A total of 36 HCW were studied, of which 10 had no clinical symptoms of latex allergy; the remaining 26 subjects had clinically proven latex allergy 35. Among the 21 SB patients studied, 13 had clinical latex allergy 11. Latex allergy in health care workers was diagnosed by (a) a history of skin and respiratory symptoms often progressing from contact dermatitis through urticaria to asthma and anaphylaxis on latex contact, usually with latex glove powder inhalation, or (b) immediate wheal and flare skin reaction to latex glove antigens, (c) a history of reaction to cross-reactive latex antigens such as bananas or other fruits, and/or (d) serum IgE antibodies to latex glove extracts carried out by ELISA. Latex allergy in SB patients was diagnosed by a history of perioperative anaphylaxis and/or the demonstration of respiratory symptoms on latex glove powder contact, and/or the demonstration of antibody to latex antigens and a history of cross-reaction to food allergens 11. All sera were evaluated for latex specific IgE antibody using a Malaysian non-ammoniated latex extract, two glove extracts routinely used in our laboratory to confirm the diagnosis 1112.

Four crude latex extracts and 11 purified and recombinant allergens from H. brasiliensis latex were used for in vitro studies of latex specific serum IgE antibody. The purified allergens used in the study are listed in Table 1. All antigens were used in an ELISA to evaluate latex specific IgE antibody in sera of patients with latex allergy and normal healthy controls 51112. Latex collected after tapping H. brasiliensis trees (rubber trees) was shipped frozen to the laboratory from Malaysia. The clear serum phase of the latex was collected after centrifugation of the coagulated latex as described previously 51213. This extract designated as Malaysian non-ammoniated latex (MNA) was characterized and used in ELISA as described previously 5. Another crude latex extract was from clone RRIM 600 and was obtained from Greer Laboratory 7. Two latex glove extracts were also used in the study. These gloves were selected from two different manufacturing sources, one with more extractable protein, while the other one with a lower latex protein content. The allergens were extracted from pieces of latex gloves by stirring with PBS in a flask for 15 min at room temperature as previously described 1415. We used two different ImmunoCAPs; in one the crude latex was used to make the CAPs, while in the other in addition to the regular CAP, Hev b 5 was also supplemented. This modification was devised to remedy the lack of Hev b 5 in the clotted serum of rubber latex.

Three of the allergens Hev b 2, Hev b 4, and Hev b 13 were purified from latex by the Malaysian laboratory (HYY) as previously described 51617. The genes for Hev b 1, 3, 5, 6, 7, 8, 9, and 10 were cloned from cDNA libraries and Hev b 1, 3, 5, and 6 were expressed in the Medical College of Wisconsin laboratory (VPK), while Hev b 7, 8, 9, and 10 were cloned and expressed in the University of Vienna laboratory (HB) 1827.

The protein profile of the extract was studied by sodium dodecyl sulfate polyacrylamide gel electrophoresis. Electrophoresis was carried out by loading 10 micrograms of proteins on a 12% SDS polyacrylamide mini gel and running at 200 mv/cm for 40 to 50 minutes 5. The gels were stained with Coomassie brilliant blue R-250 and the stained bands in the gel were compared and the molecular sizes ascertained. The reactivity of antigens to serum IgE was studied using pooled sera from HCW and SB patients with latex allergy and controls by ELISA and Western blot.

The MNA, Clone RRIM 600, glove extract antigens, and purified latex proteins were coated at a concentration of 5-μg protein/ml. All dilutions and coating concentrations of the antigens and reagents were derived from checkerboard titration using latex positive and negative sera. The ELISA was performed as previously described 5. Briefly, one hundred micro liters of the preparations were added to the wells of polystyrene micro titer plates (Immunolon II HB, Therma Lab Systems, Franklin, MA). The plates were incubated at room temperature for 3 hours, followed by a further incubation at 4°C overnight. After washing the plates with PBS, containing 0.05% Tween 20 (PBS-T), the wells were blocked with 0.5% BSA in PBS-T. The wells were again washed and 100-μl of 1:25 dilution of the serum added to each well, incubated at room temperature for 3 hours, and washed as before. One hundred-μl of biotinylated mouse, anti-human IgE monoclonal antibody (Zymed Laboratories, Inc., San Francisco, CA) was added to each well and the plates were incubated for 1 hour at room temperature, washed as before and 100 μl of 1:2000 dilution of streptavidin peroxidase was added to the wells. This was followed by incubation for 30 minutes and washing again. Finally, the peroxidase activity was developed with o-phenylenediamine substrate in citrate buffer. The color was developed for 15 minutes in a dark chamber and the reaction stopped by the addition of 25 μl of 2N H₂SO₄ solution. The color was read in an ELISA plate reader using a 490 nm filter (Molecular Devices; Sterling, VA). The optical density (O.D) values were corrected by subtracting the blank values and the average of three wells was taken. A value exceeding mean plus two standard deviation (SD) of HCW and SB patients without latex allergy was taken as a cut off value for positivity.

The Pharmacia ImmunoCAP was used to demonstrate latex specific IgE in the sera of patients and controls according to the instructions of the manufacturer. Both Hev b 5 amplified (rk82) and non-amplified (k82) ImmunoCAPs were used. The protocol of the manufacturer was followed, and a value of 0.35 kU_A/L or more was considered positive.

The mean O.D values for all allergens were calculated and the results were analyzed by the multivariate analysis of variance (MANOVA). A P-value of 0.05 was considered significant. When a significant difference was detected, a stepwise discriminant analysis was also performed to select the significant allergens that delineate the positive and negative groups by their reactivity or non-reactivity with IgE by ELISA. The variables with P-value of < 0.05 were chosen as the significant variables, while the variables with P-value > 0.20 were removed at each step of the discriminant analysis. Using all the significant allergens, Fisher discriminant functions for the latex allergic and non-allergic HCW and SB subjects were calculated. Based on this, each case was assigned two Fisher discriminant scores, one for the latex allergic group and the other one for the non-allergic group. Each subject was classified into a group (latex allergic or non-allergic) based on the higher corresponding Fisher discriminant score value as the case may be 528.

The protein profiles using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) of MNA, Clone RRIM600 and the various purified latex antigens are shown in Figure 1. MNA and Clone RRIM600 showed a number of bands in SDS-PAGE, while most of the purified allergens used in the study showed single bands. A few of the purified antigens showed additional weak bands in the gel, the immunoblots of the protein reacted with IgE from latex allergic patients showing IgE binding with only the major bands and not with the weak bands. The two glove extracts showed very weak bands. The Western blots of both glove extracts, MNA, and Clone RRIM600 showed multiple reactive bands with the pooled latex allergic patient sera, but not with the normal control sera.

Of the 36 HCW evaluated, 26 were symptomatic with urticaria or asthma on latex allergen exposure; 10 subjects had no symptoms on exposure to latex products in the health care workplace. All 36 subjects included in this study were exposed to latex proteins present in gloves or other latex products. The IgE reactivity of the sera from HCW patients to11 purified latex antigens by ELISA is shown in Figure 2. Hev b 2, 4, 5, 6, and 13 showed significant binding to IgE in the patients' sera, while Hev b 1, 3, 7, 8, 9, and 10 failed to show significant binding to IgE compared to controls (Fig. 2). In ImmunoCAP, 25/26 HCW patients showed 0.35 kU_A/L or more, while one normal subject without latex allergy was also positive (Fig. 3). One out of the 26 patients who failed to show latex specific IgE by the ImmunoCAP showed strong reactivity with Hev b 5 by ELISA and Hev b 5 amplified ImmunoCAP. On the other hand, one patient negative to all crude antigens and most purified antigens showed strong reactivity to Hev b 5 and amplified Hev b 5 ImmunoCAPs. This patient also showed significant IgE to Hev b 5 and Hev b 13 by ELISA. The solitary HCW patient that showed strong reactivity with Hev b 1 and Hev b 3 also reacted with most other latex allergens studied. Hev b 7, 8, 9, and 10 invariably showed only very weak reactions with specific IgE in the sera of most patients tested, while Hev b 2, Hev b 6, and Hev b 13 consistently showed high levels of IgE in more patients compared to other purified allergens. Three out of 10 normal subjects also showed IgE to Hev b 2 in their sera, but the levels were comparatively lower than those detected in patients. Hev b 13 also failed to react with two latex allergic patients, but did not show any reactivity with the normal controls.

The reactivity of various allergens to the IgE of SB patients are shown in Figure 4. All 13 latex allergic patients showed significantly elevated IgE levels by both ImmunoCAPs and ELISA using crude latex extracts, while none of the non-allergic SB showed any reactivity (Fig. 3). Both Hev b 1 and Hev b 3 demonstrated strong reactivity with IgE of 11/13 patients; the remaining two patients had low levels of IgE against these two allergens. Hev b 6 demonstrated strong IgE binding to all but one SB patient with latex allergy, but had no reactivity with SB patients without latex allergy. Three patients each failed to react with Hev b 5 and Hev b 13, while only three showed binding of IgE to Hev b 7. Hev b 8, 9 and 10 failed to show binding with any of the patients or control sera. None of the SB patients without latex allergy showed any significant reactivity with IgE to any of the latex allergens except Hev b 2, which showed strong reactivity with only one patient.

The reactivity of HCW and SB sera against the four crude antigens are shown in Figure 5A and 5B. Both NRL extracts reacted strongly with both SB and HCW, with Clone RRIM600 showing more reactivity with HCW patients. Among the crude extracts studied, Glove 1 invariably showed less reactivity. All four antigens showed reactivity with the sera from a few control subjects.

The binding of IgE to allergens Hev b 1 to Hev b 6 and Hev b 13 showed a significant difference (P < 0.05) between SB patients with and without latex allergy when studied individually. Hev b 7 to Hev b 10 failed to show any significant IgE binding reactivity between these two groups. Among the HCW patients with latex allergy studied, strong reactivity was detected only with Hev b 2, Hev b 5, Hev b 6, and Hev b 13. When all 11 purified latex proteins were used together and analyzed the data by MANOVA, a significant difference was detected with latex allergic and non-allergic subjects from both HCW and SB groups (P < 0.05).

Stepwise discriminant analysis of the ELISA data from SB patients selected Hev b 1, Hev b 3, and Hev b 6 and together these antigens delineated all the latex allergic and non-allergic subjects. The Fisher discrimination function for the positive group is -10.549 + 11.569 Hev b 1 -9.189 Hev b 3 + 5.443 Hev b 6, and for the negative group is -0.694 + 0.098 Hev b 1 -0.068 Hev b 3 + 0.027 Hev b 6. All the SB subjects studied could be classified into latex allergic or non-allergic, and the specificity and sensitivity were found to be 100% by ELISA using these allergens.

The Stepwise discriminant analysis on HCW workers selected only Hev b 6 as the major allergen, perhaps due to the fact that this protein is the major allergen with marked specificity. The Fisher discriminant function for the latex allergic patient group when Hev b 6 alone is used is -1.511 -1.627 Hev b 6, and for the non-allergic group is -0.694 + 0.063 Hev b 6. This analysis correctly classified 17 out of 26 patients as having latex allergy and all 10 HCW subjects without symptoms. However, Hev b 2, Hev b 5, and Hev b 13 also showed significant reactivity and hence, discriminant analysis of the IgE binding of Hev b 2, Hev b 5, Hev b 6, and Hev b 13 was also carried out. The sensitivity and specificity using only Hev b 6 or using Hev b 2, Hev b 5, Hev b 6, and Hev b 13, and using all allergens Hev b 1 to Hev b 13 were carried out for all patients and the results indicate a sensitivity of 65 to 85% and a specificity of 100% (Table 2).