Patent Grant
7279295
The present invention relates to a novel allergen from timothy grass (Phleum pratense) pollen, Phl p 11, and use thereof as a reagent and in a diagnostic kit as well as for immunotherapy.
A hallmark of atopic allergy is the formation of IgE antibodies to proteins present in the sensitizing biological material. Upon contact with the allergen source, these proteins will act to crosslink IgE antibodies present on the surface of mast cells, thereby eliciting the release of inflammation mediators such as histamine. As a result, an allergic reaction occurs (1).
In the industrialized world, up to 10% of the human population shows allergic sensitization to grass pollen, making this one of the most important airborne allergen sources (2). Considerable efforts have been made towards the characterization of pollen allergens from a variety of grass species using biochemical and immunological methods. A number of IgE binding proteins have thus been identified which exhibit conserved structure and serological cross-reactivity between species. Based on these criteria, such immunologically related grass pollen allergens have been assigned to groups designated by numbers. These include group 1, group 2/3, group 4, and group 5 allergens, which are represented in pollen of most grass species (3).
To date, six different allergens from timothy grass (Phleum pratense) pollen have been cloned: Phl p 1 (4, 5), Phl p 2 (6), Phl p 5 (7-9), Phl p 6 (10, 11), Phl p 7 (12), Phl p 12 (profilin) (13), and Phl p 13 (14). These allergens have all been produced as recombinant proteins which, by different in vitro and in vivo activity assays, have been shown to share immunological and allergenic properties with their native counterparts.
Using a panel of four recombinant allergens (rPhl p 1, rPhl p 2, rPhl p 5, and profilin) in serological and skin testing procedures, positive results were obtained in 95% of a large population of grass pollen-allergic individuals (15). Sensitization to allergens such as Phl p 7 (a calcium-binding, two-EF-hand protein) and Phl p 12 occurs in a smaller proportion of grass pollen allergics, but they share IgE epitopes with homologous proteins present in pollen of trees and weeds and can therefore cause immediate-type symptoms in sensitized individuals upon contact with these unrelated allergen sources (12, 16, 17).
The present invention relates to a novel allergen from timothy grass (Phleum pratense) pollen, Phl p11. While the absolute majority of grass pollen allergics produce IgE antibodies binding to group 1 and group 5 grass pollen allergens—Phl p 1 and Phl p 5 in the case of timothy grass pollen, a subset of the patients also make IgE antibodies to a variety of other protein components such as profilin, Phl p 7 or Phl p 11. These patients are thought to have a more expansive immunological activity in their allergic disease, putting them at greater risk of developing clinical reactivity to an increasing number of allergen sources (such as weed and tree pollens, cat and dog dander, mites, etc).
Profilin and Phl p 7 are highly cross-reactive (vegetables, fruits and pollens of weeds and trees) and not specifically indicative of sensitisation to the grass pollen proteins. Antibody binding structures present on Phl p 11, on the other hand, appear to be specific to grass pollen and may therefore be regarded as a marker of multivalent sensitisation to this particular allergen source, i.e. a marker of enhanced overall propensity to produce IgE antibodies to environmental substances. Recombinant Phl p 11 may be used to identify the subset of patients which are sensitized to polypeptide structures of Phl p 11 and its homologues in pollen of other grass species. About half of all patients that show IgE reactivity to this pollen protein reportedly have their antibodies directed to glycan structures present on the group 11 allergen. These glycan structures are cross-reactive in nature and antibodies binding to them may not be informative/diagnostic in relation to grass pollen-specific sensitisation. IgE reactivity to rPhl p 11 may thus be informative with respect to state of immunological diversification of patients' allergic sensitisation to grass pollen, independent of other allergen sources.
As a defined immunotherapy reagent, recombinant Phl p 11 may be used to treat specifically sensitised patients by way of eliciting a protective or attentuating immune response. The active substance may be a protein comprised of either the natural (“wild-type”) polypeptide sequence or a derivative with improved safety or efficacy properties.
Thus, in a first aspect the invention relates to a reagent which may be recombinantly produced or chemically synthesized, Phl p11, comprising the amino aid sequence according to SEQ ID NO 1 of the enclosed sequence listing as well as essentially homologous (such as 75%) and cross-reactive variants and derivatives thereof. These variants have equivalent or similar function with respect to antibody binding. The invention also relates to DNA sequences encoding said amino acid sequence.
In a second aspect, the invention relates to a diagnostic kit comprising the above reagent. The diagnostic kit may also comprise one or more other known Phl allergens, such as Phl p 1, Phl p 2, Phl p 4, Phl p 5a, Phl p 5b, Phl p 6, Phl p 7, Phl p 12 and Phl p 13.
In a third aspect, the invention relates to an immunoassay comprising the following steps:
The immunoassay may comprise the reagent Phl p 11, either natural or recombinantly produced or chemically synthesized. The immunoassay may be in any desired format, such as ELISA.
In a fourth aspect, the invention relates to use of the above reagent or a derivative thereof for production of a drug for immunotherapy (“allergy vaccination”) of grass pollen allergic patients showing IgE antibody reactivity to Phl p 11. Preferably, the use is for immunotherapy (“allergy vaccination”) of timothy grass pollen allergic patients showing IgE antibody reactivity to Phl p 11.
Figure Legends
General Reagents, Plasmids, Oligonucleotides, Bacterial Strains and Antibodies
Salts and buffers were purchased from Sigma (St. Louis, Mo.) and Fluka (Buchs, Switzerland). Pollen from timothy grass (Phleum pratense) was obtained from Pharmacia Allergon AB (Välinge, Sweden). Protein analysis by SDS-PAGE was performed using 4-20% Tris-glycine gels (Novex, San Diego, Calif.) and for electroblotting Hybond-C Extra membrane (Amersham Life Science, Amersham, UK) was used. For immunoblot analysis of IgE binding, rabbit anti-IgE antiserum (MLAB, Uppsala, Sweden) and horseradish peroxidase-conjugated donkey anti-rabbit IgG (Amersham Life Science) were used, followed by ECL detection (Amersham Life Science). Preparation of polyadenylated RNA from total RNA and subsequent synthesis of cDNA for RT-PCR were performed using the mRNA Purification Kit and the First-strand cDNA Synthesis Kit, both from Amersham Pharmacia Biotech (Uppsala, Sweden). Plasmids pET-23a(+) and pMAL-c2 were purchased from Novagen (Madison, Wis.) and New England Biolabs (Beverly, Mass.), respectively. Restriction endonucleases EcoRI, HindIII, NdeI and XhoI, as well as Taq DNA polymerase and deoxynucleotides were from Amersham Pharmacia Biotech. Pfu DNA polymerase was purchased from Stratagene (La Jolla, Calif.). DNA from PCR and other enzyme reactions was purified using appropriate Wizard kits from Promega (Madison, Wis.). For solid phase capture of biotinylated PCR products, streptavidin-modified magnetic beads (M-280) from Dynal AS (Skøyen, Norway) were used. For large-scale plasmid preparation, the Plasmid Maxi Kit from Qiagen (Düisseldorf, Germany) was used. Oligonucleotides were obtained from Scandinavian Gene Synthesis (Köping, Sweden). DNA sequencing was performed using the T7 Sequencing Kit from Amersham Pharmacia Biotech and [α-35S]dATP from Amersham Life Science. The E. coli strains used were XL1-Blue MR (Stratagene) for cloning purposes and BL21 (Novagen) harboring plasmid pT7POL23 (18) for expression. HiTrap Chelating columns (Amersham Pharmacia Biotech) were used for immobilized metal ion affinity chromatography (IMAC). Buffer exchange and size exclusion chromatography of protein preparations were performed using an FPLC system and columns packed with Sephadex G-25 and Superdex 75, respectively (Amersham Pharmacia Biotech). Quantitative serology for the recombinant allergen was established using Pharmacia CAP System (Pharmacia Diagnostics), employing reagents and procedures as recommended by the supplier. For IgE detection in immunoblot inhibition experiments, an 125I-labeled anti-human IgE antibody from Pharmacia Diagnostics was used. Histamine release from isolated granulocytes of allergic and healthy individuals was measured by a radioimmunoassay (Immunotech, Marseille, France). As a positive control for histamine release capacity of cells, the monoclonal anti-IgE antibody E124.2.8 Dε2 (Immunotech) was used. Histamine and sodium chloride solution for skin prick tests were obtained from ALK (Hørsholm, Denmark).
Patient Samples
A total of 188 grass pollen-allergic subjects or serum samples were examined in this study. One hundred and fifty serum samples were from an in-house collection at Pharmacia Diagnostics, selected on the basis of IgE sensitization to P. pratense. Thirty-eight subjects were from a Vienna clinic and were characterized by case history indicative of grass pollen allergy, positive RAST result for timothy grass pollen, and positive skin prick test to grass pollen extract. The allergen sensitization profiles of these subjects were established with natural and recombinant timothy grass pollen allergens as described (19). Serum samples from two non-allergic individuals were included for control purposes.
Protein Extracts, SDS-PAGE and Immunoblot Analysis
Phleum pratense pollen was extracted at room temperature for 2 hrs in 5 mL of distilled water per gram of pollen. After centrifugation for 5 min at 13 000×g, the clear supernatant was divided into small aliquots and stored at −20° C. until use. The pollen extract was subjected to reducing SDS-PAGE and either stained with Coomassie Brilliant Blue or electroblotted onto nitrocellulose membrane. Protein blots were blocked for 1 hr at room temperature using either 1% (v/v) Tween-20 in PBS (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4, 1.4 mM KH2PO4) or 5% (w/v) defatted dry milk in PBS and then incubated overnight with patient serum diluted five-fold in PBS containing 0.1% Tween-20. After washing in the same buffer, bound IgE was visualized using a rabbit anti-IgE antiserum followed by horseradish peroxidase-conjugated donkey anti-rabbit IgG and ECL detection.
Protein Sequencing
An IgE-binding protein band corresponding to Phl p 11 was identified by immunoblotting using an essentially monoreactive serum sample. The band was excised from a Coomassie Brilliant Blue-stained SDS-polyacrylamide gel, homogenized and extracted in 6 M guanidinium hydrochloride. After removal of polyacrylamide fragments by centrifugation, the extracted protein was subjected to 20 cycles of sequencing from the N-terminus using a Hewlett-Packard G1000A instrument.
Cloning and Characterization of Phl p 11 cDNA
First-strand cDNA was synthesized from purified poly-A+ RNA using the primer 5′-CCA GTG AGC AGA GTG ACG AGG ACT CGA GCT CAA GC(T)18-3′ (QT; SEQ ID NO:19). For 3′-PACE, the two nested specific forward primers GSP-1 (5′-CAT TAC ATA TGG ACA AGG GCC CSG GCT TCG TSG TSA C-3′; SEQ ID NO:15) and GSP-2 (5′-CAT GAA TTC GGA CGC GTC TAC TGC GAC-3′; SEQ ID NO:20) were used, together with the two nested universal reverse primers Qo (5′-CCA GTG AGC AGA GTG ACG-3′; SEQ ID NO:21) and Q1 (5′-GAG GAC TCG AGC TCA AGC-3′; SEQ ID NO:22). Primers GSP-1 and GSP-2 were designed from the N-terminal amino acid sequence of the Phl p 11 protein while primers Qo and Q1 were identical to adjacent parts of cDNA synthesis primer QT.
First-strand cDNA was synthesized from purified poly-A+ RNA using the primer 5′-CCA GTG AGC AGA GTG ACG AGG ACT CGA GCT CAA GC(T)18-3′ (QT). For 3′-RACE, the two nested specific forward primers GSP-1 (5′-CAT TAC ATA TGG ACA AGG GCC CSG GCT TCG TSG TSA C-3′) and GSP-2 (5′-CAT GAA TTC GGA CGC GTC TAC TGC GAC-3′) were used, together with the two nested universal reverse primers QO (5′-CCA GTG AGC AGA GTG ACG-3′) and QI (5′-GAG GAC TCG AGC TCA AGC-3′). Primers GSP-1 and GSP-2 were designed from the N-terminal amino acid sequence of the Phl p 11 protein while primers QO and QI were identical to adjacent parts of cDNA synthesis primer QT.
To generate a Phl p 11-enriched template for 3′-RACE, second-strand cDNA was synthesized by 40 cycles of primer extension of biotinylated GSP-1 using first-strand cDNA as template. The cycling profile used was: 95° C./5 min, followed by 40 cycles of 95° C./60 sec, 58° C./60 sec, 72° C./90 sec. The product of this reaction was then immobilized on streptavidin-modified magnetic beads and washed with 0.1 M NaOH and TE (10 mM Tris-HCl, 1 mM EDTA, pH 8.0). In the first round of 3′-RACE, a sample of immobilized second-strand cDNA and primers GSP-1 and QO were used in the cycling profile: 95° C./5 min, followed by 40 cycles of 95° C./1 min, 72° C./2 min. One μl of a 20-fold dilution of this reaction was used as template together with primers GSP-2 and QI in the second round of 3′-RACE, with the cycling profile: 95° C./5 min, followed by 30 cycles of 95° C./60 sec, 58° C./60 sec, 72° C./90 sec. The GSP-2 and QI primers were designed to incorporate EcoRI and HindIII sites, respectively, at the ends of the amplification product. After purification and cleavage with these enzymes, the product was cloned between the EcoRI and HindIII sites of pBR322 (22). Five candidate clones were subjected to DNA sequencing, revealing a single open reading frame corresponding to Phl p 11.
Amplification of full-length Phl p 11 coding sequence from immobilized second-strand cDNA was performed using the GSP-1 primer and the reverse primer PP 11/R-X (5′-AGT CAC TCG AGT GGC GTC TCG GGG GCG TC-3′; SEQ ID NO:18), which was based on the 3′ end of the Phl p 11 open reading frame. These two primers were designed to incorporate terminal Ndel and Xhol sites, respectively, in the PCR product. The then-nocycling profile used in this reaction was the same as that in round two of the 3′-RACE experiment. The amplification product, purified and digested with Ndel and XhoI, was cloned between the Ndel and Xhol sites of a pET-23a(+)—derivative designed for expression of the gene of interest as a fusion to the maltose binding protein (MBP) of E. coil. The resulting full-length construct for expression was verified by DNA sequencing.
Expression and Purification of rPhl p 11
The P. pratense allergen was expressed in E. coli as a fusion to MBP. Plasmid DNA from one selected clone was introduced into strain BL21 harboring plasmid pT7POL23 which provides T7 RNA polymerase in a stringently controlled, temperature-dependent manner (18). LB medium (10 g/L tryptone, 5 g/L yeast extract, 5 g/L NaCl, adjusted to pH 7.0 using 1 M NaOH) was inoculated 1:500 with an overnight culture and first grown at 30° C. to mid-log phase. The incubation temperature was then raised to 42° C. for 1 hr, followed by 4 hrs at 30° C. before harvest. Cells were collected by centrifugation at 10 000×g for 10 min at 4° C. and resuspended in 5 mL of buffer A (20 mM Tris-HCl pH 8.0, 0.5 M NaCl, 100 mM β-mercaptoethanol, 5 mM imidazole) per gram (fresh weight) of cells. The resuspended cells were ruptured by sonication while kept on ice, followed by centrifugation to remove solid material. Following exchange to buffer B (20 mM Tris-HCl pH 8.0, 0.5 M NaCl, 5 mM β-mercaptoethanol) containing 5 mM imidazole using Sephadex G-25, the supernatant was loaded onto a Ni2+-charged 5 mL HiTrap Chelating column for IMAC. The column was washed with 20 mM imidazole in buffer B and elution was performed with a 20-250 mM gradient of imidazole in buffer B. Fractions containing the eluted fusion protein were pooled and subjected to a final step of size exclusion chromatography through Superdex 75, equilibrated with non-reducing buffer, to obtain a homogeneous, unaggregated preparation without visible contamination by E. coli proteins. To serve as a negative control in functional studies, MBP alone was expressed from BL21[pT7POL23] cells harboring the expression vector without insert and the protein purified as above, except that buffer B containing 5 mM imidazole was used in place of buffer A at the stage of cell homogenization. The concentration of MBP-Phl p 11 and MBP in the final preparations was determined from their absorbance at 280 run, using calculated extinction coefficients of 1.30 and 1.47 per mg/mL, respectively.
Assessment of IgE-Binding Activity of rPhl p 11 Using Pharmacia CAP System
In vitro IgE-binding activity of the purified recombinant allergen was examined in Pharmacia CAP System, an immunoassay system used for IgE antibody detection in clinical diagnosis of atopic allergy. Experimental ImmunoCAP tests were prepared by covalent immobilization of the purified allergen onto activated cellulose at a concentration chosen to achieve an adequate linear measuring range and a background for negative sera adequately below the conventional cut-off value of 0.35 kUA/L. Negative control tests carrying MBP alone were prepared using the same protein concentration at immobilization. For determination of specific IgE to the whole complement of natural P. pratense pollen proteins, the regular pollen extract-based ImmunoCAP test was used. For the purpose of comparison to a previously established recombinant allergen, all serum assays were run in parallel with rPhl p 2 ImmunoCAP tests. Assay controls and calculation of statistical parameters attesting to the quality of the assays were performed using standard assay system routines and software (Pharmacia Diagnostics).
Immunoblot Analysis of IgE Binding Properties of rPhl p 11
The proportion of timothy grass pollen-specific IgE directed against rPhl p 11 and rPhl p 5 was investigated by a RAST inhibition-based experiment. Serum samples from 10 rPhl p 11-reactive subjects were diluted 1:10 in buffer C (50 mM sodium phosphate pH 7.5, 0.5% (v/v) Tween 20, 0.5% (w/v) BSA, 0.05% (w/v) NaN3) and preadsorbed overnight at 4° C. with either rPhl p 11, MBP (negative control) or rPhl p 5 (positive control), all at a final concentration of 10 μg/mL. To ensure conditions of antigen excess on the solid phase, approximately 0.2 mg of natural timothy grass pollen protein extract was immobilized to nitrocellulose strips of exactly the same size (0.6×3 cm). Strips were blocked by preincubation with buffer C (once for 1 hour and twice for 5 minutes) and then exposed to the preadsorbed sera at 4° C. overnight. The following day, strips were washed four times in buffer C and then probed with 125I-labeled anti-human IgE antibody at room temperature overnight. Strips were washed again four times in buffer C and dried. The amount of 125I-labeled anti-human IgE antibody was determined using a gamma counter (Wallac, Turku, Finland). The percentage inhibition of IgE binding after preincubation of sera with rPhl p 5 or rPhl p 11 was calculated as follows: % inhibition=100−100×(cpm rPhl p 5/cpm MBP or rPhl p 11/cpm MBP).
The capacity of the recombinant allergen to bind Phl p 11-specific IgE antibodies was studied by IgE immunoblot inhibition experiments (17). Sera from two grass pollen allergic subjects with IgE reactivity to rPhl p 11 were preadsorbed with purified rPhl p 11 at 10 μg/mL serum, or, for control purposes, with an equal concentration of MBP or BSA. Preadsorbed sera were exposed to nitrocellulose-blotted timothy grass pollen proteins separated by SDS-PAGE and bound IgE was detected as described (17).
Histamine Release Experiments
Granulocytes were isolated by dextran sedimentation of heparinized blood samples (23, 24) from two grass pollen allergic and one non-allergic individuals. Aliquots of washed cells were incubated with a range of concentrations (0.001 μg/mL, 0.01 μg/mL, 0.1 μg/mL, 1 μg/mL) of purified rPhl p 11, MBP, and a monoclonal anti-IgE antibody. Histamine released in the supernatant was measured by radioimmunoassay. Total histamine was determined after freeze-thawing of cells. Results were displayed as mean values of triplicate determinations and represent the percentage of total histamine.
Skin Testing
After informed consent was obtained from two grass pollen-allergic and four non-allergic individuals, skin prick tests were performed on their forearms as described (25). Individuals were pricked with 20 μl aliquots of solutions containing different concentrations (0.1 μg/mL, 1 μg/mL, 10 μg/mL, 100 μg/mL) of purified rPhl p 11 and MBP, and with timothy grass pollen extract, histamine and sodium chloride. The skin reactions were recorded 20 minutes after sample application by photography and by transferring a ballpoint pen-tracing of the wheal area to paper using adhesive tape. Mean wheal diameters (Dm) were determined as follows: Dm=0.5×(D1+D2) where D1 and D2 represent the largest longitudinal and transverse diameters in mm, respectively.
Results
Immunochemical Detection, Isolation and Protein Sequencing of Natural Phl p 11
Immunoblot analysis of serum from a grass pollen-allergic subject, which lacked IgE antibodies to all purified or recombinant allergens from P. pratense currently available (rPhl p 1, rPhl p 2, nPhl p 4, rPhl p 5, rPhl p 6, rPhl p 7, and rPhI p 12), revealed predominant IgEbinding to a single protein band at approximately 20 kDa. One faint band in Coomassiestained SDS-PAGE aligned perfectly with the IgE-reactive band in the immunoblot analysis (
cDNA Cloning and Sequence Analysis of Phl p 11
After back-translation of the N-terminal amino acid sequence into DNA, using the codon preference seen in other genes expressed in P. pratense pollen, two nested forward PCR primers (GSP-1 and GSP-2) were designed for use in 3′-RACE and RT-PCR. First-strand cDNA was synthesized from a poly-A+ RNA preparation, using a universal oligo-dT primer carrying terminal target sequence for two nested reverse PCR primers, QO and QT, to be used in subsequent steps of amplification. Specifically enriched second-strand cDNA, generated by 40 cycles of primer extension of GSP-1 on first-strand cDNA, was used as template in the first round of 3′-RACE, carried out with primers GSP-1 and QO. In a second round, 1/1000 of the first round reaction was used as template together with primers GSP-2 and QI. Analysis of this reaction by agarose gel electrophoresis revealed two distinct bands of similar intensity, approximately 700 and 800 bp in size (not shown). The use of raised annealing temperature did not change the appearance of the second round 3′-RACE product. The double-band product was therefore tentatively considered genuine and specific. The product was cloned and transformants harboring inserts matching both fragment sizes were identified and analyzed by DNA sequencing. All five clones examined contained inserts of nearly identical sequence and it appeared that the difference in size between the two bands seen after the second round of 3′-RACE was due to alternative sites for priming of cDNA synthesis (
The open reading frame of the cDNA defined a polypeptide of 143 amino acid residues with a calculated isoelectric point of 4.8, a molecular mass of 15.8 kDa and one potential site for N-linked glycosylation (
The most prominent difference in primary structure observed between Phl p 11 and Lol p 11 (sequence accession No. A54002) was a stretch of nine additional amino acid residues (-DLRDAPETP; residues 135-143 of SEQ ID NO:2) at the C-terminus of Phl p 11, equivalent to a 1.0 kDa increment in molecular mass. In comparison to the L. perenne homologue, the Phl p 11 sequence contained a total of six amino acid substitutions, four of which were non-conservative (D42N, K56G, D57L, K83T). At position 103, which was not determined in the case of Lol p 11, an asparagine. residue was present in the Phl p 11 sequence. The two homologues showed conservation of one potential site for N-linked glycosylation (residue 24) and six cysteine residues.
As previously shown by van Ree et al. (26), group 11 grass pollen allergens are structurally related to the soybean trypsin inhibitor and may therefore present antigenic structures similar to proteins belonging to this family. Very recently, structurally related allergens from English plantain, Plantago lanceolata, (Pla 11) and goosefoot, Chenopodium album, (Che a 1) were reported (27, 28).
The discrepancy between the observed apparent MW of the native Phl p 11 allergen by SDS-PAGE and the MW calculated from the deduced amino acid sequence is presumably explained by post-translational modification of the native allergen. In support of this is the report by van Ree et al. (26), where the homologous L. perenne protein was shown to carry N-linked glycosylation amounting to approximately 8% of the total molecular mass, and the conservation of the corresponding glycan attachment site in the amino acid sequence of Phl p 11.
Expression in Escherichia coli and Purification of rPhl p 11
With the aim of allergenic and serological characterization of the Phl p 11 allergen, the protein was expressed in E. coli and purified to homogeneity. Because of poor solubility when the allergen was initially expressed with an N-terminal hexahistidine tag as the only engineered addition, we chose instead to produce it as a fusion to the E. coli maltose binding protein as a means to aid solubility. After preparing a construct where transcription of the fusion was under control of the T7 promoter, using E. coli XL1-Blue as a cloning host, the plasmid was transferred to strain BL21 harboring plasmid pT7POL23. In this binary system the construct is quiescent at 30° C. and recombinant protein expression induced by a temperature shift to 42° C. (
Using this strain for expression, accumulation of MBP-Phl p 11 to approximately 10% of total cellular protein was obtained, as estimated from Coomassie-stained SDS-PAGE (
Analysis of Antibody Recognition of rPhl p 11
To examine the IgE antibody binding capacity of the recombinant allergen and investigate the frequency and magnitude of Phl p 11-specific IgE sensitization among grass pollen-allergics, serological tests were prepared for use in Pharmacia CAP System. As a control for antibody binding to the MBP part of the fusion protein, tests carrying MBP alone were prepared and used in parallel. Upon analysis of serum samples of 184 grass pollen-sensitized subjects using these tests, 59 (32%) of them were found to contain specific IgE reactivity to the recombinant allergen (Table II). The average level of IgE to rPhl p 11 in the specifically reactive sera was 16 kUA/L, as compared to 79 kUA/L of IgE to natural extract of P. pratense pollen. Thus, it appears that on average among these subjects, approximately 20% of the IgE reactivity to P. pratense pollen allergens was directed to rPhl p 11.
In two of the sera that showed a positive result with the rPhl p 11 test, there was also an apparent binding of IgE to MBP alone. For one of these sera the IgE determination was in fact higher with the MBP test, and this serum was therefore regarded as lacking IgE to rPhl p 11. For the other serum, the contribution by MBP to the IgE binding by the fusion protein was only about 1%, which was considered insignificant. In total, only four sera of all 184 tested (2%) showed detectable IgE binding to MBP alone, indicating that MBP may be a suitable fusion partner for recombinant allergen production in instances when a soluble non-fusion protein cannot be efficiently produced in E. coli.
For the purpose of comparison, the 184 serum samples were also tested with an assay specific for a previously established major grass pollen allergen, rPhl p 2. IgE antibody reactivity directed to this allergen was found in 103 (56%) of all tested subjects, with an average IgE level of 11.4 kUA/L. Binding to rPhl p 2 would thereby account for approximately 15% of the total level of IgE to whole, natural extract of P. pratense pollen in this subset of sera, which was 74 kUA/L on average. In summary, the serological analysis shows that the E. coli-expressed rPhl p 11 has significant and specific IgE antibody binding capacity, comparable in frequency and magnitude to that of rPhl p 2.
Inhibition of IgE Binding to Natural Grass Pollen Extract by Soluble rPhl p 11
To compare in a more direct way the IgE binding characteristics of recombinant and natural Phl p 11, an immunoblot inhibition experiment was performed. In this analysis, competition for IgE binding to immobilized natural allergen by soluble rPhl p 11 would be visualized as attenuation of IgE binding to immobilized natural Phl p 11 after preincubation of patient serum with the recombinant allergen. As a control for unspecific inhibition, both serum samples used were preincubated with BSA and MBP in parallel with the rPhl p 11 pretreatment. While the control proteins had no visible effect on IgE binding to extract proteins, as compared to preincubation with buffer (not shown), pretreatment of the serum samples with rPhl p 11 almost completely abolished the autoradiography signal at 20 kDa molecular weight (
The contribution of Phl p 11 to the total IgE binding activity of pollen proteins was further examined by dot blot inhibition experiments in which rPhl p 5, an allergen known for its high IgE binding capacity (7), was used for comparison. Equal amounts of pollen protein extract were spotted onto identical pieces of nitrocellulose membrane and exposed to patients' sera that had been preincubated with either rPhl p 11, rPhl p 5 or MBP. From serological analyses, these sera were known to contain IgE to both Phl p 11 and Phl p 5, but not to MBP. As controls, buffer incubation and serum from one non-allergic individual were used. After washing, membrane-bound IgE was determined radiometrically and the inhibition effects of rPhl p 11 and rPhl p 5 were calculated in relation to the MBP-pretreated samples. The results of the experiment are shown in Table III. On average rPhl p 11 was found to inhibit 25% of the IgE binding to pollen extract, which corresponds to the quantitative serological data shown in Table II, while rPhl p 5 caused an average inhibition of 55%. We conclude that Phl p 11 accounts for a relevant proportion of timothy grass pollen-specific IgE antibodies, although smaller than Phl p 5.
rPhl p 11 Induces Basophil Histamine Release and Immediate Skin Reaction
In an experiment on basophils from a high-level Phl p 11-sensitized allergic individual, rPhl p 11 induced dose-dependent release of histamine, demonstrating its capacity to productively cross-link cell surface-bound IgE antibodies. Limited histamine release occurred from cells of a low-grade Phl p 11-sensitized subject and none from cells of a non-allergic upon incubation with rPhl p 11. (Table 1). Evidence of specific biological activity of rPhl p 11 in vivo was obtained from skin test experiments. In two sensitized subjects, dose-dependent wheal reactions resulted from challenge with a dilution series of the allergen while no reaction occurred in four non-allergic controls tested. (Table 2). The rPhl p 11 fusion partner MBP alone gave rise to no reaction in neither these experiments. Hence, rPhl p 11 exhibited biological activity which fulfilled criteria of specificity.
aRelease of histamine is expressed as percentage of the cells' total content of histamine.
bTwo allergics and four non-allergics were skin tested with rPhl p 11 at different concentrations, with timothy grass pollen extract and histamine. The diameter of the resulting wheal reactions were determined and are presented in mm.
Discussion
Grass pollens belong to the most frequently sensitizing and potent allergen sources. They contain a number of allergenic molecules, several of which have been identified and characterized in the recent past (3). In the present invention we report the identification, cloning and recombinant production of a novel P. pratense pollen allergen which adds new important epitopes to the growing panel of recombinant grass pollen allergens (29).
Despite the fact that the group 11 grass pollen allergens are glycoproteins (26) and contain several cysteine residues, we were able to produce soluble, monomeric and immunologically active rPhl p 11 allergen by utilizing MBP as a fusion partner for expression in E. coli. Extensive serological characterization of IgE reactivity to rPhl p 11 was carried out using a quantitative assay system where allergen is covalently immobilized onto activated cellulose. Using the rPhl p 11-specific tests, we found that about one third of all grass pollen sensitized subjects analyzed (n=184) contained serum IgE antibodies binding to rPhl p 11 and that the magnitude of binding corresponded to a significant proportion of grass pollen-specific IgE antibodies in these subjects.
Evidence supporting the authenticity of epitope presentation by rPhl p 11 was obtained from immunoblot inhibition experiments, where natural grass pollen proteins were attached on solid phase and rPhl p 11 used as fluid-phase inhibitor. Specific and extensive inhibition of IgE binding to the natural allergen occurred in both of two patient sera examined, demonstrating that rPhl p 11 could compete with natural Phl p 11 for IgE antibody binding. Taken together, the serological data show that immunoreactive rPhl p 11 can be produced using E. coli expression and that the recombinant protein shares epitopes for IgE antibodies with the natural allergen. Based on the results obtained, it is clear that rPhl p 11 represents an important addition to the panel of recombinant grass pollen allergens useful for in vitro diagnosis of grass pollen allergy.
Relevant to this discussion is the immunological analysis of chemically deglycosylated natural Lol p 11 reported by van Ree et al. (26), which suggested the involvement of carbohydrate structures in the IgE binding properties of this allergen. Thus, we cannot exclude that a qualitative difference in allergenic properties exists between natural Phl p 11 and the recombinant molecule described in this paper and that expression of Phl p 11 in a glycosylated form, using a eukaryotic host, could yield a recombinant allergen with different IgE binding characteristics. On the other hand, in view of recent notions that glycan epitopes may not be efficient elicitors of IgE-mediated reactions or informative in relation to clinical allergy manifestation (30-34), it is possible that an unmodified recombinant allergen expressed in E. coli is more useful for diagnostic purposes.
Despite the significant sequence homology among the members of the widely represented (grasses, trees and weeds) group of allergens exemplified by Phl p 11 in timothy grass and Ole e 1 in olive tree pollen, little cross-reactivity for IgE antibodies appears to exist between them. In a preliminary analysis, we have been unable to detect cross-reaction between rPhl p 11 and Ole e 1 (Niederberger, Valenta & Lidholm, unpublished data) and the results of recent studies on other members of this allergen family (27, 28) are in agreement with this observation.
One important implication of the apparent lack of significant cross-reactivity between rPhl p 11 and other members of this allergen family is that they are useful as diagnostic markers to more precisely identify the primary sensitizer of allergic individuals, as compared to natural extracts or cross-reactive components such as profilin, two-EF-hand allergens or Bet v 1 homologues. Thus, a preferential IgE recognition of Phl p 11, in relation to other members of this allergen family, may suggest a primary sensitization by grass pollen rather than another allergen source containing cross-reactive components. The use of selected recombinant allergens in this way may provide information useful for advice on allergen avoidance and adequate selection of allergen extract for specific immunotherapy treatment.
In conclusion, the present invention concerns cDNA cloning and recombinant production of an IgE-reactive and biologically active group 11 grass pollen allergen. Recombinant Phl p 11 can be used to identify group 11 allergen sensitization in patients and for specific immunotherapy.
| Number | Date | Country | Kind |
|---|---|---|---|
| 0200946 | Mar 2002 | SE | national |
| Number | Date | Country |
|---|---|---|
| 03047618 | Jun 2003 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 20040014155 A1 | Jan 2004 | US |
