Low levels of gluten and major milk allergens Bos d 5 and Bos d 11 identified in commercially available honey

Abstract

To the Editor, Despite allergy and allergic reactions to honey being widely regarded as rare, there have been documented systemic allergic reactions following ingestion of honey.1 Current literature suggests that pollens and components derived from bees are the main cause of such reactions.1,2 However, interestingly and perhaps unknown to many allergic patients and allergists, there are also reports of supplementary bee feeding with food allergenloaded mixtures of soybean flour, dried brewer's yeast (containing high levels of residual gluten from brewing processes) and dry skimmed milk with sugar and water.3 Furthermore, there have been reports of mould contamination within beehives.4 Both factors suggest a potential for gluten, food and mould allergen protein presence in honey, which could account for some of the reported reactions following honey consumption. As such, the aim of this study was to determine if commercially available honey contained undeclared gluten and/or food or mould allergens, and at levels which could present a risk to individuals with hypersensitivities. To investigate this, honey samples (n = 40) of UK, EU, NonEU and blended NonEU/EU origin were extracted and analysed for gluten using the Neogen Veratox Gliadin R5 Gluten ELISA, which is regarded as the gold standard for gluten measurements in the food industry. Major allergen content was measured using InBio MARIA and MARIA for Foods quantitative multiplex arrays for cow's milk, egg, peanut, soy, hazelnut, cashew and mould allergens. The MARIA immunoassay is based on xMAP® technology (Luminex Corp.) which uses polystyrene or magnetic microspheres that are internally labelled to create distinct sets of beads. Separate bead sets are covalently coupled with allergenspecific monoclonal antibodies, enabling the simultaneous capture and detection of multiple allergens in a single sample.5 Gluten (gliadin) assays were conducted according to manufacturer instruction. Honey samples were extracted by transferring 250 mg of honey to a 50ml sterile centrifuge tube to which 2.5 ml of renaturing cocktail solution (Neogen 8515, 8515B, 8515S) was added. The resultant suspension was vortex mixed for 30 s and incubated in a water bath at 50°C for 40 min. Samples were then cooled for 10 min at room temperature (RT), and 80% v/v ETOH was added. Samples were mixed as previously described for 20 s and then rotated at 200 rpm for 60 min before 100 μl of the resultant solution was added to 1.25 ml of phosphate buffered saline (PBS). A negative honey control was employed during testing for the presence of Gliadin R 5, this was produced on the day of testing and consisted of; 28 g glucose, 14 g fructose and 8 g of Sterile distilled water. In all tests, the negative control did not produce a result above the assay limit of detection of 2.5 parts per million (ppm) of gliadin, or 5 ppm of gluten. Analysis was repeated on two separate occasions and results are an average of these two measurements. Approximately, 50% of gluten is available as gliadin. Therefore, results for gliadin were multiplied by two to determine the levels of gluten. Two MARIA arrays were used for sample analysis. A MARIA for foods multiplex array allowed the simultaneous quantification of Peanut (Ara h 3, Ara h 6), Cow's Milk (Bos d 5, Bos d 11), Egg (Gal d 1, Gal d 2), Cashew (Ana o 3), Hazelnut (Cor a 9) and Soy (Gly m 5) allergen. A second MARIA multiplex array allowed for the simultaneous quantification of mould (Asp f 1, Alt a 1) allergens. The arrays use highly purified allergen standards to quantify specific allergen proteins from samples. Assays were carried out as described by Filep and Chapman.5 Prior to analysis, samples (1 g) were extracted in 20 ml PBS 0.05% Tween20, pH 7.4. Samples were briefly vortexed and incubated on a rocking platform for 120 min at RT. The resulting extracts were stored at −20°C prior to analysis. Positive results were confirmed by repeat extractions (n = 4– 5 samples per honey) with results presented as an average of replicate extractions. Sample extracts were analysed in twofold dilution series, ranging from neat up to 1:80. Recovery of allergen proteins from honey was verified through production of an incurred honey sample (as described in online Open Science Framework repository: https://osf.io/vd28j/, 10.17605/ OSF.IO/VD28J). Of the 40 samples analysed, it was observed that 8 of the 40 (20%) samples contained gluten in the range of 5 ppm to 13.8 ppm. Positive honey samples categorized by origin are detailed in Table 1. Of the 21 nonEU honey samples analysed, six were positive for gluten. From the nine UK and nine EU/nonEU blend honey samples, each had one sample positive for gluten. This represents a glutenpositive sample rate of 28.6% and 11.1% for nonEU and both UK and EU/nonEU blends respectively. Milk allergens Bos d 5 and Bos d 11 were detected in 7.5% of samples. The positive results ranged from 0.368 ppm (mg/kg), up to 0.567 ppm for Bos d 5, and from 0.030 ppm up to 0.182 ppm for