Allergy to Prebiotic Galacto-Oligosaccharides: House Dust Mites-the Putative Primary Sensitizer.

Galacto-oligosaccharides (GOS) have been increasingly used globally to supplement commercially available milk formulations and dairy products for their health benefits.1Vandenplas Y. Zakharova I. Dmitrieva Y. Oligosaccharides in infant formula: more evidence to validate the role of prebiotics.Br J Nutr. 2015; 113: 1339-1344Crossref PubMed Scopus (115) Google Scholar Intriguingly, when GOS were introduced into Southeast Asian countries as an ingredient in milk formula, a few acute allergic reactions and anaphylactic responses to GOS were promptly observed.2Vo T.H. Le N.H. Patel M.S. Phan L.T. Tran Minh N.N. Acute allergic reactions in Vietnamese children after drinking a new milk product.Foodborne Pathog Dis. 2012; 9: 156-159Crossref PubMed Scopus (16) Google Scholar, 3Chiang W.C. Huang C.H. Llanora G.V. Gerez I. Goh S.H. Shek L.P. et al.Anaphylaxis to cow's milk formula containing short-chain galacto-oligosaccharide.J Allergy Clin Immunol. 2012; 130: 1361-1367Abstract Full Text Full Text PDF PubMed Scopus (28) Google Scholar, 4Soh J.Y. Chiang W.C. Huang C.H. Woo C.K. Ibrahim I. Heng K. et al.An unusual cause of food-induced anaphylaxis in mothers.World Allergy Organ J. 2017; 10: 3Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar Anaphylaxis to GOS occurred on first consumption in all our patients with allergic reactions, which was an indication of the presence of pre-existing GOS sensitization. GOS allergy is unique because the allergen is a pure carbohydrate. Vivinal GOS (vGOS) (FrieslandCampina, Amershfoort, The Netherlands) is produced with a β-galactosidase enzyme derived from Bacillus circulans. The composition of GOS has been shown to contain mostly β-1,4 linkages between glucose and galactose or between galactoses.5Coulier L. Timmermans J. Bas R. Van Den Dool R. Haaksman I. Klarenbeek B. et al.In-depth characterization of prebiotic galacto-oligosaccharides by a combination of analytical techniques.J Agric Food Chem. 2009; 57: 8488-8495Crossref PubMed Scopus (102) Google Scholar However, β-1,6 and β-1,3 linkages are also found in vGOS, although to a lesser extent. GOS does not contain any fucose or sialic acid moieties. In view of the geographic restriction of GOS allergy to Southeast Asian countries when GOS is used to supplement milk formulas globally, we postulated that GOS allergy might be a result of sensitization to an agent that is likely geographically confined to Southeast Asia. We sought to identify the primary sensitizer for GOS allergy. The clinical symptoms/reactions, skin prick test responses, and basophil activation test results to vGOS in 33 patients with GOS allergy, including 23 subjects who reacted to vGOS after consumption of vGOS-containing milk or water and 10 GOS-sensitized subjects who did not consume GOS before testing, are shown in Table I. Sera from these subjects were tested to determine the sensitization profile to 54 allergens (see Table E1 in this article’s Online Repository at www.jacionline.org for the list of allergens) by using the EUROLINE multiplex immunoblot assay (EUROIMMUNE, Lübeck, Germany) and to Curvularia lunata and Aedes communis by using the ImmunoCAP assay (Phadia, Uppsala, Sweden). This study was approved by the ethics review boards of both National University Hospital and Kandang Kerbau Women’s and Children’s Hospital, and written informed consent was obtained from all participants.Table IClinical and immunologic features of GOS-sensitized subjectsPatient no.∗Subjects 1 to 17 had GOS allergy, subjects 19 to 23 had allergic symptoms on vGOS oral challenge, and subjects 24 to 33 were sensitized to GOS but did not undergo oral GOS challenge. The 5 subjects described in Fig 1 are indicated inside parentheses.Age of onset (years)Dose eliciting reaction (g)SymptomsSPT to vGOS (mm)BAT threshold dose†The lowest dose that induced both upregulation of CD203c and expression of CD63 on basophils was used. (μg/mL)Specific IgE‡GOS-specific IgE levels were determined by means of ELISA. An sIgE level of 1.14 ng/mL or greater was considered positive. Levels of B tropicalis– and D pteronyssinus–specific IgE were determined by using EUROLINE. An sIgE level of 0.35 kU/L or greater was considered positive.GOS (ng/mL)B tropicalis (kU/L)D pteronyssinus (kU/L)Cow’s milk (kU/L)Shellfish mix (kU/L)160.2RS, Resp3 × 30.1103.6941001.7<0.35271.9AE, RS, Resp7 × 6141.566901.70.353 (1)381.5AE, RS, Resp5 × 4102.5291.4<0.35<0.35461.9AE, GU, GI, RS, Resp5 × 51<1.14100870.5218580.4AE, GU, RS, Resp3 × 31<1.1486450.3517.4621.1AE, GU, RS, Resp5 × 41<1.140.430.75.50.357302AE, LUNDND<1.142.7<0.350.4362861.9AE, GU, GI, Resp4 × 419.418<0.351<0.35931.6AE, LU, RS, Resp3 × 71<1.145.59.511.4681040UnknownAE, GU, RS, Resp5 × 61<1.14925.5<0.35<0.3511321.5AE, RS, Resp5 × 511.80.60.43<0.351.412291.9AE, GU, RS5 × 50.1<1.141.70.35<0.35<0.3513331.5-1.9AE, RS, Resp5 × 5101.210.43<0.352014260.7AE, RS, Resp5 × 5102.140370.43<0.3515100.9AE, RS, Resp5 × 5ND1.58.57.50.432716131.5AE, LU, GI, Resp10 × 80.1<1.142.4720.511720.5AE, LU, RS, Resp8 × 6101.716.45913.419372RS, Resp6 × 610<1.147.512.4<0.35<0.3518302AE, GU, RS4 × 51<1.14812.40.35<0.3520 (2)402AE, RS5 × 6101.8615.5<0.350.4321 (3)270.6AE, RS, Resp8 × 50.11.6402.4<0.350.3522 (4)432Resp5 × 8>1000§The subject had a negative BAT response to the highest dose (1000 μg/mL) of GOS used in the BAT assay.2.7422.4<0.35<0.3523224LU, GI, RS3 × 311.38448<0.35<0.3524NANANC3 × 35001.372570.35<0.3525NANANC3 × 310<1.1481360.43<0.3526NANANC6 × 511.453452.76.527NANANC4 × 411.510092<0.35<0.3528NANANC8 × 5#Results of SPT to 14.4 mg/mL of vGOS.>1000§The subject had a negative BAT response to the highest dose (1000 μg/mL) of GOS used in the BAT assay.<1.14591000.43<0.3529NANANC5 × 410<1.140.65.50.43<0.3530NANANC3 × 310<1.144.574<0.350.4331 (5)NANANC4 × 3112.97035<0.350.5232NANANC4 × 61084.185950.3515.433NANANC5 × 5101.1413.429<0.35<0.35AE, Angioedema; BAT, basophil activation test; GI, vomiting; GU, generalized urticaria; LU, localized urticaria; NA, not applicable; NC, not challenged; ND, not done; Resp, wheeze, dry persistent cough, or dyspnea; RS, rhinorrhea and bouts of sneezing; SPT, skin prick test.∗ Subjects 1 to 17 had GOS allergy, subjects 19 to 23 had allergic symptoms on vGOS oral challenge, and subjects 24 to 33 were sensitized to GOS but did not undergo oral GOS challenge. The 5 subjects described in Fig 1 are indicated inside parentheses.† The lowest dose that induced both upregulation of CD203c and expression of CD63 on basophils was used.‡ GOS-specific IgE levels were determined by means of ELISA. An sIgE level of 1.14 ng/mL or greater was considered positive. Levels of B tropicalis– and D pteronyssinus–specific IgE were determined by using EUROLINE. An sIgE level of 0.35 kU/L or greater was considered positive.§ The subject had a negative BAT response to the highest dose (1000 μg/mL) of GOS used in the BAT assay.# Results of SPT to 14.4 mg/mL of vGOS. Open table in a new tab AE, Angioedema; BAT, basophil activation test; GI, vomiting; GU, generalized urticaria; LU, localized urticaria; NA, not applicable; NC, not challenged; ND, not done; Resp, wheeze, dry persistent cough, or dyspnea; RS, rhinorrhea and bouts of sneezing; SPT, skin prick test. Of the 56 allergens tested, the common group of allergens to which all patients with GOS allergy were sensitized was dust mites. All patients with GOS allergy had specific IgE to Blomia tropicalis, Dermatophagoides farinae, and Dermatophagoides microceras. The sensitization rates to Tyrophagus putrescentiae and Dermatophagoides pteronyssinus were 97% and 94%, respectively, in patients with GOS allergy. The sensitization rates to D pteronyssinus, D farinae, D microceras, and T putrescentiae ranged from 79% to 83% in atopic control subjects. Atopic control subjects were atopic subjects (n = 24) with a positive skin prick test response to B tropicalis but not to GOS. Sensitization rates to the rest of allergens were all less than 50%, except for cow’s milk and shellfish mix 1, with approximately 55% of both patients with GOS allergy and control subjects reacting to these allergens. B tropicalis and D pteronyssinus are 2 common mite species found in Singapore. Five serum samples with high levels of GOS-specific IgE (Table I and see Fig E1 in this article’s Online Repository at www.jacionline.org) and available serum were selected for determining the IgE cross-reactivity between GOS and B tropicalis/D pteronyssinus by using a competitive ELISA. A control extract, Curvularia species, which showed low IgE sensitization rates (3%) in patients with GOS allergy, was included. Both dust mite extracts inhibited GOS-specific IgE in a dose-dependent manner for all 5 subjects except for one subject in whom there was no inhibition observed with D pteronyssinus extract (Fig 1, A). B tropicalis extract was a very potent inhibitor because the concentration of B tropicalis extract needed to inhibit 50% of GOS-specific IgE (IC50) was 11 times less than that of GOS itself (86 and 954 ng/mL, average IC50 values of 5 subjects for B tropicalis and GOS, respectively). The IC50 for D pteronyssinus extract (1032 ng/mL, average IC50 values of 4 subjects) was comparable with that for GOS. On the other hand, Curvularia species extract showed minor ability (approximately 40%) to inhibit GOS-specific IgE only at high concentrations (30 μg/mL; Fig 1). IgE cross-reactivity between mite culture medium and GOS was measured. The results of preliminary experiments make it unlikely that the mite culture medium contributes to GOS cross-reactivity of Blomia species extract (data not shown). B tropicalis instead of D pteronyssinus was selected as the most relevant allergen for further evaluation of IgE cross-reactivity with GOS. This mite is highly prevalent in tropical regions and could explain the geographic restriction of GOS allergy to Southeast Asia. Another reason that Dermatophagoides species might not be the primary sensitizer for GOS allergy is the global importance of D pteronyssinus and D farinae allergens. Dermatophagoides species sensitization is common in Europe, where GOS-supplemented formula and dairy products have been widely available in the market for years, and yet there have not been any reports of GOS-related allergy in that region. Several dust mite allergens were shown to be glycosylated.6Al-Ghouleh A. Johal R. Sharquie I.K. Emara M. Harrington H. Shakib F. et al.The glycosylation pattern of common allergens: the recognition and uptake of Der p 1 by epithelial and dendritic cells is carbohydrate dependent.PLoS One. 2012; 7: e33929Crossref PubMed Scopus (63) Google Scholar,7Halim A. Carlsson M.C. Madsen C.B. Brand S. Moller S.R. Olsen C.E. et al.Glycoproteomic analysis of seven major allergenic proteins reveals novel post-translational modifications.Mol Cell Proteomics. 2015; 14: 191-204Crossref PubMed Scopus (28) Google Scholar GOS and B tropicalis extracts were subjected to deglycosylation by means of periodate oxidation. The epitope for GOS-specific IgE was sensitive to periodate oxidation because periodate-treated GOS lost its ability to inhibit GOS-specific IgE (see Fig E2, A, in this article’s Online Repository at www.jacionline.org). There was a decrease in GOS-specific IgE inhibition from 80% to 100% in untreated extracts to 0% to 30% in periodate-treated, deglycosylated B tropicalis extracts (Fig 1, B). Periodate treatment did not significantly affect the protein-specific IgE epitopes. Periodate-treated B tropicalis extract was still able to inhibit more than 75% of B tropicalis–specific IgE (Fig 1, B). We went on to validate our findings by treating B tropicalis extract with Proteinase K (New England Biolabs, Ipswich, Mass) and trypsin. The proteolysed B tropicalis extract showed 50% to 95% reduction in its ability to inhibit B tropicalis–specific IgE. In contrast, proteolysed B tropicalis extract still retained its ability to inhibit GOS-specific IgE, which was comparable with the untreated B tropicalis extract (>97%; see Fig E2, B). Taken together, these results indicate that protein-specific IgE epitopes in the B tropicalis extract do not play a role in inhibition of GOS-specific IgE. The carbohydrate moieties of the B tropicalis allergens were responsible for the IgE cross-reactivity between GOS and B tropicalis. The presence of a primary sensitizer for carbohydrate-induced anaphylaxis has been demonstrated in 2 other studies: (1) α-gal allergy, in which the deer tick was a primary sensitizer in the United States,8Commins S.P. James H.R. Kelly L.A. Pochan S.L. Workman L.J. Perzanowski M.S. et al.The relevance of tick bites to the production of IgE antibodies to the mammalian oligosaccharide galactose-alpha-1,3-galactose.J Allergy Clin Immunol. 2011; 127: 1286-1293.e6Abstract Full Text Full Text PDF PubMed Scopus (411) Google Scholar and (2) allergy to GOS-supplemented beverages in a group of oyster shuckers having sea squirt allergy in Japan.9Jyo T. Kuwabara M. Kodommari Y. Tanemori N. Asaoku Y. Katsutani T. et al.Cases of immediate type allergy in oyster shuckers due to galactooligosaccharide.Med J Hiroshima Prefectural Hospital. 1993; 25: 19-27Google Scholar Antibody cross-reactivity between the crude extract of sea squirt body fluid and the GOS was observed. In this study we have identified B tropicalis as a potential primary sensitizer for allergy to GOS in Singaporean subjects and determined that the cross-reactive IgE epitope or epitopes are carbohydrates and not proteins. Based on our small study, this is the first time that B tropicalis contains glycoallergens that have IgE cross-reactive epitopes specific to GOS. We thank Ms Corinne Kwek for her assistance with the subjects, the subjects for their participation, and Linqiu Cao and Mireille Gadella (FrieslandCampina, Amersfoort, The Netherlands) for providing the enriched DP4-GOS sample. Skin prick tests and basophil activation tests to vGOS were conducted, as described previously.E1Chiang W.C. Huang C.H. Llanora G.V. Gerez I. Goh S.H. Shek L.P. et al.Anaphylaxis to cow's milk formula containing short-chain galacto-oligosaccharide.J Allergy Clin Immunol. 2012; 130: 1361-1367Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar The wheal size for each allergen was recorded and used as the degree of skin test reactivity. One hundred microliters of blood was stimulated with up to 1000 μg/mL vGOS. The threshold dose was recorded as the lowest dose that triggered both the expression of CD63 and CD203c above the cutoff value of a positive basophil activation test result, as determined in our previous study.E2Soh J.Y. Huang C.H. Chiang W.C. Llanora G.V. Lee A.J. Loh W. et al.Anaphylaxis to galacto-oligosaccharides—an evaluation in an atopic population in Singapore.Allergy. 2015; 70: 1020-1023Crossref PubMed Scopus (8) Google Scholar Levels of specific IgE to a panel of 54 allergens and cross-reactive carbohydrate determinants were measured by using the Multiplex Immunoblot (EUROLINE “Atopy Indonesia 1”; DP 3707-1601-1 E; EUROIMMUNE). Multiplex immunoblots was performed according to the manufacturer’s instructions. Briefly, test strips were incubated with subjects’ sera using the volume-optimized version, as described in the manufacturer’s instructions. Results were analyzed with EUROLineScan (EUROIMMUN) software. In addition, levels of specific IgE to A communis and C lunata were measured by using the ImmunoCAP assay, which was run with the Phadia 100 system (Phadia AB, Uppsala, Sweden). In both assays results were considered positive when the level of specific IgE to a particular allergen measured was greater than 0.35 kU/L. Cultured D pteronyssinus was purchased from Mahidol University (Bangkok, Thailand). B tropicalis mites were obtained in house, as described previously.E3Yi F.C. Chew F.T. Jimenez S. Chua K.Y. Lee B.W. Culture of Blomia tropicalis and IgE immunoblot characterization of its allergenicity.Asian Pac J Allergy Immunol. 1999; 17: 189-194PubMed Google Scholar In brief, the B tropicalis mite was cultured with TetraMin fish feed (Tetra, Blacksburg, Va). The mite culture was separated through a series of 500- and 125-μm sieves by using a mechanical sieve shaker in which vibration was applied for about 20 minutes to remove most of the culture media. Mites with sizes of greater than 120 μm were transferred to a modified Tullgren funnel, in which mites were harvested by using the heat-escape method, which collects living mites escaping from cultures exposed to low heat. A mixture of spores from Curvularia species was a kind gift from Associate Professor Chew Fook Tim (Department of Biological Science, National University of Singapore). Mites and spores were kept at −80°C until use. Frozen mites were homogenized with a mortar and pestle in the presence of liquid nitrogen. PBS containing 0.1% aprotinin (Sigma-Aldrich, St Louis, Mo), 1 mmol/L phenylmethylsulfonyl fluoride (Sigma-Aldrich), 15 μmol/L trans-epoxysuccinyl-L-leucylamido(4-guanidino)butane (Sigma-Aldrich), and 10 μmol/L Pepstatin A (Sigma-Aldrich) was used for protein extraction at 4°C overnight. The supernatant was obtained after centrifugation at 14,000g for 10 minutes and kept at −80°C until use. The protein extract of Curvularia species was obtained by using a similar method, except the extraction buffer was changed to 12.5 mmol/L ammonium bicarbonate. After centrifugation, the supernatant was lyophilized and reconstituted in PBS. Total protein concentrations were determined by using a Bio-Rad Protein Assay Kit (Bio-Rad Laboratories, Hercules, Calif). To synthesize a GOS-fraction enriched in allergenicity, an enzymatic reaction was started with 19% lactose wt/wt (Lactochem; FrieslandCampina, Amersfoort, The Netherlands) and a lactase (Lactoles L3; Amano, Yokohama, Japan) dosage of 50 LU/g lactose at 50°C for 24 hours while stirring with a magnet bar. The obtained GOS solution was loaded to an active carbon column (PN2 type; Norit NV, Amersfoort, The Netherlands) and subsequently eluted with 5 bed volumes of water to remove nonadsorbed components, 5 bed volumes of 5% ethanol aqueous solution (Sigma), and finally 5 bed volumes of 30% ethanol aqueous solution. After evaporation of water and ethanol from the 30% ethanol fraction, water was added to reach a concentration of greater than 45% (wt/wt). Subsequently, this concentrated fraction was further refractioned by using a Bio-Gel Column (Bio-Rad Laboratories). The degree of polymerization 4 (DP4) fraction was collected, pooled, and further concentrated to approximately 20% solution (wt/wt) by using a Rotavap (Rotary vacuum evaporation; Salm en Kipp, Breukelen, The Netherlands). The sample was stored at −20°C until use. The dose required to induce basophil activation for this enriched DP4 fraction is 3 times lower than that of the DP4 of vGOS (data not shown). The DP4-GOS fraction was conjugated to human serum albumin (HSA; Sigma-Aldrich) by means of the Maillard reaction to bind the enriched DP4-GOS fraction to the solid phase. In brief, the DP4-GOS fraction and HSA were dissolved in distilled water and then mixed together at a 1:1 (wt/wt) ratio. The mixture was lyophilized. The Maillard reaction was conducted by placing the lyophilized sample in an airtight container containing saturated sodium chloride solution to maintain relative humility at about 75%. The container was put in an incubator at 50°C for 9 days. The sample (HSA-GOS) was then reconstituted in PBS. Specific IgE to GOS was measured by means of ELISA. All the ELISA procedures with 25 μL/well were carried out at room temperature for 1 hour unless specified. Half-area ELISA plates (Corning, Corning, NY) were coated with 5 μg/mL HSA-GOS extract in 0.1 mol/L NaHCO3, pH 8.3. After washing with Tris-buffered saline containing 0.05% Tween-20 (TBS-T; Sigma-Aldrich), plates were blocked with 100 μL of 1% BSA (Sigma-Aldrich) in TBS-T. Sera were incubated overnight with serially diluted B tropicalis, D pteronyssinus, and Curvularia species extracts or GOS together with 50 μg/mL HSA at 4°C. Preabsorbed sera were incubated overnight with HSA-GOS–coated wells in duplicates at 4°C. Plates were washed and incubated with horseradish peroxidase–conjugated anti-human IgE (B3102E8; Southern Biotechnology, Birmingham, Ala) at room temperature for 1 hour. Signals were developed with 3, 3′, 5, 5′-tetramethylbenzidine (Invitrogen, Vienna, Austria), and reactions were stopped with the addition of 12.5 μL of 1 mol/L sulfuric acid. Absorbance of 450/570 nm was determined by using the Synergy 2 Multi-Detection Microplate Reader (BioTek, Winooski, Vt). For semiquantification purposes, a standard curve was generated by adding serially diluted semipurified human IgE (Chemicon International, Temecula, Calif) to wells coated with 2 μg/mL anti-human IgE. GOS-specific IgE titers were extrapolated from standard curves. The percentage of inhibition was calculated as follows:(1-IgE1evelwithinhibitor/IgElevelwithoutinhibitor)×100 In some experiments sera were preincubated with deglycosylated or proteolysed B tropicalis extract. GOS-specific IgE or B tropicalis–specific IgE levels were measured thereafter. The lyophilized B tropicalis extract (100 μg) was reconstituted in 250 μL of 0.1 mol/L sodium acetate, pH 5.5. Deglycosylation was performed by adding 250 μL of 20 mmol/L Sodium (meta)periodate (Sigma-Aldrich) and incubated at room temperature for 30 minutes in the dark. The deglycosylated B tropicalis extract was then dialyzed against PBS. The total protein concentration was subsequently determined by using a Bio-Rad Protein Assay Kit (Bio-Rad Laboratories). The lyophilized B tropicalis extract (100 μg) was reconstituted in 25 μL of 50 mmol/L Tris buffer (pH 8.0) with 0.1% SDS and 20 mmol/L dithiothreitol and heated at 95°C for 10 minutes. The denatured extract was first treated with 5 μg of trypsin (Pierce Biotechnology, Rockford, Ill) at 37°C for 24 hours and subsequently treated with 20 μg of Proteinase K at 50°C for 1 hour. The sample was subjected to heat inactivation at 95°C for 10 minutes after each digestion.Fig E2GOS-specific IgE is sensitive to periodate oxidation. Sera of patients with GOS allergy (n = 5) were preabsorbed with HSA-GOS conjugates or periodate-treated HSA-GOS (A) or B tropicalis extract or proteolysed B tropicalis (Blo t) extract (B). Levels of B tropicalis–specific IgE and GOS-specific IgE were detected. The ability of HSA-GOS, periodate-treated HSA-GOS, B tropicalis extract, or proteolysed B tropicalis extract in inhibition of GOS sIgE or B tropicalis–specific IgE (as indicated) compared with sera without preabsorption was presented as a percentage of inhibition.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Table E1List of allergens in EUROLINE Atopy Indonesia 1 panelAeroallergens/venomMitePollenMoldAnimal danderInsect•Dermatophagoides pteronyssinus•Dermatophagoides farinae•Dermatophagoides microceras•Bermuda grass•Timothy grass•Grass mix 5 (Sweet vernal grass, Bermuda grass, Timothy grass, cultivated rye)•Mold mix 1 (Penicillium notatum, Cladosporium herbarum, Aspergillus fumigatus, Alternaria alternata)•Cat•Dog•Horse•Honeybee venom•Tyrophagus putrescentiae•Glycyphagus domesticus•Blomia tropicalis•Acacia•White pine (Australian)•Oil palm•Kapok•Candida albicans•Feather mix 1 (chicken feathers, duck feathers, goose feathers)•Cockroach (German)Food allergensEgg/milk/nutFish/shellfishMeatOthers•Cow’s milk•nBos d 4•nBos d 5•nBos d 6•Codfish•Tuna•Salmon•Duck meat•Beef (cooked)•Tomato•Garlic•Strawberry•Kiwi•nBos d 8•Egg white•Cheddar cheese•Crab•Shrimp (Pacific)•Lobster•Pork (cooked)•Chicken•Coffee•Wheat flour•Gluten•Chocolate•Peanut•Hazelnut•Almond•Shellfish mix 1 (spiny lobster, oyster, clam)•Lamb•Soybean•Baker’s yeast•Glutamate Open table in a new tab