Detection of allergenic ingredients using real-time PCR: a case study on hazelnut (Corylus avellena) and soy (Glycine max)

Compliance with the European allergen labeling legislation (Directive 2007/68/EC) is only possible when coupled with appropriate methods to detect allergens in food. The aim of the current study was to develop new real-time PCR assays for the detection of hazelnut and soy and evaluate these assays via comparison with commercially available kits. Although the new assays were not as sensitive as the commercial qualitative assays, they proved to be more specific. Moreover, the cross-reactivity study indicated contamination of some of the food products used with either hazelnut or soy, which presents a risk for the allergic consumer. The assays were able to quantify as few as 5-15 genome copies. This unit, used to express analytical results for allergen detection by means of PCR, needs to be converted to a unit expressing the amount of allergenic ingredient in order to be informative. This study emphasizes that the use of real-time PCR for allergen quantification is complicated by the lack of appropriate reference materials for allergens.     

Quantitative sandwich ELISA for determination of traces of hazelnut (Corylus avellana) protein in complex food matrixes.

A hazelnut-specific sandwich-type ELISA based on polyclonal antisera was developed for detection of hidden hazelnut protein residues in complex food matrixes. In the absence of a food matrix, extractable protein from different native and toasted hazelnuts was detected at rates of 94 +/- 13 and 96 +/- 7% applying standards prepared from native and toasted hazelnuts, respectively. From complex food matrixes, 0.001-10% of hazelnut was recovered between 67 and 132%, in average by 106 +/- 17%. Depending on the food matrix, hazelnut protein could be detected down to the ppb (ng/g) level. Intraassay precision was <6% for hazelnut >/= 0.001% and interassay precision was <15% for hazelnut >/= 0.01%. In 12 of 28 commercial food products without labeling or declaration of hazelnut components, between 2 and 421 ppm of hazelnut protein was detected, demonstrating a remarkable presence of potentially allergenic hazelnut protein "hidden" in commercial food products.     

Differential detection of shrimp and crab for food labeling using polymerase chain reaction

Shrimp and crab are well-known as allergenic ingredients. According to Japanese food allergy labeling regulations, shrimp species (including prawns, crayfishes, and lobsters) and crab species must be differentially declared when ≥10 ppm (total protein) of an allergenic ingredient is present. However, the commercial ELISA tests for the detection of crustacean proteins cannot differentiate between shrimp and crab. Therefore, two methods were developed to discriminate shrimp and crab: a shrimp-PCR method with postamplification digestion and a crab-PCR method that specifically amplifies a fragment of the 16S rRNA gene. The sensitivity and specificity of both PCR methods were verified by experiments using DNA extracted from 15 shrimp species, 13 crab species, krill, mysid, mantis shrimp, other food samples (cephalopod, shellfish, and fish), incurred foods, and commercial food products. Both PCR methods could detect 5 pg of DNA extracted from target species and 50 ng of genomic DNA extracted from incurred foods containing 10 ppm (μg/g) total protein of shrimp or crab. The two PCR methods were considered to be specific enough to separately detect species belonging to shrimp and crab. Although false-positive and false-negative results were obtained from some nontarget crustacean species, the proposed PCR methods, when used in conjunction with ELISA tests, would be a useful tool for confirmation of the validity of food allergy labeling and management of processed food safety for allergic patients.     

Detection of potentially allergenic hazelnut (Corylus avellana) residues in food: a comparative study with DNA PCR-ELISA and protein sandwich-ELISA

Allergen detection is of increasing interest for food labeling purposes. A comparative study with a commercial hazelnut-specific PCR-ELISA and a sandwich-type ELISA detecting hazelnut protein was performed to investigate to what extent immunochemical and DNA-based techniques would correlate in the detection of trace amounts of potentially allergenic hazelnut residues. Both methods were highly sensitive and allowed the detection of even <10 ppm of hazelnut in complex food matrixes. The protein-ELISA was highly specific for hazelnut. However, some foods could lead to false-positive results at the 10 ppm level. The PCR-ELISA did not show any cross-reactions with non-hazelnut foods, thus reducing the probability of having false positives at the trace level. Forty-one commercial food products with and without hazelnut components on their labels were analyzed for the presence of hazelnut. Of the 27 products in which hazelnut components were detected, two samples were not identified by the protein-ELISA, and only one sample, namely one white chocolate having <1 ppm of hazelnut protein, was not detected by PCR-ELISA. The good correlation of the results of PCR-ELISA and protein-ELISA suggested that both PCR-based and immunochemical techniques are suitable for reliable detection of potentially allergenic hazelnut residues in foods at the trace level.     

Development of a real-time PCR and a sandwich ELISA for detection of potentially allergenic trace amounts of peanut (Arachis hypogaea) in processed foods

Hidden allergens in food products are, especially for peanut-allergic consumers, a serious problem because even low amounts (approximately 200 microg) of peanut can elicit allergic reactions. Undeclared peanut traces can be found in processed food products, because contaminations with peanut during production processes are frequent. To minimize the risk of such cross-contaminations, it is necessary to develop sensitive analytical methods for the detection of hidden allergens in foods. For this approach we developed two peanut-specific assays based on the detection of peanut protein by specific antibodies (sandwich ELISA) and by the detection of peanut-specific DNA (part of the coding region of Ara h 2) by a real-time PCR. Both tests did not show any cross-reactivity with 22 common food ingredients (cereals, nuts, legumes), and the limit of detection is <10 ppm peanut in processed foods. Thirty-three random samples of food products were tested for the presence of peanut to compare both assay types with each other and to evaluate the percentage of foods on the German market that are contaminated with peanut traces. We found that four products (13.3%) without peanut in the list of ingredients contained peanut protein in a range from 1 to 74 ppm peanut protein and that the results of both tests correlated well. The real-time PCR was able to detect one more positive sample than the sandwich ELISA. In conclusion, both assays are sensitive and specific tools for the detection of hidden allergens in processed foods.     

A novel approach for the detection of potentially hazardous pepsin stable hazelnut proteins as contaminants in chocolate-based food

Contamination of food products with pepsin resistant allergens is generally believed to be a serious threat to patients with severe food allergy. A sandwich type enzyme-linked immunosorbent assay (ELISA) was developed to measure pepsin resistant hazelnut protein in food products. Capturing and detecting rabbit antibodies were raised against pepsin-digested hazelnut and untreated hazelnut protein, respectively. The assay showed a detection limit of 0.7 ng/mL hazelnut protein or <1 microg hazelnut in 1 g food matrix and a maximum of 0.034% cross-reactivity (peanut). Chocolate samples spiked with 0.5-100 microg hazelnut/g chocolate showed a mean recovery of 97.3%. In 9/12 food products labeled "may contain nuts", hazelnut was detected between 1.2 and 417 microg hazelnut/g food. It can be concluded that the application of antibodies directed to pepsin-digested food extracts in ELISA can facilitate specific detection of stable proteins that have the highest potential of inducing severe food anaphylaxis.     

Single-tube nested real-time PCR as a new highly sensitive approach to trace hazelnut

Hazelnut is one of the most commonly consumed tree nuts, being largely used by the food industry in a wide variety of processed foods. However, it is a source of allergens capable of inducing mild to severe allergic reactions in sensitized individuals. Hence, the development of highly sensitive methodologies for hazelnut traceability is essential. In this work, we developed a novel technique for hazelnut detection based on a single-tube nested real-time PCR system. The system presents high specificity and sensitivity, enabling a relative limit of detection of 50 mg/kg of hazelnut in wheat material and an absolute limit of detection of 0.5 pg of hazelnut DNA (1 DNA copy). Its application to processed food samples was successfully achieved, detecting trace amounts of hazelnut in chocolate down to 60 mg/kg. These results highlight the adequacy of the technique for the specific detection and semiquantitation of hazelnut as potential hidden allergens in foods.     

Sandwich immunoassays for the determination of peanut and hazelnut traces in foods

People suffering from food allergies are dependent on accurate food labeling, as an avoidance diet is the only effective countermeasure. Even a small amount of allergenic protein can trigger severe reactions in highly sensitized patients. Therefore, sensitive and reliable tests are needed to detect potential cross-contamination. In this paper two fast sandwich immunoassays are described for the determination of peanut (Arachis hypogaea) and hazelnut (Corylus avellana) traces in complex food matrices. Mouse monoclonal antibodies were used as capture antibodies, and labeled rabbit polyclonal antibodies were used as detection antibodies in both assays. The assay time was 30 min in total, and cross-reactivities against a variety of fruits and seeds were found to be in the low 10(-4)% (ppm) level or in some cases not detectable. The recoveries in all tested food matrices ranged from 86 to 127%, and the limits of detection were in the range of 0.2-1.2 mg/kg (ppm) in food for both peanut and hazelnut, respectively.     

Comparison of DNA extraction methods and development of duplex PCR and real-time PCR to detect tomato, carrot, and celery in food

Traceability is of particular importance for those persons who suffer allergy or intolerance to some food component(s) and need a strict avoidance of the allergenic food. In this paper, methodologies are described to fingerprint the presence of allergenic species such as carrot, tomato, and celery by DNA detection. Three DNA extraction methods were applied on vegetables and foods containing or not containing the allergens, and the results were compared and discussed. Fast SYBR Green DNA melting curve temperature analyses and duplex PCR assays with internal control have been developed for detection of these allergenic vegetables and have been tested on commercial foods. Spiking food experiments were also performed, assessing that limits of detection (LOD) of 1 mg/kg for carrot and tomato DNA and 10 mg/kg for celery DNA have been reached.     

Comparison of real-time PCR detection chemistries and cycling modes using Mon810 event-specific assays as model

The most widely accepted methods for accurate quantitative detection of genetically modified organisms rely on real-time PCR. Various detection chemistries are available for real-time PCR. They include sequence-unspecific DNA labeling dyes such SYBR-Green I and the use of both universal (e.g., AmpliFluor) and sequence-specific double-labeled probes, the latter comprising hybridization (e.g., Molecular Beacon) and hydrolysis (e.g., TaqMan or MGB) probes. Also, new real-time PCR devices and reagents allowing fast cycling reactions exist. Five Mon810 real-time PCR assays were developed in which the event specificity was based on the detection of transgene and plant rearranged sequences found to 3' flank the insertion site. Every assay was specifically designed for one particular detection chemistry, that is, AmpliFluor, Molecular Beacon, MGB, TaqMan, and SYBR-Green I. When possible, the assays were adapted to fast cycling mode. All assays displayed satisfactory performance parameters, although Molecular Beacon, MGB, and TaqMan chemistries were the most suitable for quantification purposes in both conventional and fast cycling modes.     

Reliable enzyme-linked immunosorbent assay for the determination of soybean proteins in processed foods

Among allergenic foods, soybean is known as a food causing adverse reactions in allergenic patients. To clarify the validity of labeling, the specific and sensitive detection method for the analysis of the soybean protein would be necessary. The p34 protein, originally characterized to be p34 as an oil-body associated protein in soybean, has been identified as one of the major allergenic proteins and named Gly m Bd 30K. A novel sandwich enzyme-linked immunosorbent assay (ELISA) for the detection and quantification of the soybean protein in processed foods was developed using polyclonal antibodies raised against p34 as a soybean marker protein and the specific extraction buffer for extract. The developed sandwich ELISA method was highly specific for the soybean protein. The limit of detection (LOD) and the limit of quantification (LOQ) of the developed ELISA were 0.47 ng/mL (equivalent to 0.19 microg/g in foods) and 0.94 ng/mL (equivalent to 0.38 microg/g in foods), respectively. The recovery ranged from 87.7 to 98.7%, whereas the intra- and interassay coefficients of variation were less than 4.2 and 7.5%, respectively. This study showed that the developed ELISA method is a specific, precise, and reliable tool for the quantitative analysis of the soybean protein in processed foods.     

Specific detection of potentially allergenic kiwifruit in foods using polymerase chain reaction

Kiwifruit (Actinidia deliciosa and Actinidia chinensis) is allergenic to sensitive patients, and, under Japanese regulations, it is one of the food items that are recommended to be declared on food labeling as much as possible. To develop PCR-based methods for the detection of trace amounts of kiwifruit in foods, two primer pairs targeting the ITS-1 region of the Actinidia spp. were designed using PCR simulation software. On the basis of the known distribution of a major kiwifruit allergen (actinidin) within the Actinidia spp., as well as of reports on clinical and immunological cross-reactivities, one of the primer pairs was designed to detect all Actinidia spp. and the other to detect commercially grown Actinidia spp. (i.e., kiwifruit, Actinidia arguta, and their interspecific hybrids) except for Actinidia polygama. The specificity of the developed methods using the designed primer pairs was verified by performing PCR experiments on 8 Actinidia spp. and 26 other plants including fruits. The methods were considered to be specific enough to yield target-size products only from the target Actinidia spp. and to detect no target-size products from nontarget species. The methods were sensitive enough to detect 5-50 fg of Actinidia spp. DNA spiked in 50 ng of salmon testis DNA used as a carrier (1-10 ppm of kiwifruit DNA) and 1700 ppm (w/w) of fresh kiwifruit puree spiked in a commercial plain yogurt (corresponding to ca. 10 ppm of kiwifruit protein). These methods would be expected to be useful in the detection of hidden kiwifruit and its related species in processed foods.     

Real-time polymerase chain reaction based assays for quantitative detection of barley, rice, sunflower, and wheat

Quality assurance is a major issue in the food industry. The authenticity of food ingredients and their traceability are required by consumers and authorities. Plant species such as barley (Hordeum vulgare), rice (Oryza sativa), sunflower (Helianthus annuus), and wheat (Triticum aestivum) are very common among the ingredients of many processed food products; therefore the development of specific assays for their specific detection and quantification are needed. Furthermore, the production and trade of genetically modified lines from an increasing number of plant species brings about the need for control within research, environmental risk assessment, labeling/legal, and consumers' information purposes. We report here the development of four independent real-time polymerase chain reaction (PCR) assays suitable for identification and quantification of four plant species (barley, rice, sunflower, and wheat). These assays target gamma-hordein, gos9, helianthinin, and acetyl-CoA carboxylase sequences, respectively, and were able to specifically detect and quantify DNA from the target plant species. In addition, the simultaneous amplification of RALyase allowed bread from durum wheat to be distinguished. Limits of detection were 1 genome copy for barley, sunflower, and wheat and 3.3 copies for rice real-time PCR systems, whereas limits of quantification were 10 genome copies for barley, sunflower, or wheat and approximately 100 haploid genomes for rice real-time PCR systems. Real-time PCR cycling conditions of the four assays were stated as standard to facilitate their use in routine laboratory analyses. The assays were finally adapted to conventional PCR for detection purposes, with the exception of the wheat assay, which detects rye simultaneously with similar sensitivity in an agarose gel.     

Specific and sensitive enzyme-linked immunosorbent assays for analysis of residual allergenic food proteins in commercial bottled wine fined with egg white, milk, and nongrape-derived tannins

Regulations introduced by the Food Standards Australia New Zealand in December 2002 require all wine and wine product labels in Australia to identify the presence of a processing aid, additive or other ingredient, which is known to be a potential allergen. The objective of this study was to establish sensitive assays to detect and measure allergenic proteins from commonly used processing aids in final bottled wine. Sensitive and specific enzyme-linked immunosorbent assays (ELISA) were developed and established for the proteins casein, ovalbumin, and peanut. Lower limits of detection of these proteins were 8, 1, and 8 ng/mL, respectively. A panel of 153 commercially available bottled Australian wines were tested by these ELISA, and except for two red wines known to contain added whole eggs, residuals of these food allergens were not detected in any wine. These findings are consistent with a lack of residual potentially allergenic egg-, milk-, or nut-derived processing aids in final bottled wine produced in Australia according to good manufacturing practice at a concentration that could cause an adverse reaction in egg, milk, or peanut/tree-nut allergic adult consumers.     

A rapeseed-specific gene, acetyl-CoA carboxylase, can be used as a reference for qualitative and real-time quantitative PCR detection of transgenes from mixed food samples

Polymerase chain reaction (PCR) methods are very useful techniques for the detection and quantification of genetically modified organisms (GMOs) in food samples. These methods rely on the amplification of transgenic sequences and quantification of the transgenic DNA by comparison to an amplified reference gene. Reported here is the development of specific primers for the rapeseed (Brassica napus) BnACCg8 gene and PCR cycling conditions suitable for the use of this sequence as an endogenous reference gene in both qualitative and quantitative PCR assays. Both methods were assayed with 20 different rapeseed varieties, and identical amplification products were obtained with all of them. No amplification products were observed when DNA samples from other Brassica species, Arabidopsis thaliana, maize, and soybean were used as templates, which demonstrates that this system is specific for rapeseed. In real-time quantitative PCR analysis, the detection limit was as low as 1.25 pg of DNA, which indicates that this method is suitable for use in processed food samples which contain very low copies of target DNA.     

Multivariate analysis for the evaluation of fiber, sugars, and organic acids in commercial presentations of table olives

Table olives constitute an important part of the Mediterranean diet and the diet of many non-olive-producing countries. The aim of this work was to determine the fiber, sugar, and organic acid contents in Spanish commercial presentations of table olives and characterize them by means of a multivariate analysis. The selection of variables was carried out on the basis of a canonical analysis and their classification, according to processing styles and cultivars, through a linear discriminant analysis. Values of dietary fiber in table olives ranged from 2 to 5 g/100 g edible portion (e.p.). Some stuffing materials (almond, hot red pepper, and hazelnut) or the addition of capers produced a significant increase in the total dietary fiber in green olives. Glucose, fructose, and mannitol were usually found in the ranges of 0-55, 0-70, and 0-107 mg/100 g e.p., respectively. Succinic acid was detected only in green and directly brined olives (0-40 mg/100 g e.p.), while lactic and acetic acids were used within the ranges of 0-681 and 5-492.8 mg/100 g e.p., respectively. A multivariate analysis showed that fiber, mannitol, and succinic, lactic, and acetic acids can be used to discriminate between processing styles (95.5% correct assignations) and cultivars (61.20%). Current data can also be used in the evaluation of the dietary value of table olives.     

Interlaboratory evaluation of a real-time multiplex polymerase chain reaction method for identification of salmon and trout species in commercial products

This interlaboratory study evaluated a real-time multiplex polymerase chain reaction (PCR) method for identification of salmon and trout species in a range of commercial products in North America. Eighty salmon and trout products were tested with this method by three independent laboratories. Samples were collected in the United States and Canada, and only the collecting institution was aware of the species declaration. Following analysis with real-time PCR, all three laboratories were able to identify species in 79 of the 80 products, with 100% agreement on species assignment. A low level of fraud was detected, with only four products (5%) found to be substituted or mixtures of two species. The results for two of the fraudulent products were confirmed with alternate methods, but the other two products were heavily processed and could not be verified with methods other than real-time PCR. Overall, the results of this study show the usefulness and versatility of this real-time PCR method for the identification of commercial salmon and trout species.     

Fast real-time PCR for the detection of crustacean allergen in foods

Crustaceans are one of the most common allergens causing severe food reaction. These food allergens are a health problem, and they have become very important; there are various regulations that establish that labeling must be present regarding these allergens to warn consumers. In the present work a fast real-time PCR, by a LNA probe, was developed. This allows the detection of crustaceans in all kinds of products, including processed products in which very aggressive treatments of temperature and pressure during the manufacturing process are used. This methodology provides greater sensitivity and specificity and reduces the analysis time of real-time PCR to 40 min. This methodology was further validated by means of simulating products likely to contain this allergen. For this, products present on the market were spiked with crustacean cooking water. The assay is a potential tool in issues related to the labeling of products and food security to protect the allergic consumer.     

Development of sensitive crop-specific polymerase chain reaction assays using 5S DNA: applications in food traceability

The 5S intergenic spacers were amplified using a common pair of primers and sequenced from four species (Brassica napus, Zea mays, Helianthus annuus, and Glycine max). Crop-specific assays were developed from primers designed from the spacers and tested to amplify corresponding DNAs in both conventional end-point and real-time polymerase chain reactions (PCRs). The high copy numbers of the 5S DNA in plants make it possible to detect very small amounts of DNA using this marker. This sensitivity made it possible to compare different DNA extraction methods for highly processed food products using 5S spacers, even allowing dilution of templates to overcome PCR inhibition.     

Validation of a tomato-specific gene, LAT52, used as an endogenous reference gene in qualitative and real-time quantitative PCR detection of transgenic tomatoes

Toward the development of reliable qualitative and quantitative Polymerase Chain Reaction (PCR) detection methods of transgenic tomatoes, one tomato (Lycopersicon esculentum) species specific gene, LAT52, was selected and validated as suitable for using as an endogenous reference gene in transgenic tomato PCR detection. Both qualitative and quantitative PCR methods were assayed with 16 different tomato varieties, and identical amplified products or fluorescent signals were obtained with all of them. No amplified products and fluorescent signals were observed when DNA samples from 20 different plants such as soybean, maize, rapeseed, rice, and Arabidopsis thaliana were used as templates. These results demonstrated that the amplified LAT52 DNA sequence was specific for tomato. Furthermore, results of Southern blot showed that the LAT52 gene was a single-copy gene in the different tested tomato cultivars. In qualitative and quantitative PCR analysis, the detection sensitivities were 0.05 and 0.005 ng of tomato genomic DNA, respectively. In addition, two real-time assays employing this gene as an endogenous reference gene were established, one for the quantification of processed food samples derived from nontransgenic tomatoes that contained degraded target DNA and the other for the quantification of the junction region of CaMV35s promoter and the anti-sense ethylene-forming enzyme (EFE) gene in transgenic tomato Huafan No. 1 samples. All of these results indicated that the LAT52 gene could be successfully used as a tomato endogenous reference gene in practical qualitative and quantitative detection of transgenic tomatoes, even for some processed foods derived from transgenic and nontransgenic tomatoes.