Adequate quality is essential for any medicinal product to be eligible for marketing. Quality includes verification of the identity, content and purity of a medicinal product in combination with a specified production process and its control. Allergen products derived from natural sources require particular considerations to ensure adequate quality. Here, we describe key aspects of the documentation on manufacturing and quality aspects for allergen immunotherapy products in the European Union and the United States. In some key parts, requirements in these areas are harmonized while other fields are regulated separately between both regions. Essential differences are found in the use of Reference Preparations, or the requirement to apply standardized assays for potency determination. As the types of products available are different in specific regions, regulatory guidance for such products may also be available in one specific region only, such as for allergoids in the European Union. Region-specific issues and priorities are a result of this. As allergen products derived from natural sources are inherently variable in their qualitative and quantitative composition, these products present special challenges to balance the variability and ensuring batch-to-batch consistency. Advancements in scientific knowledge on specific allergens and their role in allergic disease will consequentially find representation in future regulatory guidelines.

German cockroach (GCr) allergen extracts are complex and heterogeneous products, and methods to better assess their potency and composition are needed for adequate studies of their safety and efficacy.The objective of this study was to develop an assay based on liquid chromatography and multiple reaction monitoring mass spectrometry (LC-MRM MS) for rapid, accurate, and reproducible quantification of 5 allergens (Bla g 1, Bla g 2, Bla g 3, Bla g 4, and Bla g 5) in crude GCr allergen extracts.We first established a comprehensive peptide library of allergens from various commercial extracts as well as recombinant allergens. Peptide mapping was performed using high-resolution MS, and the peptide library was then used to identify prototypic and quantotypic peptides to proceed with MRM method development. Assay development included a systematic optimization of digestion conditions (buffer, digestion time, and trypsin concentration), chromatographic separation, and MS parameters. Robustness and suitability were assessed following ICH (Q2 [R1]) guidelines. The method is precise (RSD < 10%), linear over a wide range (r > 0.99, 0.01-1384 fmol/μL), and sensitive (LLOD and LLOQ <1 fmol/μL). Having established the parameters for LC-MRM MS, we quantified allergens from various commercial GCr extracts and showed considerable variability that may impact clinical efficacy.Our data demonstrate that the LC-MRM MS method is valuable for absolute quantification of allergens in GCr extracts and likely has broader applicability to other complex allergen extracts. Definitive quantification provides a new standard for labelling of allergen extracts, which will inform patient care, enable personalized therapy, and enhance the efficacy of immunotherapy for environmental and food allergies.

Regulatory approaches for allergen immunotherapy (AIT) products and the availability of high-quality AIT products are inherently linked to each other. While allergen products are available in many countries across the globe, their regulation is very heterogeneous. First, we describe the regulatory systems applicable for AIT products in the European Union (EU) and in the United States (US). For Europe, a depiction of the different types of relevant procedures, as well as the committees involved, is provided and the fundamental role of national agencies of the EU member states in this complex and unique network is highlighted. Furthermore, the regulatory agencies from Australia, Canada, Japan, Russia, and Switzerland provided information on the system implemented in their countries for the regulation of allergen products. While AIT products are commonly classified as biological medicinal products, they are made available by varying types of procedures, most commonly either by obtaining a marketing authorization or by being distributed as named patient products. Exemptions from marketing authorizations in exceptional cases, as well as import of allergen products from other countries, are additional tools applied by countries to ensure availability of needed AIT products. Several challenges for AIT products are apparent from this analysis and will require further consideration.

The FDA lot release protocol review and testing program insures that the potencies of standardized allergenic extracts distributed in the U.S. are within established limits. For cat and short ragweed pollen allergenic extracts, potency is defined according to the concentration of Fel d 1 and Amb a 1, respectively, and is measured with the radial immunodiffusion assay (RID). The RID is labor intensive and subjective, and dependent on devices that are no longer manufactured. We have therefore developed a sandwich ELISA to more accurately, precisely, and reproducibly measure the potency of these standardized allergenic extracts. Three candidate single chain variable fragment antibodies (scFv) against each allergen were purified by affinity chromatography and used as the coating (capture) antibody. Purified native Fel d 1 (Indoor Biotechnologies) and Amb a 1 were used as antigens. Revealing antibodies were polyclonal sheep anti-cat or anti-short ragweed pollen sera in combination with HRP-conjugated rabbit anti-sheep antibody (KPL). Data were analyzed using a 4-parameter logistic model. The assay was validated with CBER's current standardized cat and ragweed pollen reference standards, and with extracts purchased from the manufacturers. Each combination of capture scFv antibody and revealing polysera was sensitive, highly specific, and linear within a wide range of concentrations. Sandwich ELISAs for cat and short ragweed allergen extracts will enhance the FDA lot release protocol review and testing program's ability to measure and confirm the potency of standardized cat and short ragweed pollen allergenic extracts.

Neurospora crassa has been used as a platform for rapid and cost-effective vaccine production (Allgaier et al. Biologicals 2009; 37:128). The purpose of this study is to screen Neursopora crassa extract for the existence of potential allergens using human scFv antibodies. A highly complex human scFv phage library (Creative Bio-labs) was panned and screened against spent soytone-based Neurospora medium. Affinity-purified soluble scFv antibodies were screened against spent medium. Antibody-binding Neurospora proteins were identified by electrophoresis and immunoblotting and, when possible, isolated using protein-L based immunoprecipitation. One such target protein was excised from a Coomassie-stained gel and identified using laser capture mass spectrometry. Out of 65 positive clones four unique clones were identified by sequencing. All of the Neurospora-specific antibodies were affinity purified. All 4 of the antibodies recognized a specific band of about 70 kDa molecular mass by immunoblot analysis. SDS-PAGE of immunoprecipitated product also revealed protein running at the level of ∼70 kDa molecular weight. Mass spectrometric analysis performed on the protein band was consistent with glucoamylase I precursor of Neurospora crassa (XP_956966). The glucoamylase shares 54% identity and 65% similarity with glucoamylase of Aspergillus niger (Q870G8). Aspergillus-derived glucoamylase has been associated with occupational allergies (Quirce et al. Ann Allergy Asthma Immunol 2002; 89:197). Human scFv antibodies were successfully used for identification of a potential allergen, glucoamylase I precursor, in Neurospora crassa extract.

RATIONALE: To improve the safety and efficacy of allergen immunotherapy, better methods are needed to measure the potency and composition of allergen extracts. The purpose of this study is to produce GCr-specific scFv antibodies from a chicken antibody cDNA library, and use these antibodies to analyze GCr extracts.METHODS: An avian antibody library was generated as previously described (Finlay, et al., Clin Exp Allergy. 2005; 35:1040-8). Unique GCr-specific antibodies were identified by sequencing. The phagemids of selected clones were transformed in E. coli, induced with IPTG and the soluble scFvs were purified by metal chelate affinity chromatography. The purified antibodies were used to analyze a GCr extract (E6CG) by ELISA and immunoblot. E6CG is a GCr internal standard extract with an estimated protein content of 20 g/L.RESULTS: The selected unique antibodies were purified and 32 kDa bands were analyzed using SDS-PAGE. 14 high affinity antibodies recognized specific proteins of varying molecular masses in immunoblot. Protein bands recognized by the expressed antibodies migrated at 17, 32, 50, 70, 76, and 102 kDa. Direct ELISA was used to define differential response to different doses of extracts. EC50s calculated from dose response curves ranged from 10−4 to 7 x 10−2 g/L of E6CG.CONCLUSION: Chicken scFv libraries proved to be a useful tool in preparation of unique scFv antibodies. We used these antibodies for analyses and identification of proteins in highly complex GCr extracts. RATIONALE: To improve the safety and efficacy of allergen immunotherapy, better methods are needed to measure the potency and composition of allergen extracts. The purpose of this study is to produce GCr-specific scFv antibodies from a chicken antibody cDNA library, and use these antibodies to analyze GCr extracts. METHODS: An avian antibody library was generated as previously described (Finlay, et al., Clin Exp Allergy. 2005; 35:1040-8). Unique GCr-specific antibodies were identified by sequencing. The phagemids of selected clones were transformed in E. coli, induced with IPTG and the soluble scFvs were purified by metal chelate affinity chromatography. The purified antibodies were used to analyze a GCr extract (E6CG) by ELISA and immunoblot. E6CG is a GCr internal standard extract with an estimated protein content of 20 g/L. RESULTS: The selected unique antibodies were purified and 32 kDa bands were analyzed using SDS-PAGE. 14 high affinity antibodies recognized specific proteins of varying molecular masses in immunoblot. Protein bands recognized by the expressed antibodies migrated at 17, 32, 50, 70, 76, and 102 kDa. Direct ELISA was used to define differential response to different doses of extracts. EC50s calculated from dose response curves ranged from 10−4 to 7 x 10−2 g/L of E6CG. CONCLUSION: Chicken scFv libraries proved to be a useful tool in preparation of unique scFv antibodies. We used these antibodies for analyses and identification of proteins in highly complex GCr extracts.

RATIONALE: German cockroach (GCr) has been reported to cause asthma-related disease in the US. Seven GCr allergens have been reported. We have developed single chain variable fragment antibodies (scFv) to GCr for a novel multiplex assay of GCr extract potency. To qualify these antibodies, we have examined target allergens by immunoprecipitation. METHODS: 14 unique scFv antibodies selected by phage display and panning were purified by affinity chromatography. Two of these, 2A1 and 6A2, were coupled to agarose beads, and incubated with GCr extract. The eluted proteins were analyzed by SDS-PAGE, N-terminal sequencing and mass spectrometry (MS). In addition, the gene sequence for the likely 2A1 target was obtained (from FTC), and the protein was expressed in E. coli and used for screening of GCr-allergic human serum pools for presence of specific IgE. RESULTS: The 2A1 target has an apparent mass of 78 kDa, and the N-terminal sequence is ATIPADKVFL. Sequence BLAST searches, using the N-terminal sequence and internal sequences obtained from MS, indicated the closest match to isoform 2 of a putative Per a 3 homologue (ACY40651). Direct ELISA and dot blot data using expressed recombinant Per a 3 homologue indicated the presence of allergen-specific IgE in screened pools. The 6A2 target has an apparent mass of 33 kDa, and the peptide sequences obtained from mass spectrometry suggest that it is Bla g 7. CONCLUSIONS: We have identified two GCr antigens, one of which appears to be a recently reported Per a 3 homologue, by immunoprecipitation using scFv.

RATIONALE: Multiplex microbead flow cytometry has been used to diagnose allergy by binding allergens to beads and incubating these beads in the presence of patient IgE. We have shown that this technology may also be used to determine the potency of allergen extracts. In this study we examine the effect of the presence of multiple bead-bound antibodies on antibody binding. METHODS: Six anti-Fel d 1 recombinant antibodies were generated and were covalently bound to carboxy-labeled beads. Antibody-bound beads were then used to measure Fel d 1 in commercial cat hair extracts using bead-based flow cytometry (Bioplex, Biorad). These potencies were compared to those obtained by manufacturers and CBER/FDA using a conventional antibody-based method. Potencies were calculated by comparing dose response curves using Prism software (GraphPad). Bead-antibody pairs were tested to determine if the presence of additional bead-antibody pairs affected the apparent potency of the extract. RESULTS: Bead-based flow cytometry assay using anti-Fel d 1 recombinant antibodies yields potency numbers within the range of potencies determined using conventional methods. Minimal differences were detected in the apparent potencies of the extracts even when four additional bead-antibody pairs were added. Six lots of cat hair tested using six bead-antibody pairs had potencies (2, 6, 10, 4, 13, 4) similar to those obtained by a manufacturer (4, 6, 15, 6, 18, 7) (r2 > 0.95). CONCLUSIONS: We have developed a bead-based assay using recombinant antibodies that accurately determines Fel d 1 levels in cat allergenic extracts. This assay is reproducible and consistent with data obtained by other methods and testers.

RATIONALE: D. farinae extracts contain more endotoxin than D. pteronyssinus extracts. We have shown that this endotoxin is associated with bacterial DNA in D. farinae mites. Sequence analysis suggests that this bacterial DNA is from Bartonella. In this study, we have probed for bacterial DNA in D. farinae extracts and other mite species. METHODS: We tested 27 lots of mite extracts (13 lots D. farinae and 14 lots D. pteronyssinus) obtained from six US manufacturers, as well as 10 non-mite extracts (cat hair/pelt, cockroach, honeybee, Bermuda and ryegrass). In addition, we tested whole mite preparations from five additional species (C. arcuatus, L. destructor, E. maynei, A. siro, and T. putrescentiae,). DNA from commercial extracts (extracted with QIAamp DNA Minikit) and DNA from whole mites (extracted with DNAzol), was amplified as previously described using high fidelity PCR Supermix (Invitrogen). Sequencing was performed on PCR product from D. farinae extracts (5 different lots) and 44 whole mite DNA preparations. The sequences were analyzed for homologies by BLAST. RESULTS: Among the mite extracts, bacterial DNA was identified in 6/13 D. farinae extracts but 0/14 D. pteronyssinus extracts. Among the non-mite extracts, bacterial DNA was identified in only 1/10 (German cockroach). Among the non-Dermatophagoides mite preparations, bacterial DNA was detected in all five. However, Bartonella homologies were detected only in C. arcuatus. CONCLUSIONS: Consistent with our previous findings with whole mites, bacterial DNA may be detected in D. farinae but not D. pteronyssinus allergen extracts. In addition, bacterial DNA of different homologies are detected in other mite species.

Multiplex microbead flow cytometric technology has been used to identify allergies in individuals by binding recombinant allergens to bead surfaces and incubating these beads in the presence of patient IgE. Others have bound antibodies to the beads to detect a wide variety of proteins, such as signal transduction molecules, cytokines, and viral or bacterial surface antigens. In principal, this technology may be used to determine the potency and composition of complex protein mixtures, such as allergen extracts. Four anti-Fel d 1 recombinant antibodies were gerenareted by the phage display method, and were covalently bound to carboxy-labeled beads using EDC and sulfo-NHS. Antibody-bound beads were then used to detect levels of Fel d 1 in commercial cat hair extracts using a bead-based flow cytometry assay (Bioplex, Biorad). A bead-based flow cytometric potency assay using anti-Fel d 1 recombinant antibodies yields potency numbers within the range of potencies determined using a radial immunodiffusion assay by the manufacturers and CBER. Three lots of cat hair tested using this assay had potencies, as determined by the manufacturer, of 14.6, 15.4, and 14.4 Fel d 1 U/mL (95% CI = ± 25%). The potencies obtained by the microbead method were 11.6, 15.9, and 12.4 Fel d 1 U/mL (± 15%), respectively. We have developed a bead-based flow cytometric assay using recombinant antibodies that can accurately determine Fel d 1 levels in cat hair allergenic extracts.

Background Cockroach allergy is an important cause of inner city asthma. To perform valid studies on the diagnosis and treatment of cockroach allergy, biological potencies of test extracts need to be established, and a surrogate in vitro test for biological potency should be chosen.Methods Sixty-two cockroach-allergic adult subjects were recruited for quantitative skin testing with three commercial German cockroach extracts. The intradermal D50 values were determined using linear interpolation, and the biologic potencies were determined from D50 data. The extracts were also analysed for relative potency, using a competition ELISA, and for specific allergen content, using a two-site ELISA.Results Estimates of each extract's D50 were analysable in 48-55 subjects, with D50s between 10.3 and 11.8. All three extracts were bioequivalent using pre-set criteria. The biological potencies of the extracts were 1738-8570 bioequivalent allergy units (BAU)/mL (geometric mean=3300), and these relative potencies were similar to those estimated by competition ELISA and specific allergen content. IgE against cockroach allergens were detected in sera from 34 subjects with analysable D50s, and 17 subjects had IgE directed against specific cockroach allergens. Although the presence of anti-Bla g 5 correlated with the subjects' skin test responses for 2/3 extracts, no single allergen was immunodominant. Antibody responses among the subjects were heterogeneous.Conclusions Although commercial cockroach extracts are relatively low in potency, immunotherapeutic doses should be achievable. Biological potency may be estimated using D50 testing, a combination of specific allergen determinations, or by an overall potency assay such as the competition ELISA.Capsule Summary The biological potency of three German cockroach allergen extracts, determined in an inner city population, was 1738-8570 BAU/mL. No one allergen was immunodominant, and surrogate in vitro testing methods were examined.

RATIONALE: Protein microarrays have been used to identify specific IgE to solid phase allergens on the array surface. However, few papers have been published that utilize solid phase antibodies on the array surface to detect specific proteins in a complex mixture. Here we show that recombinant antibodies can be bound to the microarray surface and used to quantitavely detect allergens in allergen extracts. METHODS: Recombinant scFv antibodies that specifically recognize either nAmb a 1 or rFel d 1 (Clin Exp Allergy 2005;35:1040-1048) were spotted in 5 serial dilutions on nitrocellulose microarray slides using a pin and ring arrayer and blocked with ovalbumin. Three different ragweed or cat hair commercial extracts of predetermined potency (by radial immunodiffusion assay (RID)), as well as the existing CBER standards for cat and ragweed, were then incubated with the array and detected with secondary antibodies bound to HRP. Chemiluminescent signal was captured using a CCD camera and spot intensities were determined using Imagequant software. Relative potencies were determined by parallel line analysis of linear regression data. RESULTS: For the cat extracts, the potencies (in Fel d 1 U/mL) were 19.6 (95% CL = 13.6-28.4), 18.1 (13.9-23.4), and 12.5 (8.4-18.6). For the same extracts, the RID values were 20.5, 19.7 and 12.2, respectively. Ragweed results were similar. CONCLUSIONS: We have developed a microarray assay using recombinant antibodies that can detect Amb a I and Fel d 1 in allergenic extracts with accuracy similar to that of currently used standardized methods.

SummaryRationale The competition ELISA assay is used to determine the potency of US standardized allergen extracts. We have been concerned that the competition ELISA is not sensitive to changes in individual allergen levels. This study was designed to determine the sensitivity of the competition ELISA to detect the specific loss of Bla g 1 and Bla g 2 in cockroach extracts.Methods German cockroach extract E3Cg was made from defatted German cockroaches. New Zealand White rabbits were immunized with rBla g 1 or rBla g 2. Optimal dilutions of anti‐Bla g 1 and anti‐Bla g 2 sera were established by ELISA. E3Cg was selectively depleted of Bla g 1 or Bla g 2 by immunoabsorption with anti‐Bla g 1 or anti‐Bla g 2 attached to Protein G agarose beads. Competition ELISA using pooled human sera, or mixed anti‐Bla g 1 and anti‐Bla g 2 serum, was performed on the depleted extracts, and on depleted extracts reconstituted with rBla g 1 or rBla g 2.Results Unlike pooled human‐allergic IgE sera, anti‐Bla g 1 and anti‐Bla g 2 IgG – in dilutions as low as 10−6, could be used in the competition ELISA to measure the loss of allergen in depleted E3Cg. As little as 0.001 μg/mL of added rBla g 1 and 0.1 μg/mL of added rBla g 2, could be detected.Conclusion The competition ELISA can be highly sensitive to compositional differences in complex allergen mixtures, even when the specific detecting antibody is present in relatively small amounts.

Background Monoclonal antibodies are a valuable tool in the study of allergens, but the technology used in their generation can be slow and labour-intensive. Therefore, we have examined recombinant antibody development by phage-display against single allergens and protein mixtures.Objective We used the avian immunoglobulin system (generated from single V-H and V-L genes) to provide a rapid method for generating highly specific recombinant antibody fragments from a minimal number of animals.Methods A single-chain antibody fragment (scFv) library was generated from a single chicken immunized with model allergens. ScFvs were isolated by phage-display and their properties investigated by ELISA and Western blot.Results Mono-specific scFvs were generated against recombinant Fel d 1 and native Amb a 1. Pannings against yellow jacket venom extracts only yielded clones that reacted with multiple proteins in the venom extract. The scFvs from each panning type were effectively expressed in Escherichia coli and readily purified. Highly specific and sensitive recognition of Fel d 1 and Amb a 1 was demonstrated in ELISA, with scFvs displaying antibody-concentration-dependent absorbance curves down to picogram levels of antibody. The specificity of selected antibodies for their cognate antigen was further confirmed in Western blot analysis, with scFvs directed to either Fel d 1 or Amb a 1 showing no reactivity for the other antigens used in immunization. Anti-Amb a 1 scFvs also mapped Amb a 1-isoform location in Western blot of ragweed extracts separated by 2D SDS-PAGE. DNA sequence analysis of scFvs showed that multiple different clones had been generated against Fel d 1 and Amb a 1. Using two anti-Fel d 1 scFv for ELISA analysis of Fel d 1 content in crude cat pelt extracts, we could produce data which were highly similar (P=0.33 and 0.89 by paired t-test analysis) to those obtained using conventional assays (radial immunodiffusion).Conclusion Phage-display technology may generate multiple allergen-specific recombinant antibody fragments from a single chicken, to allergens from mammalian, plant and insect sources. The resulting antibody fragments are of demonstrable use in allergen identification and quantification, in comparison with standard immunoassays.

Allergen vaccines are complex extracts of natural products, and are used for the diagnosis and treatment of allergic diseases. In the U.S., 19 allergen extracts have been standardized. For these vaccines, the potency is estimated by the skin test responses of highly allergic individuals, and surrogate in vitro tests are established for lot release and quality control. The surrogate tests differ for different extracts. National reference standards to which manufactured lots are compared are maintained at FDA/CBER. Allergen standardization has facilitated the establishment of data-driven release limits.

Background: Allergen extracts prepared from Dermatophagoides farinae contain significantly more endotoxin than Dermatophagoides pteronyssinus extracts, and extracts from both mite extracts contain more endotoxin than pollen extracts. Attempts to culture bacteria from mite cultures have failed to establish the sources of the endotoxin.Objective: To determine the bacterial sources of endotoxin in mite extracts.Methods: Live mites of both species were obtained from 2 sources, DNA was extracted from the mites, and DNA encoding bacterial 16S ribosomal RNA was amplified by using specific primers. The amount of bacterial DNA in each mite DNA sample was determined by quantitative PCR using an internal standard, and sequence homologies were determined from amplifications performed by using a high-fidelity DNA polymerase.Results: DNA from D farinae appeared to contain between 11-fold and 24-fold more 16S ribosomal gene copies than the genomic DNA from D pteronyssinus (P <= .003). Sequence analysis indicated the dominant presence of at least 3 phylogenetic clusters of Bartonella species (henselae, quintana, vinsonii, and grahamii), as well as uncharacterized alpha-proteobacteria, from both D farinae and D pteronyssinus. In a few clones, sequences from Escherichia coli, Pseudomonas species, and Acinetobacter species were also identified.Conclusion: House dust mite DNA contains evidence of Bartonella and other Gram-negative species. These Gram-negative species are likely to be the sources of the endotoxin found in mite allergenic extracts.

RATIONALE: D. farinae dust mite extracts contain more endotoxin than D. pteronyssinus extracts. Initial attempts to identify microbial sources of this endotoxin by standard culture techniques were unsuccessful. In this study, we attempted to detect 16S ribosomal sequences in DNA extracted from both species. METHODS: Live mites were obtained by overnight courier from two sources, washed of growth media, eggs and feces with sterile water in a stainless steel sieve (45 micron mesh), and ground with a dounce tissue grinder with glass pestle. DNA was extracted with DNAzol, and amplified in a Stratagene Robocycler (94°C for 5 min; 30 cycles of 94°C for 1 min, 60°C for 1 min and 72°C for 4 min; and 72°C for 26 min) using Platinum PCR Supermix (Invitrogen) with the following reverse primer: CCCGGATCCAAGCTTACGGTTACCTTGTTACGACTT and these two forward primers: CCGAATTCGTCGACAACAGAGTTTGATCCTGGCTCAG (fD1) and CCGAATTCGTCGACAACAGAGTTTGATCATGGCTCAG (fD2). Sequencing was performed (MWG Biotech) on 40 PCR products (20 with each primer pair). The sequences were analyzed for homologies by BLAST. RESULTS: Of the sequences identified using the fD1 primer pair, 5/10 of the D. farinae and 0/10 of the D. pteronyssinus were gram negative (p=0.03). Of these, 4/5 were bartonella species. Of the sequences identified using the fD2 primer pair, there was no significant difference between the number of bacterial sequences in the two mite species. CONCLUSIONS: Bacterial DNA encoding 16S ribosomal subunits were detected in genomic DNA extracted from both species of mites, with predominant homology to bartonella species. Sequences were detected more often in D. farinae than in D. pteronyssinus.

RATIONALE: The competition ELISA assay is used to determine the relative potency of US standardized allergen extracts. With awareness of the importance of immunodominant allergens in determining immune responses, we have been concerned that the competition ELISA is not sensitive to changes in individual allergen levels. This study was designed to determine the sensitivity of the competition ELISA to detect the specific loss of Bla g 1 in cockroach extracts. METHODS: German cockroach extract E3CG was made from frozen defatted German cockroaches. New Zealand White rabbits were immunized with either rBla g 1 or rBla g 2 (Indoor Biotechnologies). Optimal concentrations of anti-Bla g 1 and anti-Bla g 2 sera were established by ELISA. E3CG was selectively depleted of Bla g 1 by immunoabsorption with anti-Bla g 1-loaded Protein G agarose. Competition ELISA was performed by coating plates with E3CG, blocking, and adding mixed anti-Bla g 1 and anti-Bla g 2 serum containing serial dilutions of depleted E3CG (depE3CG)), or depleted E3CG reconstituted with rBla g 1 (recE3CG). RESULTS: Anti-Bla g 1 and anti-Bla g 2 could be used in the competition ELISA to measure allergens in E3CG. Anti-Bla g 1 (1:500,000) in anti-Bla g 2 (1:20,000) could readily distinguish between E3CG and depE3CG. As little as 1.25 ng/mL of added rBla g 1 could be detected; the addition of 12.5 ng/mL made recE3CG indistinguishable from E3CG. CONCLUSIONS: The competition ELISA can be highly sensitive to compositional differences in complex allergen mixtures, even when the specific detecting antibody is present in relatively small amounts.

RATIONALE:We investigated the use of new methods to rapidly produce novel antibodies for the study of model allergens and complex allergen extracts.Since phage display of recombinant antibodies is a powerful method for generating useful antibodies we examined its application in antibody development against single allergens and protein mixtures.The animal model used in this study was the chicken immune system as their immunoglobulin repertoire is generated from single V H and V L genes, simplifying the antibody library construction process.METHODS: A single chicken (White Leghorn, 6 months old) was immunized with 50 µg Fel d 1, 150 µg Amb a 1 and 65 µg yellow jacket venom per dose (1st dose complete Freund's adjuvant, subsequent doses incomplete Freund's adjuvant).The spleen and bone marrow of the chicken were then harvested, RNA was isolated and RT-PCR was performed to generate an scFv library.Single-chain antibodies were then isolated by phage-display.The specificity of the antibodies was determined by ELISA and western blot analyses (1D and 2D).RESULTS: Mono-specific antibody fragments were generated against all target proteins and the scFv products were effectively expressed in E. coli.Purified antibody fragments exhibited highly specific recognition of their relevant proteins in ELISA.All antibodies displayed concentration-dependent absorbance curves to a dilution of 10 -5 (against cognate antigen only).No cross-reactivity was observed between antigens recognized in western blots.CONCLUSIONS: This study describes the efficient generation of allergen-specific recombinant antibody fragments from chicken, to allergens from mammalian, plant and insect sources.