Figure S5 in the Supporting Information for a flow diagram of the extraction method. Yoghurt. To extract the SPED, 0.5 g of the yoghurt was mixed with 1 mL of SDS solution (10 wt %) in a 2 mL Eppendorf tube (Figure 2). After a quick vortex step, the sample was sonicated for 30 s. The tubes were centrifuged for 3 min at 21500g. The resulting supernatant was discarded, while 1 mL of SDS solution (10 wt %) was added to the pellet. One minute of vortexing was followed by another centrifugation step for 3 min at 21500g. The supernatant was discarded, while 50 μL of ultrapure water was given to the pellet. The sample was then vortexed for another minute. Finally, 70 μL of BOE solution was pipetted to the sample. As a last step, the QIAquick PCR Purification Kit was applied. The resulting liquid was directly used for qPCR analysis. See Figure S6 in the Supporting Information for a flow diagram of the extraction method. Cheese. From the cheese, a cylindrical-shaped piece was cut out (Figure 2). From the cross section, three 50 mg samples from the cylinder’s rind, core, and the space between the rind and the core were combined to a collective sample and placed into a 2 mL microtube. Next, 1.5 mL of SDS solution (10 wt %) and two speed mixing beads (chrome steel, d = 6.35 mm) were added. The sample was then homogenized with a Mini-Beadbeater (Biospec Products) for 70 s at a speed of 4200 rpm. After the beads had been removed from the tubes, the latter were centrifuged for 7 min at 21500g. Following a repetition of the mechanical homogenization at the same conditions, the sample was washed twice with 1 mL of SDS solution (10 wt %) and once with ultrapure water, with intermediate centrifugation steps. After addition of 50 μL of ultrapure water, the sample was vortexed for 1 min, and 70 μL of the BOE solution was added. The sample was then shaken at 24 °C for 30 min before it was purified using the QIAquick PCR Purification Kit for qPCR analysis. See Figure S7 in the Supporting Information for a flow diagram of the extraction method.



RESULTS AND DISCUSSION SPED are aggregated silica particles with a primary particle size of ∼140 nm and an agglomerate size of ∼1200 nm (see Figure 1a), much resembling the structure of commercial silica products (e.g., Aerosil, fumed silica, and precipitated silica). The particles are easily dispersed in water and can be directly mixed with aqueous liquids at various concentrations. The silica can be loaded with ∼0.1 wt % of DNA, and in this work both a random (MB1) and a naturally occurring (TOM1) DNA code were utilized. The DNA molecules are well encapsulated within the silica matrix, as proven by exposing the particles to aggressive radical oxygen species (stability >90%, see Table S2 in the Supporting Information).23 Upon mixing with fluoridecontaining solutions (e.g., buffered oxide etch (BOE), a pHstabilized NH4F solution) the silica can be dissolved, releasing the DNA and allowing its analysis by qPCR without further treatment.22 In the case of food products, the particles first have to be extracted from the complex matrix, as food components such as proteins and fats are known to interfere with PCR analysis. Therefore, for every food type, a processing procedure was elaborated (see Figure 2 and Figures S5−S7 in the Supporting Information). These processing procedures essentially include multiple SDS washing and particle centrifugation steps. In the case of the cheese, the semisolid matrix was first mechanically broken down utilizing a mini-bead-beater. Following the washing and centrifugation procedure, all cases resulted in (hardly visible) particle deposits in the tubes, without any visual residue of proteins or fat. The deposits were dissolved in BOE, and then the DNA was purified with commercial DNA purification kits and directly analyzed via qPCR using the corresponding primers. qPCR analysis was done in a traditional way, where the cycle was monitored (= Ct) at the point at which the fluorescence signal exceeded a given threshold (threshold computed automatically from signalto-noise ratio by Lightcycler instrument). In this way the DNA, and by extension also the particles, could be detected down to 0.1 ppb for milk and yoghurt and down to 1 ppb for cheese (Figure 3 and Supporting Information). In comparison, the analysis of pesticide residues in fruit and vegetable samples with gas chromatography is possible to a limit of detection of 20 ppb.27 To investigate whether the DNA/particle detection method is quantitative, the foodstuffs under investigation were labeled with various SPED concentrations in the ppb range. Based on the exponential nature of the qPCR method, we did not expect to be able to measure small concentration differences, but rather concentrations on a logarithmic scale. Indeed, the Ct depended linearly (R2 = 0.98 for yoghurt, R2 = 0.99 for milk and cheese) on logarithmic particle concentration added to the individual foods (see Figure S1 and Table S1 in the Supporting Information). Additionally, in milk, yoghurt, and cheese, particle concentrations could be statistically discriminated on a log 10 scale, i.e., 0.1 ppb could be discriminated from 1 ppb, and 1/10 ppb samples were statistically different (p < 0.05, Figure 3). Further control experiments (data not shown) C

dx.doi.org/10.1021/jf503413f | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

utilized a fragment (121 bp) of tomato DNA (species Solanum lycopersicum) as code. The sequence was selected from chromosome 1 of the tomato’s DNA for optimal primers utilizing an online primer search algorithm (Primer3)28 and tested for uniqueness using the Basic Local Alignment Search Tool (BLAST).26 Yoghurt manufactured with 10 ppb of these particles could be identified by qPCR if the corresponding TOM1 primers were utilized (Ct = 28.77 ± 0.27). Utilizing the same procedure but analyzing with non-corresponding primers (MB1 primers) gave no significant PCR amplification (Ct = Ct(blank)), proving the selectivity of PCR analysis, which is well known and reported in medical diagnostics29 and bacteria detection.30 The SPED labeling method is much more quantifiable compared to tagging approaches where a microbial cell with a DNA discriminable from the somatic cells of the milk was applied. Bacterial cells pass through a growth cycle with replication and decline. Their cell number and DNA amount is therefore not at all times the same. With our method, we could prove that the amount of DNA encapsulated in silica was stable over a course of several weeks. The SPED ensure, therefore, a more precise tagging than other methods do. When silica particles loaded with the natural source DNA sequence (TOM1) are utilized to mark yoghurt at a particle loading of 1 ppb, a consumer eating 250 g of the yoghurt would consume 0.25 μg of silica and 0.25 ng of tomato DNA. This is equivalent to 0.0002% of the average daily dietary silica intake31 and equivalent to the DNA present in 0.0001% of one fresh tomato.32,33 Silica particles have a long history as food additives (E551) and are approved for several uses in European and U.S. markets. Based on the availability of billions of base-pair sequences from food sources, the use of naturally occurring DNA sequences as a coding tool can easily be envisioned. Combinations of mixtures of particles carrying several codes may further increase the variability of the code library. In order to estimate the price of the DNA in silica food tagging technology presented here, we utilized currently available price data. The main cost driver of the technology is the synthetic DNA code, at a price of 9000 USD/g (as per commercial offer from Microsynth in March 2014; costs of additional raw materials are negligible). From 1 g of DNA, about 1 kg of particles can be produced (loading of ∼1 μg/mg). If a SPED concentration of 10 ppb is applied to a food product, the labeling would cost 0.00009 USD per kg (i.e., 9¢ per ton) of food plus the labor costs, which will depend on the cost structure of the particle-producing company. This short calculation shows that in a potential application of the technology in tracing food, the cost of the DNA label is not critical. From a purely technical and chemical point of view, above justification of material choice and cost estimation shows that the application of DNA in silica particles as a food-marking tool is of both low risk and low cost. Still, it will be evident to the reader that the ongoing public discussion on the risk of both DNA technology and nanotechnology (including discussions on the general use of amorphous silica as a food additive)34 may slow down the application of these tracers in foodstuff. As with most new technologies, the potential risk of the additive must be balanced with the potential opportunity of uniquely tagging foodstuff. The answer to this question can only be the result of future discussions between many stakeholders, including consumers, producers, customs agencies, and local governments. In terms of food additive regulation, the use of an

Figure 3. qPCR results (cycle of threshold, Ct) for the DNA within the particles detected at various particle loadings (ppb = μg particles per kg of food). *Denotes statistical significance at p < 0.05. The linearity of the Ct vs log particle concentration is shown in Figure S1 in the Supporting Information.

revealed that the particle concentration was independent of the sampling position in the cheese sample (yoghurt and milk samples were homogenized prior to particle extraction). An additional aspect that is of importance with regard to quantitative analysis is the stability of the marker within the matrix over time. We have previously shown that, based on its encapsulation in silica, the DNA is protected from environmental decay over long storage times.23 In order to display that this is true within the food products under investigation, additional tests were performed: SPED particles were added to cheese and yoghurt at 1 ppb and analyzed over time (2 weeks for yoghurt; 4 weeks for cheese). As visible in Figure 4, the

Figure 4. Cycle of threshold for cheese and yoghurt labeled with 1 ppb SPED. The samples remained stable over a course of 4 and 2 weeks for cheese and yoghurt, respectively.

qPCR readings (Ct values) did not increase over time, indicating that the DNA within the particles is unharmed during storage. This further shows that absolute quantification was feasible, even after several weeks of storage. Of course, for cheese, longer storage durations are of interest and are under current investigation. The application of natural DNA as a unique sequence may be an advantage for consumer acceptance. For this reason we D

dx.doi.org/10.1021/jf503413f | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

(8) Kelly, S.; Heaton, K.; Hoogewerff, J. Tracing the Geographical Origin of Food: The Application of Multi-Element and Multi-Isotope Analysis. Trends. Food Sci. Technol. 2005, 16, 555−567. (9) Zhao, Y.; Zhang, B.; Chen, G.; Chen, A.; Yang, S.; Ye, Z. Tracing the Geographic Origin of Beef in China on the Basis of the Combination of Stable Isotopes and Multielement Analysis. J. Agric. Food Chem. 2013, 61, 7055−7060. (10) Camin, F.; Wietzerbin, K.; Cortes, A. B.; Haberhauer, G.; Lees, M.; Versini, G. Application of Multielement Stable Isotope Ratio Analysis to the Characterization of French, Italian, and Spanish Cheeses. J. Agric. Food Chem. 2004, 52, 6592−6601. (11) Scampicchio, M.; Mimmo, T.; Capici, C.; Huck, C.; Innocente, N.; Drusch, S.; Cesco, S. Identification of Milk Origin and ProcessInduced Changes in Milk by Stable Isotope Ratio Mass Spectrometry. J. Agric. Food Chem. 2012, 60, 11268−11273. (12) Versini, G.; Monetti, A.; Reniero, F. Monitoring Authenticity and Regional Origin of Wines by Natural Stable Isotope Ratios Analysis. Wine: nutritional and therapeutic benefits; ACS Symposium Series 661; American Chemical Society: Washington, DC, 1997. (13) Ehtesham, E.; Hayman, A. R.; McComb, K. A.; Van Hale, R.; Frew, R. D. Correlation of Geographical Location with Stable Isotope Values of Hydrogen and Carbon of Fatty Acids from New Zealand Milk and Bulk Milk Powder. J. Agric. Food Chem. 2013, 61, 8914− 8923. (14) Bugaud, C.; Buchin, S.; Coulon, J.-B.; Hauwuy, A.; Dupont, D. Influence of the Nature of Alpine Pastures on Plasmin Activity, Fatty Acid and Volatile Compound Composition of Milk. Lait 2001, 81, 401−414. (15) Mauriello, G.; Moio, L.; Genovese, A.; Ercolini, D. Relationships Between Flavoring Capabilities, Bacterial Composition, and Geographical Origin of Natural Whey Cultures Used for Traditional Water-Buffalo Mozzarella Cheese Manufacture. J. Dairy Sci. 2003, 86, 486−497. (16) Sampaio, O. M.; Reche, R. V.; Franco, D. W. Chemical Profile of Rums as a Function of Their Origin. The Use of Chemometric Techniques for Their Identification. J. Agric. Food Chem. 2008, 56, 1661−1668. (17) Mafra, I.; Ferreira, I. M.; Oliveira, M. B. P. Food Authentication by PCR-Based Methods. Eur. Food Res. Technol. 2008, 227, 649−665. (18) Conyers, C. M.; Allnutt, T. R.; Hird, H. J.; Kaye, J.; Chisholm, J. Development of a Microsatellite-Based Method for the Differentiation of European Wild Boar (Sus Scrofa Scrofa) from Domestic Pig Breeds (Sus Scrofa Domestica) in Food. J. Agric. Food Chem. 2012, 60, 3341− 3347. (19) Mininni, A. N.; Pellizzari, C.; Cardazzo, B.; Carraro, L.; Balzan, S.; Novelli, E. Evaluation of Real-Time PCR Assays for Detection and Quantification of Fraudulent Addition of Bovine Milk to Caprine and Ovine Milk for Cheese Manufacture. Int. Dairy J. 2009, 19, 617−623. (20) Fang, W.; Meinhardt, L. W.; Mischke, S.; Bellato, C. M.; Motilal, L.; Zhang, D. Accurate Determination of Genetic Identity for a Single Cacao Bean, Using Molecular Markers with a Nanofluidic System, Ensures Cocoa Authentication. J. Agric. Food Chem. 2014, 62, 481− 487. (21) Casey, M. G.; Isolini, D.; Amrein, R.; Wechsler, D.; Berthoud, H. Naturally Occurring Genetic Markers in Lactobacilli and Their Use to Verify the Authenticity of Swiss Emmental PDO Cheese. Dairy Sci. Technol. 2008, 88, 457−466. (22) Paunescu, D.; Puddu, M.; Soellner, J. O. B.; Stoessel, P. R.; Grass, R. N. Reversible DNA Encapsulation in Silica to Produce ROSResistant and Heat-Resistant Synthetic DNA “Fossils”. Nat. Protoc. 2013, 8, 2440−2448. (23) Paunescu, D.; Fuhrer, R.; Grass, R. N. Protection and Deprotection of DNAHigh-Temperature Stability of Nucleic Acid Barcodes for Polymer Labeling. Angew. Chem. 2013, 52, 4269−4272. (24) Mora, C. A.; Paunescu, D.; Grass, R. N.; Stark, W. J. Silica Particles with Encapsulated DNA as Trophic Tracers. Mol. Ecol. Resour. 2014, DOI: 10.1111/1755-0998.12299. (25) Puddu, M.; Paunescu, D.; Stark, W. J.; Grass, R. N. Magnetically Recoverable, Thermostable, Hydrophobic DNA/Silica Encapsulates

additive as internal food label has not been addressed so far, so neither the Codex Alimentarius nor the European numbering system (EFSA) has a corresponding functional category. This study shows that DNA-loaded silica particles can be utilized as tracers in milk, yoghurt, and cheese. The technology requires only marginal amounts of resources and has a broad range of label generation possibilities. While quantification has to be normalized for every food type (standard curve per food type), the result is quantitative on a logarithmic scale for tracer concentrations as low as 0.1 ppb (i.e., 0.1 μg of tracer per kg of foodstuff). The stability of the DNA tracer in the processing of milk to yoghurt and cheese shows the stability of the tracer over the production chain. The possibility of tracer read-out from food matrixes as complex as cheese and yoghurt displays the robustness of the method, and the applicability of the tracer to other food types is anticipated.



ASSOCIATED CONTENT

S Supporting Information *

More information on the validity of the method, cheese manufacture, calculation of sedimentations times, and process flowcharts. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*Tel: + 41 44 633 63 34. Fax: + 41 44 633 10 83. E-mail: [email protected]. Funding

The Institute for Chemical and Bioengineering (ICB) at ETH Zurich and the Swiss National Science Foundation (no. 200021-150179) provided financial support. Notes

The authors declare the following competing financial interest(s): R.N.G. and W.J.S. declare financial interest in the form of a patent application on DNA encapsulation licenced to TurboBeads Llc, of which R.N.G. and W.J.S. are shareholders and R.N.G. is a part-time employee. M.S.B., D.P., P.R.S., and C.A.M. have no competing financial interests..



REFERENCES

(1) European Union. Regulation (EU) No. 1151/2012 on Quality Schemes for Agricultural Products and Foodstuffs, 2012. (2) Dalvit, C.; De Marchi, M.; Cassandro, M. Genetic Traceability of Livestock Products: A Review. Meat Sci. 2007, 77, 437−449. (3) Chansud, W.; Wisanmongkol, J.; Ketprom, U. RFID for Poultry Traceability System at Animal Checkpoint. In Electrical Engineering/ Electronics, Computer, Telecommunications and Information Technology; ECTI-CON 2008, 5th International Conference; IEEE: New York, 2008; Vol. 2, pp 753−756. (4) Chao, C.-C.; Yang, J.-M.; Jen, W.-Y. Determining Technology Trends and Forecasts of RFID by a Historical Review and Bibliometric Analysis from 1991 to 2005. Technovation 2007, 27, 268−279. (5) Regattieri, A.; Gamberi, M.; Manzini, R. Traceability of Food Products: General Framework and Experimental Evidence. J. Food Eng. 2007, 81, 347−356. (6) Shanahan, C.; Kernan, B.; Ayalew, G.; McDonnell, K.; Butler, F.; Ward, S. A Framework for Beef Traceability from Farm to Slaughter Using Global Standards: An Irish Perspective. Comput. Electron. Agric. 2009, 66, 62−69. (7) Cabana, G.; Rasmussen, J. B. Modelling Food Chain Structure and Contaminant Bioaccumulation Using Stable Nitrogen Isotopes. Nature 1994, 372, 255−257. E

dx.doi.org/10.1021/jf503413f | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

and Their Application as Invisible Oil Tags. ACS Nano 2014, 8, 2677− 2685. (26) Altschul, S. F.; Gish, W.; Miller, W.; Myers, E. W.; Lipman, D. J. Basic Local Alignment Search Tool. J. Mol. Biol. 1990, 215, 403−410. (27) Fillion, J.; Sauvé, F.; Selwyn, J. Multiresidue Method for the Determination of Residues of 251 Pesticides in Fruits and Vegetables by Gas Chromatography/Mass Spectrometry and Liquid Chromatography with Fluorescence Detection. J. AOAC Int. 2000, 83, 698−713. (28) Koressaar, T.; Remm, M. Enhancements and Modifications of Primer Design Program Primer3. Bioinformatics 2007, 23, 1289−1291. (29) Rohr, U.-P.; Wulf, M.-A.; Stahn, S.; Steidl, U.; Haas, R.; Kronenwett, R. Fast and Reliable Titration of Recombinant AdenoAssociated Virus Type-2 Using Quantitative Real-Time PCR. J. Virol. Methods 2002, 106, 81−88. (30) Hein, I.; Lehner, A.; Rieck, P.; Klein, K.; Brandl, E.; Wagner, M. Comparison of Different Approaches to quantifyStaphylococcus Aureus Cells by Real-Time Quantitative PCR and Application of This Technique for Examination of Cheese. Appl. Environ. Microbiol. 2001, 67, 3122−3126. (31) Dekkers, S.; Krystek, P.; Peters, R. J.; Lankveld, D. P.; Bokkers, B. G.; van Hoeven-Arentzen, P. H.; Bouwmeester, H.; Oomen, A. G. Presence and Risks of Nanosilica in Food Products. Nanotoxicology 2011, 5, 393−405. (32) Bertin, N.; Gautier, H.; Roche, C. Number of Cells in Tomato Fruit Depending on Fruit Position and Source-Sink Balance during Plant Development. Plant Growth Regul. 2002, 36, 105−112. (33) Arumuganathan, K.; Earle, E. D. Nuclear DNA Content of Some Important Plant Species. Plant Mol. Biol. Rep. 1991, 9, 208−218. (34) Dekkers, S.; Bouwmeester, H.; Bos, P. M.; Peters, R. J.; Rietveld, A. G.; Oomen, A. G. Knowledge Gaps in Risk Assessment of Nanosilica in Food: Evaluation of the Dissolution and Toxicity of Different Forms of Silica. Nanotoxicology 2013, 7, 367−377.

F

dx.doi.org/10.1021/jf503413f | J. Agric. Food Chem. XXXX, XXX, XXX−XXX