The major grass pollen group 5 allergen from Dactylis glomerata and its C-terminal split product both behave as dimers: implications for allergen standardization.

BACKGROUND: On SDS-PAGE grass pollen group-5 allergens migrate as a doublet with an apparent molecular mass (M(r)) of 25 kDa. Immunoblot analysis revealed additional group 5 reactivity at double and half this M(r). The aim of this study was to investigate these group 5 molecular entities and to compare their allergenicity and behavior in quantitative immunoassays. METHODS: Group-5-specific monoclonal antibodies were produced and used for the development of a group-5-specific sandwich ELISA. Affinity-purified Dac g 5 was separated by SDS-PAGE/Western blotting; individual bands were analyzed by N-terminal sequencing. Size exclusion chromatography (SEC) in conjunction with group-5-specific ELISA, competitive RIA and RAST inhibition were used to analyze the size distribution of Dac g 5. Basophil histamine release assays were used to assess biological activity. RESULTS: The lower band of the typical group 5 doublet was identified as a truncated form lacking the typical group 5 N-terminus AD(L)/(A)GY, observed in the upper band. The 12-kDa peptide was shown to be the C-terminal half of Dac g 5 (amino acid 127 onwards). SEC in conjunction with competitive RIA revealed that around 45% of Dac g 5 is represented by the 12-kDa peptide. Both the C-terminal half and the whole allergen dimerize under nondenaturing conditions. In competitive RIA and RAST inhibition both forms are equally well detected. In contrast, the half molecule is poorly recognized in sandwich ELISA and displays negligible biological activity in basophil histamine release tests with purified IgE. CONCLUSIONS: These observations stress the need to evaluate the performance of allergen standardization protocols in detail, with special attention to allergen size distribution.

Characterization of natural Dac g 1 variants: an alternative to recombinant group 1 allergens.

BACKGROUND: Production of soluble correctly folded recombinant group 1 allergens has proven to be difficult. Purified natural group 1 allergens could be an alternative for application in immunotherapy. OBJECTIVE: Cloning and expression of recombinant Dac g 1; purification of natural Dac g 1 variants and immunochemical characterization of these molecules. METHODS: Dac g 1 was cloned and expressed in the yeast Pichia pastoris. Hydrophobic interaction (HIC), size exclusion, and/or affinity chromatography were used to purify Dac g 1 from Dactylis glomerata pollen extract. Dac g 1 variants were analyzed by N-terminal sequencing. Immune reactivity was assessed by sandwich ELISA, competitive RIA, RAST (inhibition), and in vitro basophil histamine release tests. RESULTS: Dac g 1 was cloned, revealing up to 98% amino acid sequence homology to other group 1 allergens. Purification of natural Dac g 1 revealed at least 3 variants, with an apparent molecular mass (Mr) on SDS-PAGE of 33 kd (HMr), 30 kd (IMr) and 28 kd (LMr). Extraction of IMr Dac g 1 required 0.9% saline, whereas the other 2 variants were also extractable in water. The N-terminus of HMr and IMr Dac g 1 differs at 2 positions, and LMr Dac g 1 was shown to be N-terminally truncated, lacking the first 30 amino acids. The nonretarded fraction of HIC commonly used in group 1 purification protocols does not contain this LMr molecule. IMr Dac g 1 was poorly recognized in 2 of 3 sandwich ELISAs and competitive RIA but demonstrated similar biological activity compared with HMr Dac g 1. CONCLUSIONS: Natural Dac g 1 variants can be separated by extraction of pollen in the presence or absence of saline followed by HIC and size exclusion chromatography. Thus, purified Dac g 1 is an alternative to recombinant group 1 allergens.

Substitution of Pichia pastoris-derived recombinant proteins with mannose containing O- and N-linked glycans decreases specificity of diagnostic tests.

BACKGROUND: Recombinant proteins from Pichia pastoris need to be fully evaluated before used as diagnostic tools. OBJECTIVE: The objective of this study was to investigate whether glycosylation by P. pastoris interferes with the specificity of diagnostic tests. METHODS: An autoantigen involved in Wegener's disease (protease 3) and 2 major inhalant allergens from grass pollen (Dac g 5) and house dust mite (Der p 1) were produced as recombinant molecules in P. pastoris. O-linked glycans on Dac g 5 were characterized by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. The immune reactivity of the recombinant proteins was compared to that of their natural counterparts by ELISA and a radio-allergosorbent test (RAST) as well as by ELISA and RAST inhibition. RESULTS: In contrast to the non-glycosylated natural allergen, recombinant Dac g 5 was shown to carry at least 2 small mannose-containing O-glycans. We showed that both these O-glycans and the N-linked glycans on recombinant protease 3 and recombinant Der p 1 were recognized in ELISA by IgG antibodies in sera of healthy individuals. These IgG responses were closely correlated. The natural autoantigen and allergens were not recognized by IgG antibodies from healthy subjects. The carbohydrate nature of the epitopes recognized by IgG on the recombinant proteins was confirmed by inhibition studies with mannose and yeast mannan. IgE recognition of yeast glycans was observed in 2 out of 9 positive sera from patients with allergic bronchopulmonary aspergillosis. CONCLUSION: Production of recombinant molecules in yeast (or moulds) can introduce IgG-binding glycans that negatively affect the specificity of diagnostic tests.

Lack of correlation between bronchial late allergic reaction to Dermatophagoides pteronyssinus and in vitro immunoglobulin E reactivity to histamine-releasing factor derived from mononuclear cells.

BACKGROUND: Activity of immunoglobulin (Ig)E-dependent histamine-releasing factor (HRF) is dependent on the IgE molecules bound to the surface of basophils. Sera capable of passively sensitizing basophils to release histamine to HRF were designated IgE+ sera. IgE+ and HRF have been suggested to play a role in late allergic reaction (LAR). OBJECTIVE: The working hypothesis was tested that IgE+ induces a LAR. Further, activity of HRF produced by mononuclear cells (HRF(mn)) was compared with that of recombinant HRF p23. METHODS: Atopic patients (n = 82) were bronchially provoked with Dermatophagoides pteronyssinus extract and the change in forced expiratory volume in 1 second was monitored. A LAR was defined as forced expiratory volume in 1 second as percentage of baseline < 80% 4 to 10 hours after allergen challenge. The presence of HRF-responsive IgE in serum was determined using basophils sensitized in vitro by serum. RESULTS: The presence of HRF(mn)-responsive IgE (IgE(mn+)) in serum was shown not be essential for a LAR: 63% of the patients with a LAR had no IgE(mn+) in their serum. Further, 71% of patients with IgE(mn+) did not have a LAR. HRF(mn) and recombinant HRF p23 were not equivalent in the bioassay: serum of 38 of 82 atopic patients sensitized basophils to release histamine to HRF(mn), whereas this was found with serum of 1 of 82 patients to HRF p23. CONCLUSIONS: The results do not support the hypothesis that IgE(mn+) induces a LAR, but do not exclude the alternative hypothesis that HRFs are released during a LAR and contribute to asthma severity.

Studies on the association between immunoglobulin E autoreactivity and immunoglobulin E-dependent histamine-releasing factors.

It has been reported that serum immunoglobulin E (IgE) from certain atopic patients can sensitize basophils to release histamine in response to IgE-dependent histamine-releasing factors (HRFs). It has also been shown that patients suffering from severe forms of atopy may contain IgE autoantibodies. It was investigated whether HRF-responsive sera contained IgE autoantibodies and if there was an association between IgE autoreactivity and IgE-dependent responsiveness to HRF. The presence of HRF-responsive IgE (IgE+) in serum of patients with respiratory atopy was determined by stimulating stripped human basophils sensitized by serum with peripheral blood mononuclear cell (PBMC)-derived HRF, and measuring the release of histamine. In parallel, these sera were screened for the presence of IgE autoantibodies to nitrocellulose-blotted human cellular extracts. The capacity of IgE autoantigen-containing preparations to induce histamine release was tested in the stripped basophil assay. Eleven out of 52 sera contained IgE autoantibodies to blotted cellular extracts of human PBMCs or of the human epithelial cell line A431. No significant association was found between IgE autoreactivity and IgE-dependent responsiveness to HRF: 7/26 IgE+ sera contained IgE to human cellular extracts, and 4/26 of the sera without IgE+ did also. IgE autoantigen-containing extracts did not induce histamine release of appropriately sensitized basophils. By size-exclusion chromatography it was shown that a 32 000 MW autoantigen eluted in the >55 000 MW fraction, which indicates that this protein forms polymers or complexes with other macromolecules. This might explain the discrepancy between binding and histamine-releasing activity. A 20 000 MW IgE-defined autoantigen cross-reacted with a shrimp allergen. Our results indicate that IgE-reactivity to immunoblotted human protein and IgE-dependent HRF activity are distinct entities that may co-occur in atopic patients.

Maturation of Pichia pastoris-derived recombinant pro-Der p 1 induced by deglycosylation and by the natural cysteine protease Der p 1 from house dust mite.

The mature cysteine protease from Dermatophgoides pteronyssinus, Der p 1, is a major house dust mite allergen. Its enzymatic activity has been shown to have pro-inflammatory effects that could also negatively influence efficacy of allergen-specific immunotherapy. The aim of this study was to express recombinant pro-Der p 1 (rpro-Der p 1) in the yeast Pichia pastoris and to study its maturation. Expression was achieved at a concentration ranging from 45 mg.L-1 (methanol-induced expression) to 168 mg.L-1 (constitutive expression). No significant spontaneous maturation of the secreted proenzyme was observed. rpro-Der p 1 with a sequence-based molecular mass of 34 kDa was hyperglycosylated by the yeast, migrating at 50-60 kDa on SDS/PAGE. Compared with its natural counterpart (nDer p 1), the recombinant proenzyme demonstrated decreased IgE reactivity, resulting in a 30-fold lower capacity to induce histamine release from human basophils. Decreased immunoreactivity was also shown by competitive RIA and sandwich ELISA with Der p 1-specific antibody reagents. CD spectra of rpro-Der p 1 and nDer p 1 revealed significant structural differences. Deglycosylation of rpro-Der p 1 with endoglycosidase H resulted in a decrease in apparent molecular mass from 50 kDa to 34 kDa, but did not affect nDer p 1. On removal of N-glycans from rpro-Der p 1, which harbours two putative N-glycosylation sites in both propeptide and mature sequence, the mature rDer p 1 appeared. This suggests that hyperglycosylation hampers spontaneous maturation. Maturation of the recombinant pro-enzyme was also achieved by addition of the active natural cysteine protease, nDer p 1. In conclusion, high-level expression of rpro-Der p 1 in P. pastoris results in a stable hypoallergenic proenzyme with potential for use in allergen-specific immunotherapy.