The 026 sdab:IgE Fc complex was concentrated to 5 mg ml−1 in 20 mM HEPES, 50 mM NaCl, pH 7.2, and crystallized by vapor diffusion in sitting drops formed by mixing protein and reservoir solution containing 0.1 M imidazole pH 7.0, 11–12% polyethylene glycol (PEG) 20,000 in the ratio 1:1. Crystals were cryocooled in liquid nitrogen after transfer to a cryo-protectant composed of 0.1 M imidazole pH 7.0, 12% w/v PEG 20,000 and 30% glycerol.
Data collection and structure determination and refinement
Data were collected at Diamond I2445 and processed with XDS46. The structure was determined by molecular replacement with PHASER47 using the pdb ID: 2WQR as search model21. The model was rebuild in Coot48 and refined with Phenix.refine and iMDFF49,50. Figures were prepared with the PyMOL Molecular Graphics System (Schrödinger LLC).
Protein expression and purification
The IgE Fc constant region cDNA was obtained from an IgE expression vector initially constructed from a human cDNA library. The cDNA was introduced into an expression vector providing a human immunoglobulin signal sequence via SmiI and MssI restriction enzymes51. Point mutations were introduced by PCR Human embryonic kidney cells (HEK-293, ATCC) were cultivated in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% (v/v) heat-inactivated fetal calf serum (FCS), 100 IU ml−1 penicillin and 100 μg ml−1 streptomycin. HEK-293 cells were transfected with the expression vector using Nanofectin (GE Healthcare) according to the recommendations of the manufacturer. After selection by addition of zeocin stably transfected cells showed an expression yield of 10–15 mg l−1. Cell culture supernatant obtained in roller flasks (Greiner) was collected and subsequently subjected to purification of the IgE Fc via nickel based affinity chromatography. The supernatant was loaded on a 1 ml HisTrap excel column (GE Healthcare) equilibrated with PBS (500 mM NaCl, 40 mM Na2HPO4, 10 mM NaH2PO4, pH 7.4). After washing with 10 column volumes (CV) of PBS and 20 CV of 5% PBS/imidazole (100 mM NaCl, 40 mM Na2HPO4, 10 mM NaH2PO4, 300 mM imidazole, pH 7.4) the IgE Fc was eluted in a 5–100% gradient of PBS/imidazole. Pooled fractions were dialyzed against Buffer A (20 mM NaOAc, 50 mM NaCl, pH 6.5) and subsequently applied to a 1 ml Mono S 5/50 column (GE Healthcare) equilibrated with Buffer A. After washing with Buffer A, the protein was eluted with 20 CV of Buffer B (20 mM NaOAc, 1 M NaCl, pH 6.5) with a 0–50% gradient. The IgE Fc was further purified by size exclusion chromatography using a 24 ml Superdex 200 20/300 GL column (GE Healthcare) equilibrated in 20 mM HEPES, 50 mM NaCl, pH 7.2.
The DNA of the sdab 026 (patent WO 2012’/175740 A1) was obtained as a synthetic gene and cloned into the bacterial expression vector pET22+ providing a pelB signal sequence and a C-terminal histidine tag (Supplementary Table 3). Expression of the sdab 026 and its variants was performed for 3 h at 30 °C after IPTG induction. The supernatant was subjected to nickel-based affinity chromatography on a 1 ml HisTrap excel column (GE Healthcare) equilibrated with PBS. After washing with 10 CV of PBS and 20 CV of 5% PBS/300 mM imidazole the protein was eluted with a 5 CV gradient from 5–100% of PBS/imidazole. The sdab mutants were purified accordingly. After dialysis against Buffer A (20 mM Tris, 50 mM NaCl, pH 8.5) the sdab was further purified using ion exchange chromatography using a 6 ml Resource Q column (GE Healthcare) equilibrated in Buffer A. After washing with 5 CV of Buffer A, the protein was eluted with a 5 CV gradient from 0–50% Buffer B (20 mM Tris, 1 M NaCl, pH 8.5).
Prior to SAXS data collection IgE Fc was further purified to remove the Cε3–4 fragment by hydrophobic interaction chromatography using a 1 ml Source 15PHE column (GE Healthcare) equilibrated in Buffer A (1.6 M NH3SO4, 20 mM HEPES, pH 7.2). The sample obtained from size exclusion chromatography was dialyzed against Buffer A and applied to the column. After washing with 5 CV of Buffer A, the protein was eluted with a 15 CV gradient from 0–100% Buffer B (50 mM NaCl, 20 mM HEPES, pH 7.2). IgE Fc Cε2–4 from the Source 15Phe purification was mixed with purified 026 sdab and the complex was purified by size-exclusion chromatography as described above.
Characterization of IgE Fc and 026
The binding of 026 sdab to IgE Fc was assessed in ELISA. Purified IgE Fc (50 µg ml−1) was coated on microtiter plates (Greiner) at 4 °C and blocked with 40 mg ml−1 milk powder in PBS. Thereafter, 026 sdab at a concentration of 10 µg ml−1 was incubated for 2 h at RT. After washing, binding to IgE Fc was detected using alkaline phosphatase-conjugated anti-human kappa antibody and anti-human IgG antibody diluted 1:30,000 and 30 µl of substrate solution.
For comparative assessment of anti-IgE antibodies, ELISA microtiter plates were coated with 026 sdab (25 µg ml−1) at 4 °C and blocked with 40 mg ml−1 milk powder in PBS. After incubation with IgE Fc, the anti-IgE antibodies were incubated in a final volume of 20 µl for 4 h at room temperature. After washing bound antibodies were detected using an alkaline phosphatase-conjugated anti-human IgG antibody diluted 1:20,000, and 30 μl of substrate solution (5 mg ml−1 4-nitro-phenylphosphate, AppliChem).
The affinity of the sdab was determined by using a T200 Biacore system. IgE Fc was immobilized to a total of 110 resonance units (RU) onto a CM5 sensor chip (GE Healthcare) using NHS/EDC coupling procedures. Analyses were performed at 25 °C in a buffer containing 10 mM monosodium phosphate, 40 mM disodium phosphate and 100 mM NaCl, pH 7.4 supplemented with 0.01% Tween-20. For kinetic analyses the 026 sdab was injected at concentrations from 0.003 to 50 µg ml−1 at a flow rate of 25 µl min−1. Association and dissociation were assessed for 600 s. Regeneration of sensor surfaces was performed by subsequent injection of 1 M Tris buffer, pH 10. The dissociation constant at equilibrium KD was calculated using a 1:1 binding model and the Biacore T200 evaluation software.
Basophil activation tests and mediator release assays
Peripheral EDTA blood from allergic donors was preincubated with 026 sdab, mu112 or omalizumab (Novartis) at concentrations from 0 to a maximum of 45 µmol l−1 (sdab 026 and mu112) and from 0 to 90 µmol l−1 (omalizumab), respectively, in IL-3-containing stimulation buffer (Bühlmann) for up to 60 min on a shaker. Samples were stained with anti-CCR3-PE in a 37 °C water bath for 15 min and anti-IgE-FITC (both BioLegend) for 20 min at 4 °C. Erythrocytes were lysed for 7 min followed by centrifugation at 500 × g for 5 min. Cell pellets were resuspended in 100 µl washing buffer and the mean fluorescence intensity (MFI) of IgE on CCR3+SSClow basophils was measured by flow cytometry (FACS Calibur, BD Biosciences).
To analyze the impact of IgE026 on basophil activation, peripheral EDTA blood from six patients with birch pollen allergy was incubated ±45 µmol l−1 sdab 026 for 60 min on a shaker. Afterwards, Bet v 1 (Biomay) diluted at different concentrations (0.01–1000 ng ml−1) in stimulation buffer was added. In each case one sample was left unstimulated to exclude background activation. Next, all samples were incubated with 15 µl of staining reagent (Bühlmann) consisting of anti-CCR3-PE and anti-CD63-FITC in a 37 °C water bath for 15 min. As above, erythrocytes were lysed for 7 min followed by centrifugation at 500 × g for 5 min. Cell pellets were resuspended in 100 µl washing buffer and basophil activation was measured by flow cytometry (FACS Calibur, BD Biosciences). Basophils were identified as CCR3+SSClow cells and CD63+ basophils were considered as activated. All results were analyzed using BD FACSDiva software (BD Biosciences).
CDsens analyses were performed as basophil allergen threshold stimulation. Basophil allergen sensitivity was measured as the allergen concentration eliciting 50% (EC50) of maximum CD63 upregulation. The CDsens is defined as inverted value for EC50 multiplied by 100, and was calculated by the following formula: CDsens = 1/EC50 × 10052.
In order to analyze in vitro degranulation RBL-SX38 cells (kindly provided by J.P. Kinet) were sensitized with rIgE Fc53. After washing with incomplete Tyrode’s buffer, receptor cross-linking was performed by incubation with goat anti-human IgE for 60 min at 37 °C. Release of β-hexosaminidase from viable vs. lysed cells was assessed for 60 min at 37 °C using p-nitrophenyl N-acetyl-glucosaminide (Sigma-Aldrich) as a substrate. After stopping the reaction by addition of carbonate buffer (0.1 M, pH 10) the absorbance was recorded at 405 nm.
Analyses of CD23 binding using the ELIFAB assay
In order to evaluate the capability of 026 sdab to inhibit the binding of IgE:allergen complexes to CD23 the ELISA-based ELIFAB assay, a cell-free variant of the FAB assay, was performed32,54. To allow for formation of IgE:allergen complexes 20 μl of indicator serum containing either high Api m 1-, Bet v 1- or Ves v 5-specific IgE concentrations (all >100 kUA l−1) was preincubated with the respective allergen at 37 °C for 1 h in the presence of different 026 sdab concentrations. Following preincubation, IgE:allergen complexes were transferred to plates coated with CD23 (R&D Systems, Bio-Techne) and incubated for 1 h at RT. IgE:allergen complexes bound to immobilized CD23 were detected by adding biotin-conjugated anti-human IgE antibody (BD Biosciences), streptavidin-peroxidase (Sigma-Aldrich), and 3,3′,5,5′-tetramethylbenzidine (TMB) substrate (Calbiochem, Merck Millipore). All samples were analyzed in duplicates.
In a second set of ELIFAB experiments, preformed IgE:allergen complexes bound to CD23 were incubated with 5 µl 026 sdab up to 60 min evaluating the ability of 026 sdab to displace IgE:allergen complexes from CD23. Displacing IgE:allergen complexes from immobilized CD23 was determined at a 026 sdab concentration of 12 μg ml−1 for Api m 1 (Latoxan), 8 μg ml−1 for Bet v 1 (Biomay) and 11 μg ml−1 for Ves v 5, respectively.
Small-angle X-ray scattering data collection and modeling
The SAXS measurements of IgE Fc and IgE Fc in complex with sdab 026 were performed in batch mode at the BM29 beamline at the European Synchrotron Radiation Facility (ESRF), Grenoble, France55. The data were collected using a PILATUS 1 M pixel detector and λ = 0.992 Å in a temperature controlled capillary at 4 °C. The sample-to-detector distance was 2.872 m, covering a range of momentum transfer 0.04 < q < 5 nm−1 (
, where 2θ is the scattering angle). Samples in 20 mM HEPES, 50 mM NaCl, pH 7.2 were investigated in the concentration ranges 0.5–1.7 and 2.2–9.0 mg ml−1 for the IgE Fc:026 sdab complex and IgE Fc, respectively. Data were collected with 10 exposures of 2 s. Radial averaging, buffer subtraction and concentration scaling were performed using the beamline pipeline56. Bovine serum albumin (Sigma, A7638-5GP) used for calculation of the molecular weight (Supplementary Table 1) was solubilized in a buffer containing 50 mM Hepes pH 7.5 at concentration of 5 mg ml−1. For the IgE Fc:026 sdab complex a concentration of 1.7 mg ml−1 was used for further modeling. For IgE Fc the measurements at 2.2 and 9.0 mg ml−1 were merged using ALMERGE57. The pair distribution function was calculated by indirect Fourier Transform using GNOM58. Rigid body refinements were performed using a momentum transfer range of q < 3.0 nm−1 with CORAL59. The 026 sdab and the Fc Cε3 and Cε4 domains were used as a single rigid body taken directly from our crystal structure. The two Cε2 domains from the pdb entry 2WQR were grouped into a single rigid body, and the C-terminal of both of the Cε2 domains were linked to the N-terminal residue of their corresponding Cε3 domain with a distance restraint of 25 Å. Calculation of the angle of rotation between the SAXS models and the extended (RCSB: 4J4P) and bent (RCSB: 2WQR) conformation of IgE Fc was performed using DynDom60. Ensemble optimization of both unbound IgE Fc and the IgE Fc:026 sdab complex was performed using the EOM 2.0 programs RANCH and GAJOE33. A total of 5000 models were generated utilizing the same rigid bodies as for CORAL rigid body modeling, but with a randomly generated alpha carbon trace connecting the two rigid bodies, instead of a distance restrain. GAJOE was then used to select the ensemble of models best representing the scattering curve.
The study was approved by the Ethics Committee of the Medical Faculty of Philipps University, Marburg, Germany; all patients provided written informed consent to participate in the trial.
Coordinates and structure factors have been deposited in the Protein Data Bank under accession code 5NQW. Other data are available from the corresponding authors upon reasonable request.