|Home | About | Journals | Submit | Contact Us | Français|
Cockroach is an important allergen in inner-city asthma. Diagnosis and treatment of cockroach allergy has been impeded by the lack of standardized cockroach extracts.
We investigated the utility of a mediator release assay based on rat basophil leukemia (RBL) cells for comparing potency of German cockroach extracts.
RBL cells (line 2H3) transfected with human Fce receptor-1 were passively sensitized with sera from cockroach-allergic subjects and stimulated with serial dilutions of 3 commercial cockroach extracts (1:10 weight/volume). In addition, the in-house prepared extract was tested in separate experiments with pooled sera that gave optimal performance in the RBL assay. N-hexosaminidase release (NHR) was used as a marker of RBL degranulation and was examined in relation to intradermal skin test (ID50EAL), serum cockroach-specific and total IgE levels.
The median cockroach-specific IgE concentration in 60 subjects was 0.72 kUA/L (inter-quartile range; 0.35, 2.97); 19 sera (“responders”) produced a minimum 10% NHR to > 1extract. Responders had higher median cockroach-specific IgE (7.4 vs. 1.0 kUA/L) and total IgE (429 vs.300 kU/L) than non-responders. Ranking of extract potency was consistent between mediator release assay and the ID50EAL. For the in-house prepared cockroach extract, the dose response curves were shifted according to the concentration of the extract. NHR was reproducible between different experiments using pooled sera.
The mediator release assay measures biological potency and correlates with ID50EAL. It should be further evaluated to determine whether it could be used to replace intradermal skin test titration for assessing the potency of cockroach extract.
Mediator release assay utilizing RBL cells and human serum measures cockroach extract potency and deserves evaluation as safer, more efficient method of cockroach extract standardization for diagnosis and immunotherapy.
German cockroach (Blatella germanica) is an important allergen for asthmatics in urban areas in the US. (1–3) Exposure to high levels of the major cockroach allergen Bla g 1 is associated with asthma morbidity in cockroach-sensitized children. (1) Cockroach mitigation is difficult; sensitization has been detected in the setting of low household cockroach allergen levels. (4;5) (6) Specific immunotherapy is a proven treatment for environmental allergens. (7) Immunotherapy with cockroach allergen is an attractive option for cockroach-associated respiratory disease, but it requires well characterized, potent allergenic extracts. The current Food and Drug Administration-approved method of standardization of allergenic extract potency is based on the in vivo skin test titration (ID50EAL).(8) This methodology is uncomfortable, time and labor intensive, and carries the risk of systemic reaction.
A previous publication from our group assessed biological potency of cockroach extracts by three methods: ID50EAL, in vitro competition ELISA with human and rabbit sera, and specific allergen content (Bla g 1, Bla g 2, and Bla g 5). (9) The purpose of this study was to determine the utility of a functional in vitro mediator release assay based on rat basophil leukemia cells (RBLs) transfected with human high affinity IgE receptor type 1 and passively sensitized with human IgE for assessment of German cockroach extract biological potency and to compare this assay with ID50EAL.
Sera were obtained from cockroach-allergic participants (age 18–65 years) in the Cockroach Allergen Standardization Evaluation study. Subjects self-reported perennial respiratory symptoms (rhinitis or asthma) and had a positive prick skin test with a commercial German cockroach extract at 1:10 weight/volume (extract C). (9) All enrolled subjects underwent evaluation with intradermal skin test titration (ID50EAL).
Three cockroach extracts A, B (1:20 weight/volume), and C (1:10 weight/volume) were purchased from major manufacturers in the USA.(9)Cockroach powder for extract E was purchased from the manufacturer of extract C and was mixed in-house. In vitro testing of the allergen extracts was performed as previously published. (9)
Serum cockroach-specific IgE was measured with UniCAP (Phadia, USA); lower limit of detection is 0.35 kUA/L, upper limit is 100 kUA/L. The allergen extract used to produce the UniCAP sorbent was not one of those tested in this study. Specific IgE to recombinant cockroach allergens: rBla g 1, rBla g 2, rBla g 4, and rBla g 5 in sera with detectable cockroach-specific IgE was measured with treptavidin CAP assay (INDOOR Biotechnologies Ltd., Charlottesville, VA).(9)
RBL-2H3 cell line transfected with human Fcε receptor type-1 and the protocol for the assay were kindly provided by Dr. S. Vieths. (10) RBL cells were cultured in Eagle’s MEM, 15% RPMI with 10% FCS and G418 sulfate (pH 7.4, in 20 mM Hepes; ACROS, New Jersey, USA). RBLs were incubated with serum at a final dilution of 1:40 at 37°C in 5% CO2 for 18–20 hours in 96-well tissue-culture plates (BD Falcon™, Bedford, MA, USA). Sensitized cells were stimulated with 100 μl/well of the dilutions of cockroach extracts in a release buffer with 50% D2O (ACROS, New Jersey, USA) at 37°C in 5% CO2 for 1 hour. Rabbit IgG anti-human polyclonal IgE (Bethyl Laboratories, Inc., Montgomery, TX, USA) was used as a positive control for IgE-mediated degranulation. To determine N-hexosaminidase release (NHR), 30 μl of supernatant was gently mixed with 50 μl of P-nitrophenyl-N-acetyl-β-D-glucosaminide solution (pH 4.5, Sigma-Aldrich, St. Louis, MO, USA). After 1 hour at 37°C in 5% CO2, 100 μl of 0.2 M glycine solution (pH 10.7) was added and absorbance at 405 nm was measured. RBLs were lysed with 1% Triton X-100 (Sigma Chemical Co., St. Louis, MO, USA) for total release. Results were expressed as the percentage of release from cells sensitized with individual serum minus spontaneous release (with buffer) which was then divided by total release. Responders were arbitrarily defined as those sera that produced ≥ 10% NHR to at least 1 cockroach extract.
Intradermal skin test titration in 60 subjects was performed according to the protocol described by Turkeltaub et al. (8;9) In this test, the biological potency is estimated by determining the extract dilution at which the sum of perpendicular erythema diameters is 50 mm. Briefly, serial three-fold dilutions of cockroach extracts A, B and C (starting from the lowest concentration) were injected intradermally on the back. Erythema was measured at 15 minutes, and the sum of erythema diameters was calculated by adding the longest possible diameter across the area of erythema and the shorter diameter perpendicular to and through the midpoint of the longest diameter. The objective was to establish a dose-response curve in each subject, with the sum of erythema diameters ranging from 0 to 125 mm and containing at least four valid data points that bracketed 50 mm.
The results of the RBL assay and skin test data were analyzed using the drc package(11) in R. (12) The data were fit by four parameter logistic models. (11) The parameters in the model estimate the minimum and maximum responses, ED50 (extract concentration that induced half maximal response), and relative slope of the ED50. Results from the fitted model were used to compare the maximum and minimum responses as well as to compare the relative fit of different extracts. Each individual’s data were fit with a single model producing a different curve for each extract to account for subject specific variation across the extracts. Models were fit separately for skin test data and RBL assay data. Additionally, models were fit combining 6 subjects’ responses producing one set of curves for the skin test results and a second set based on the RBL data. Interpolation on the fitted models from the skin test data was used to determine ID50EAL values. Potency of the extracts using the skin test data was calculated as bioequivalent allergy units (BAU)(13) using the formula
Potency measures evaluated using the RBL assay data were calculated using ED50 values. ED50 values were obtained from the parameter estimates of the models, and the potency measured was defined as 1/ED50. Plotting the curves produced by the modeling allowed for comparison of the different extracts.
The study was approved by the Institutional Review Boards of the participating institutions and the informed consent was obtained prior to subject enrollment.
Sixty serum samples from subjects evaluated with ID50EAL method were screened;(9) of those, 40 had cockroach-specific IgE antibody ≥ 0.35 kUA/L with a median level (IQR) of 0.72 kUA/L (0.35, 2.97). Comparisons between responders and non-responders are shown in Table I.
In our initial experiments, we noted a significant non-IgE-mediated release with cockroach extract alone (without presence of human IgE) when RBL cells were stimulated with higher concentrations (1st-3rd 3-fold dilutions and 10−2 dilution) of cockroach extracts. In subsequent experiments we considered the 4th 3-fold dilution and 10−3 dilution to be representative of the highest IgE-mediated release. Of note, we have not observed this non-IgE-mediated (presumably pharmacologic) effect with high concentrations of other allergens, such as birch pollen, dog dander, cow’s milk, egg white, and shrimp extracts (not shown).
Serum was considered to be responsive if NHR to at least one extract (A, B, or C) was ≥ 10% at the 4th 3-fold dilution. There were 19 responders and 41 non-responders. Cockroach-specific IgE was necessary but not sufficient for good performance in the mediator release assay. None of the sera with cockroach-IgE < 0.35 kUA/L gave ≥ 10% NHR; among the sera with detectable cockroach-IgE, only nineteen (19/40; 46%) sera gave ≥10% NHR. Responders had significantly higher levels of cockroach–specific IgE, total serum IgE, and specific/total IgE ratio. (Table I) Among the 18 responders with detectable IgE levels to recombinant cockroach allergens, 3 (16.7%), 9 (50%), 5 (27.8%), and 11 (61%) had detectable IgE to rBla g 1, r Bla g 2, rBla g 4 and rBla g 5 respectively, versus 0 (0%), 2 (15.4%), 3 (23%), and 3 (23%) among the 13 non-responders. There was a significant positive correlation between the NHR and cockroach-specific IgE level, and ratio of cockroach-specific/total IgE for all extracts (Table II).
The mediator release assay and ID50EAL dose response curves for cockroach extracts A, B, and C are shown side by side for three representative subjects (Fig 1). For 16 of the 19 responders (84%), the three extracts were ranked in identical order in both assays (Fig E1).
We chose sera from six RBL responders with optimal performance in the RBL assay (defined as highest NHR at 4th 3-fold dilution for all three extracts) to calculate biological potency of cockroach extracts and to compare with the biological potency based on ID50EAL (Fig 2, Table III). Extract B was the most potent and extract A was the least potent in both assays. Extract potency ranking by these two methods was also consistent with the ranking by competition ELISA using rabbit and human sera and specific allergen content (Bla g 1, Bla g 2, and Bla g 5).(9)
In-house prepared cockroach extract E was mixed at 1:2.5, 1:10, 1:25, and 1:250 weight/volume. Serial 10-fold dilutions were tested with a serum pool (cockroach-specific IgE was 28.3 kUA/L) made of equal parts of sera from 7 responders as well as with serum from an individual subject (cockroach-specific IgE was 19.6 kUA/L). The dose response curves had the same slopes but ED50 differed by about a factor of 5. Half maximal release occurred at the higher dilution of the 1:2.5 extract than 1:25 and than 1:250, demonstrating that the performance of the cockroach extract in the mediator release assay reflects the concentration of cockroach source material. (Fig 3)
Identical experiments were performed on 4 different days. RBL cells were sensitized with serum from 6 individual subjects and stimulated with serial 10-fold dilutions of extract E at 1:2.5, 1:10, and 1:25 weight/volume. (Fig 4) Individual variances were stable within each extract dilution. In each case, 88.9% to 90.1% of the variance was attributed to differences between individuals, and the remaining portion represented the within-individual variance.
We report for the first time the comparison between the intradermal skin test titration method (ID50EAL) of allergen extract standardization and an in vitro mediator release assay based on the RBL cell-line transfected with human IgE receptor type 1 and passively sensitized with human IgE. We demonstrate that while an in vitro mediator release assay measures potency of cockroach extract with comparable accuracy as ID50EAL, it has significant advantages over ID50EAL for allergen standardization. The in vitro mediator release assay is less laborious and less expensive than ID50EAL, and it has no risk for a systemic allergic reaction. Furthermore, it can be performed with sera selected for optimal performance and stored frozen for prolonged periods of time, resulting in less variability as compared to ID50EAL.
Accurate assessment of cockroach extract biological potency is crucial for diagnosis and immunotherapy. Allergen extracts have inherent bio-variability due to collection methods, storage, processing of raw materials, and extraction and manufacturing techniques. (14) Standardization based on the concentration of the source material (expressed as weight per volume, wt/vol) or the protein content in the source material (expressed as protein nitrogen units, PNU) is not reliable because extracts from natural sources have varying protein patterns over time. (15) In addition, there is little correlation between these designations and biological measures of allergen potency. (16;17) In the US, cockroach extracts vary in protein content, electrophoretic banding patterns, relative potency, and Bla g 2 levels. (18)
Immunochemical assays such as IgE inhibition tests and allergen detection by Western blotting are subject to epitope alteration/destruction due to adsorption of allergens to solid matrices, are susceptible to the competition from IgG of the same specificity, and often do not correlate well with data obtained by skin testing. (19–22) Measurement of major allergens depends on the use of polyclonal or monoclonal antibodies and does not measure biological activity. (23) (24) Although this approach is favored in Europe; standardization based on a selected major allergen may neglect the minor allergens of clinical importance. (24) Furthermore, it requires identification of clearly immuno-dominant allergens, which has not been accomplished in the case of cockroach allergy. (9;25)
In the US, standardization of extracts is based on a functional in vivo ID50EAL assay. (8) This method requires serial intradermal skin test (titration), is labor-intensive, expensive, and unpleasant to the tested individuals. In addition to the potential variability associated with the skin test (e.g., different batch of extract, amount of allergen injected, depth of the injection, location of the test on the back), it is usually not feasible to bring back the original subjects for repeat testing year after year, which further limits the reproducibility of the ID50EAL.
Functional assays based on histamine release from human donors’ basophils, and basophil activation (expression of CD63 and CD203c) detected by flow cytometry have been evaluated in research studies. (26–28) Practical application of basophil activation assays has similar limitations as that of the skin testing, namely the availability of basophil donors and time constraints (basophil assays have to be processed promptly following blood collection).
We chose to evaluate an in vitro mediator release assay utilizing the RBL cell line RBL-2H3 transfected with human high affinity Fcε receptor type 1 and passively sensitized with human IgE. RBL-2H3 cells are stable transfectants that can be maintained in culture for long periods of time. (10;29) RBL cells have the functional characteristics of mast cells in regard to IgE-induced mediator release, but in contrast to wild type mast cells, cross-linking of IgG via Fc-receptor on RBL-2H3 cells fails to induce any detectable mediator release. RBL-based mediator release assay was reported to detect very low levels of allergen, e.g, cat allergen Fel d 1 gave detectable release at the level of 10 pg/ml. (30) The RBL-based assay was previously reported to estimate potency of tree and grass pollen, house dust mite, cat dander, and peanut extracts. (10;19;30) The assay was sensitive and had a low coefficient of variation: 15% for NHR and 25% for ED50, which made it suitable for detecting differences between batches of allergen extracts. (19)
In the RBL model, the effect of interaction between the allergen and allergen-specific IgE is demonstrated via mediator release. Mediator release assay was dependent on the presence of cockroach-specific IgE and on the concentration of cockroach allergen. We found a significant positive correlation between NHR and: 1) cockroach-specific IgE levels; and 2) ratio of cockroach-specific IgE to total IgE. These observations suggest that both parameters may be important in pre-selecting sera for evaluation in the mediator release assay. Finally, we demonstrated that the mediator release assay was reproducible on different days, and thus might be utilized for standardization of different batches of extract over periods of time.
Our group has recently reported that the ID50EAL method provided a valid estimate of cockroach extract biological potency that was in agreement with the relative potency determinatiom with in vitro competition ELISA using rabbit and human sera, and specific content of Bla g 1, Bla g 2, and Bla g 5. (9) We demonstrated that the in vitro mediator release assay paralleled the results of ID50EAL skin testing for the potency of individual German cockroach extracts in adult subjects sensitized to cockroach. We chose to express biological potency as 1/ED50 rather than to establish a more precise unit of biological potency due to the small number of serum samples with optimal performance in the mediator release assay. This limitation reflected the selection strategy, which was based on the results of the prick skin test reactivity to cockroach extract and resulted in a cohort that included many subjects with undetectable or low levels of cockroach-specific IgE. In future studies, selection of serum samples should be based upon detection of moderate to high levels of cockroach-specific IgE and pre-screening of serum samples to identify “high responders”.
As in many bioassays, there are subjects (sera) with detectable levels of cockroach-specific IgE that do not induce good mediator release in the mediator release assay (non-responders). For instance, about 10% of atopic donors have non-responsive basophils in the histamine release assay.(31) Discordance between the magnitude of mediator release in the in vitro assay and serum allergen-specific IgE has been reported previously. (10;32;33) Lack of responsiveness in the mediator release assay despite detectable allergen-specific IgE might be due to factors that are not accounted for in the immunochemical tests. These factors potentially include: 1) low affinity and avidity of IgE(34), 2) effect of the steric site of allergen recognition(35;36), 3) availability of free (not bound in complexes) IgE(37;38), and 4) low ratio of cockroach-specific IgE to total serum IgE (dilution effect). (33;39) However, once “high performing” sera are identified, they can be stored for long periods of time (estimated stability of IgE immunoreactivity in frozen serum samples stored at temperatures lower than −20°C is about 5 years-R.G. Hamilton, personal communication) and used to standardize different batches of allergenic extracts. This would minimize the variability from performing skin test titration in different allergic individuals over time and would also be less laborious and less expensive.
We observed a consistent non-IgE-mediated release upon stimulation of non-sensitized RBL cells (without added human serum) at high concentrations of cockroach extracts (present in the experiments performed on different days and with different extracts), which has not been reported with other allergen extracts (pollen, cat dander, peanut, dust mite). Although none of the known cockroach allergens are proteolytically active, protease activity has been observed in the whole-body extract of German cockroach and might potentially contribute to the non-IgE mediated effect. (24;40–42) We were able to reduce this non-IgE-mediated effect by incubating cockroach extract with protease inhibitors prior to exposing RBL cells to cockroach extract without human serum (not shown). Additional reasons for the non-IgE-induced mediator release might be the presence of endotoxin and/ or (1,3)-β-d-glucans. (43) The mechanism of the non-IgE mediated effect of cockroach extract will need to be further explored.
In conclusion, the in vitro mediator release assay utilizing RBL cells transfected with human high affinity IgE receptor type-1 and passively sensitized with human IgE estimates biological potency of cockroach extract with similar accuracy as intradermal skin test titration ID50EAL. Considering a number of significant advantages over the laborious ID50EAL, the in vitro mediator release assay should be further evaluated for application in the standardization of commercial cockroach allergen extracts for diagnosis and immunotherapy.
This project has been funded in whole or in part with Federal funds from the National Institute of Allergy and Infectious Diseases, National Institutes of Health, under Contracts number N01-AI-25496 and NO1-AI-25482. A. Nowak-Wegrzyn is supported in part by NIH NIAID AI 059318.
We thank Dr. Wayne Shreffler for helpful discussions and Russell Castro for excellent technical assistance.
Conflict of Interest Statement
All authors report no conflict of interest.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.