There have been more than 20 studies describing allergy to soybean (Besler et al. 2000 [660]). In general individuals report cutaneous, gastrointestinal and respiratory symptoms and some studies report specifically oral allergy syndrome (Kleine-Tebbe et al. 2002 [658]). Symptoms can be severe, requiring emergency treatment (Kleine-Tebbe et al. 2002 [658]) and there is one report of exercise induced anaphylaxis to soy (Taramarcaz et al. 2001 [675]). The types of symptom reported by selected representative studies are summarised below.

1. Sampson & Ho (1997) [652] reported cutaneous (68%), gastrointestinal (77%) and respiratory (23%) symptoms in 31 soybean-allergic children.

2. Sampson (2001) [651] similarly reported cutaneous (67%), gastrointestinal (62%) and respiratory (24%) symptoms in 21 soybean-allergic children.

3. Niggemann et al. (1999) [661] reported cutaneous (58%), gastrointestinal (14%) and both cutaneous and gastrointestinal symptoms (28%) in 7 soybean-allergic children.

4. Kleine-Tebbe et al. (2002) [658] reported on symptoms from 20 patients eating a soy containing food supplement. 17/20 experienced facial swelling, 14/20 oral allergy syndrome, 11/20 throat tightness and/or swallowing discomfort, 6/20 dyspnea (chest tightness, wheezing), 6/20 hives, urticaria, 5/20 drowsiness and/or vascular dysregulation, 4/20 gastrointestinal discomfort and/or vomiting, 2/20 nasal secretion and/or nasal congestion. 13 patients required emergency treatment.

5. Mittag et al. (2004) [857] described 22 patients with oral allergy symptoms. 12 had additional symptoms: 5 had urticaria, one angioedema, one blistering of the mouth, 3 tightness of throat or chest and one each with dysnpea or oedema of the larynx, 3 experienced nausea, 1 diarrhoea, and 2 showed a fall of blood pressure.

The positive control was histamine (10mg/ml). Several criteria were compared with the positive is mean diameter greater than 3 mm rule. The definition of negative used was the mean diameter or the surface area was less than the upper 95% confidence interval of subjects with negative DBPCFC (Eigenmann and Sampson, 1998 [674]).

Gu et al. (2001) [654] and Mittag et al. (2004) [857] defined positive as a wheal of at least 3 mm diameter.

Eigenmann and Sampson (1998) [674] tested 51 patients by SPT with soy extract (DBPCFC subsequently found 25 positive and 26 negative).

Gu et al. (2001) [654] tested 6 soy-sensitive patients and 2 nonsensitive individuals.

Mittag et al. (2004) [857] tested 17 soy allergic patients.

Eigenmann & Sampson (1998) [674] conclude that skin prick tests are a useful procedure for evaluating clinical reactivity to egg, milk, peanut and wheat, but not to soy. For soy, 13/26 negative DBPCFC were SPT positive and 9/25 positive DBPCFC were SPT negative.

Gu et al. (2001) [654] report that all 6 soy sensitive patients gave positive SPT to soy extract and 2 nonsensitive indivuals were SPT negative.

Mittag et al. (2004) [857] reported that 13/17 soy allergic patients gave clear positive SPTs and 2 gave borderline positive SPTs.

CAP FEIA (Sampson 2001 [651]; Mittag et al. (2004) [857]; Rihs et al. (1999) [170])

Earlier articles generally used RAST.

Mittag et al. (2004) [857] used sera from 22 soy allergic patients and from 94 birch allergic patients (9 of whom reported symptoms with soy). 16 were DBPCFC positive, 1 had suffered anaphylaxis and 5 declined the challenge.

Sampson (2001) [651] used sera from 100 patients of whom 25 were judged soy allergic, 21 by challenge and 4 by clinical history and IgE level.

Gu et al. (2001) [654] used sera from 6 soy-sensitive individuals and 2 soy-insensitive individuals.

Rihs et al. (1999) [170] used sera from 13 patients (10 female and 3 male patients; age range, 6-57 years) who were CAP positive to soy proteins. All were highly atopic and had atopic dermatitis. Food-associated symptoms included atopic eczema, anaphylaxis, and/or asthma

Awazuhara et al. 1997 [1091] and Awazuhara et al. 1998 [1080] used sera from 30 soy-allergic patients of whom 7 were positive to oral challenge.

Eigenmann et al. (1996) [48] used sera from 5 peanut allergic patients.

Ogawa et al. (1991) [441], Ogawa et al. (1993) [131] and Ogawa et al. (1995) [653] used sera from 69 patients with atopic dermatitis.

Herian et al. (1990) [1078] used sera from 7 patients allergic to soybean of whom 6 were allergic to other legumes (4 to peanut).

Pedersen & Djurtoft (1989) [1093] used sera from 10 atopic patients with a history of reactions to soya (and some to other legumes) and a positive RAST to soy extract.

Mittag et al. (2004) [857] found that 10/22 soy allergic patients gave a positive CAP test with a commercial extract (0.35 - 5.05 kU/l) but 21/22 were positive with purified Gly m 4 (0.72->100 kU/l). All 22 showed positive reaction to Bet v 1 (only 4 to Bet v 2 and 1 to Bet v 6). 67/94 birch allergic patients showed specific IgE against Gly m 4.
Birch pollen significantly inhibited IgE binding to soy protein isolate by EAST inhibition in all tested sera, with a minimum of 45% in one patient and maximum inhibitions of more than 80% in 9 of 11 patients. The maximum inhibition of IgE binding by rGly m 4 ranged between 36% and 100%, with 9 of 11 sera showing at least 60% inhibition.

CAP IgE of 65 kU_A/L confidently predicted allergy to soy, and above 30 kU_A/L suggested soy allergy (Sampson 2001 [651]).

Gu et al. (2001) [654] report that 6 sera were positive for soy specific IgE (0.79-54.1 kU/l) and the two sera from nonsensitive individuals were negative (>0.1 kU/l).

Pedersen & Djurtoft (1989) [1093] reported that IgE from all the 10 sera bound subunits of glycinin and 9/10 bound native glycinin by ELISA. Most sera bound the subunits more strongly.

Besler et al. 2000 [660]) review 22 EAST or immunoblotting studies and details of only some recent studies are described below.

Mittag et al. (2004) [857] used 1D SDS-PAGE with 15% separation gels and 5% stacking gels. Samples were reduced by heating with 1,4-dithiothreitol and loaded onto the gel at a concentration of 1 µg/cm for rGly m 4 and rGly m 3 and 30 µg/cm for soy isolate extract, respectively.

Gu et al. (2001) [654] used 1D SDS-PAGE with a 10%-20% linear gradient separating gel (BioRad) with samples that had been heated to 95° C for 3 minutes with 2% SDS and 100 mM 2-mercaptoethanol.

Rihs et al. (1999) [170] used 1D SDS-PAGE of recombinant rGly m 3 (as a fusion protein and after cleavage) or of soybean extract on 10% NuPAGE Bis-TRIS gels (NOVEX) under reducing conditions.

Mittag et al. (2004) [857] transferred proteins onto 0.2-µm pore-sized nitrocellulose membranes by means of semidry blotting. Remaining binding sites were blocked twice by means of incubation with 0.3% (v/v) Tween 20 in Tris-buffered saline (50 mmol/L Tris/HCl and 150 mmol/L NaCl, pH 7.4). The nitrocellulose strips were incubated overnight with 130 µL of patient serum diluted to 1 mL with incubation buffer, Tris-buffered saline containing 0.05% (v/v) Tween 20 and 0.1% (w/v) BSA. Immunodetection followed standard procedures.

Gu et al. (2001) [654] transferred proteins onto polyvinylidene difluoride membranes (Millipore) by using a semidry apparatus (BioRad) at temperature for 2 h. at 0.8 mA/cm². The membranes were blocked overnight with 10 mM Tris/HCl, pH 8.0, 150 mM NaCl, 0.3% (v/v) Tween 20, 1% w/v BSA and 3 mM NaN₃, then incubated with human sera (diluted 1:20 (v/v) in blocking buffer). The membranes were washed with 10 mM Tris/HCl, pH 8.0, 150 mM NaCl, 0.3% (v/v) Tween 20 and 3 mM NaN₃ then incubated with goat anti-human IgE (diluted 1:20,000 (v/v) in blocking buffer) for 12 h. at 4°C. After 3 further washes, Amersham Pharmacia's chemiluminescent substrate solution (0.1 ml/cm²) for 5 min. at room temperature revealed IgE binding.

Rihs et al. (1999) [170] transferred proteins onto polyvinylidene difluoride membranes (Millipore) by using a semidry system (16 × 16 cm) of Roth (Karlsruhe, Germany), with a discontinuous buffer system at 0.8 mA/cm² membrane for 1 hour at room temperature. After blocking with a solution consisting of 1% (w/v) BSA and 1% (w/v) polyvinylpyrrolidone (Sigma) in Tris-buffered saline, TBS (50 mM Tris/HCl, pH 7.4, 150 mM NaCl containing 5 mM NaN₃and 1 mM EDTA) and washing with TBS, membranes were cut into individual 4-mm wide strips. The strips were incubated individually with 800-µL serum samples diluted 1:10 in TBST-BSA (2% BSA dissolved in TBS containing 0.05% (v/v) Tween 20) for 18 hours at 4° C. The binding of specific IgE antibodies to the proteins was detected with alkaline phosphatase–conjugated anti-human IgE antibodies (Sigma) diluted 1:2000 (v/v) in TBST-BSA for 2 hours at room temperature, and bands were visualized with a substrate solution containing 0.4 mM 5-bromo-4-chloro-3-indolyl phosphate disodium salt and 0.37 mmol/L p-nitro blue tetrazolium chloride in 0.1 M NaHCO₃ buffer, containing 10 mM MgCl₂ (pH 9.6).

Mittag et al. (2004) [857] detected IgE binding to rGly m 4 in 19/22 patients. rGly m 3 bound in 6 patients, including 2 patients without IgE binding to rGly m 4 in immunoblots. Fourteen patients showed IgE binding to other soy proteins in the range of 8 to 10, 18, 20, 35, and 50 kDa.

Gu et al. (2001) [654] detected IgE binding to proteins at 7, 12, 20, 39 and 57 kDa from soy lecithin. The bands at 12, 20 and 39 kDa were identified by N-terminal sequencing. The 39 Kda protein was only known from incomplete Soybean EST sequences such as gi|46492876, which are distantly related to proteins of unknown function from Arabidosis, including a glycosylphosphatidylinositol-anchored protein, P59833. The 20 kDa protein was the Kunitz trypsin inhibitor and the 12 kDa was a 2S albumin.

Rihs et al. (1999) [170] report that after preincubation of sera with 50 µg of MBP-rGly m 3, a significant inhibition of IgE binding to a double band of the blotted soybean extract in the area of 14 kDa could be observed in all 4 serum samples tested, whereas the other IgE-binding bands were not affected.

Awazuhara et al. 1998 [1080] reported that IgE antibodies in the sera from 15 out of the 30 patients reacted to the proteins in the soy lecithin by immunoblotting, whereas the IgE antibodies in only three out of 22 control sera reacted to them. 40% (12/30) of soy sensitive and 4.5% (1/22) of controls reacted to a 31 kDa protein.

Müller et al. (1998) [1092] reported that IgE bound to the proteins in soy lecithin between 14 kDa and 94 kDa. The 39 kDa heat-stable allergen was identified using a monoclonal antibody as a subunit of glycinin in this study.

Awazuhara et al. 1997 [1091] reported that IgE bound most frequently to 5 allergens. These were 20 and 58 kDa in the whey fraction (i.e. soluble after aqueous extract was adjusted to pH 4.5) and 26, 31 and 78 kDa in the globulin fraction. Other allergens were detected at 28, 41, 75 and 91 kDa in the whey fraction and 40, 47, 56, 64, 72 and 91 kDa in the globulin fraction. The 47 kDa allergen actually bound weakly to 50% of sera. The authors suggest that the 31 kDa allergen might be Gly m Bd30K and the 78 kDa allergen might be beta-conglycinin.

Eigenmann et al. (1996) [48] reported IgE binding soy fractions at 46, 40, 33, 19, and 9 kDa in sera from patients allergic to peanut. Two patients allergic to peanut and soy reacted to protein fractions of 45 and 17 kDa while three soy tolerant patients bound fractions at 45 kDa, and to a lesser degree, at 21 kDa.

Ogawa et al. (1991) [441] reported that at least 16 soybean proteins with Mr from approximately 70 kDa to 14 kDa were recognized by the sera of the patients: 10 major IgE-binding components were found in the 7S-globulin fraction, and the others mainly in the 2S-globulin and whey fractions. The IgE antibodies of the patients bound most strongly and frequently to a 30 kDa protein in the 7S-globulin fraction, which appeared to be the major allergen in soybeans and was named as Gly m Bd30K. The 11S fractions did not bind IgE strongly from the sera of these atopic dermatitis patients.

Herian et al. (1990) [1078] reported that there were different patterns of IgE reactivity between sera from patients sensitised to soybeans alone which binds to protein(s) at 20 kDa and sera from those reactive to both soybeans and peanuts which binds to proteins of 50 to 60 kDa.

The wide variation in the above results may be partly due to differences between US, often peanut allergic patients; Japanese, sensitised to soy; and Europeans sensitised via birch pollen.

Niggemann et al. (1999) [661], Sicherer et al. (2000) [673], Sampson & Ho (1997) [652] and Sampson (2001) [651] used similar methods. A maximum of 8 to 10 g dry weight of dehydrated food, or equivalent liquid form, was camouflaged in a food product and given over a 90-minute period. Six doses of food were given in progressively larger quantities such as 0.5 g, 1 g, 3 doses of 2 g, and a final dose of 2.5 g. Patients received 2 challenges per day, 1 placebo and 1 test food.

Mittag et al. (2004) [857] gave two different drinks, identical in color, texture, and taste. The active drink contained 15 g of soy protein isolate (soy protein content, 90%; Protein technologies, Du Pont & Dow, Bad Homburg, Germany), 150 mL of milk, 50 mL of cream, 9 g of Sinlac (containing rice flour, and carob flour; Nestle, Vevey, Switzerland), 25 g of sugar, 7 g of cocoa, and 6 drops of banana flavor (Givaudan, Duebendorf, Switzerland). The soy protein isolate is a mildly heated soy product that was produced through extraction of defatted soybean flakes at pH 8 to 8.5, neutralization, and spray drying. The composition of the placebo drink was identical but did not contain soy isolate. Apart from soy, all ingredients were known to be tolerated by each patient.

16 patients (Mittag et al. 2004) [857].

45 patients with 7 positive (Niggemann et al. 1999) [661].

53 patients from 1980-1993 gave a positive challenge to soy (Sicherer et al., 2000 [673]).
31 patients (Sampson & Ho 1997) [652] gave a positive challenge to soy.
21 patients (Sampson (2001) [651] gave a positive challenge to soy.

Palm et al. (1999) [1081] challenged a single 4 year old patient.

Awazuhara et al. 1998 [1080] report that 7 patients were positive on challenge.

Sicherer et al. (2000) [673] report that 28% of soy challenge positive patients reacted to the first dose (0.4g) but 19% reacted only to the last dose or to open challenge. However, patients with severe symptoms were excluded from challenge. After 1992, strong IgE reaction also excluded some patients from challenge.

Mittag et al. (2004) [857] used a 2-step spit (local mucosal challenge) and swallow procedure. For the local mucosal challenge, 4 doses containing 0.3, 0.5, 1, and 2 g of soy protein, respectively, were administered to the patients, whereas for the systemic challenge, 4 doses containing 0.5, 1, 2, and 4 g of soy protein, respectively, were used.

Mittag et al. (2004) [857] report that all 16 patients showed oral allergy syndrome symptoms, sometimes with rhinitis or flush, and 4 of them also had more severe symptoms of tighness of the throat or chest.

None of the patients described by Sicherer et al. (2000) [673] suffered severe symptoms from soy in contrast to patients challenged by egg, milk, fish and peanut.

Niggemann et al. (1999) [661] reported early reactions in 4/7 (57%) and late reactions in 3/7 (43%) of 7 positive challenges to soybean in allergic children (i.e. 16% positive challenges)