Recombinant Hypoallergenic Cat Allergy Vaccines

ABSTRACT

Background

Molecular forms of allergen‐specific immunotherapy (AIT) for cat allergy are needed. Fel d 1, Fel d 4, and Fel d 7 are the most important cat allergens.

Methods

IgE epitopes of Fel d 4 and Fel d 7 were mapped by blocking allergic patients' IgE binding with allergen peptide‐specific antisera. Five recombinant fusion proteins (PreS‐Cat 1‐PreS‐Cat 5) each containing hepatitis B virus (HBV)‐derived PreS as an immunological carrier and non‐allergenic peptides from the IgE binding sites of Fel d 1, Fel d 4, and Fel d 7 were expressed in Escherichia coli , purified, and characterized by mass spectrometry, circular dichroism (CD), and size exclusion chromatography. ImmunoCAP and basophil activation experiments demonstrated the hypoallergenic activity of PreS‐Cat 1–5. The ability of PreS‐Cat 1–5 to induce IgE‐blocking antibodies in rabbits was compared to three licensed allergen extract‐based AIT vaccines. PreS‐Cat 1‐5‐specific IgG antibodies were tested for inhibition of allergen‐specific IgE binding and specific basophil activation. T cell activation and induction of specific cytokine secretion by PreS‐Cat proteins were compared with cat allergens in PBMC cultures.

Results

Recombinant hypoallergenic, biochemically and structurally defined PreS‐Cat 1–5 were obtained. Two subcutaneous immunizations of rabbits with PreS‐Cat 1–5 induced equal (Fel d 1) or better (Fel d 4 and Fel d 7) antibodies (PreS‐Cat 5 > PreS‐Cat 1 > PreS‐Cat 3) blocking allergic patients' IgE binding to cat allergens than six to fifteen immunizations with allergen extract‐based vaccines. PreS‐Cat‐specific antibodies strongly inhibited specific basophil activation. PreS‐Cat 5 > PreS‐Cat 1 induced significantly more IL‐10 in cultured PBMCs from cat allergic patients than cat allergens.

Conclusions

PreS‐Cat 5 and PreS‐Cat 1 are highly promising molecular vaccine candidates for AIT of cat allergy, combining Fel d 1‐, Fel d 4‐, and Fel d 7‐peptides in single PreS fusion proteins.

This study reports the construction and preclinical characterization of peptide‐carrier‐based vaccines combining all three important cat allergens (Fel d 1, Fel d 4 and Fel d 7) within single vaccine antigens, termed PreS‐Cat 1‐PreS‐Cat 5. Two subcutaneous immunizations of rabbits with PreS‐Cat 1‐5 induce equal or better antibodies blocking allergic patients' IgE binding to cat allergens than 6 to 15 immunizations with allergen extract‐based vaccines. PreS‐Cat‐specific antibodies strongly inhibit specific basophil activation. PreS‐Cat induces significantly more IL‐10 in cultured PBMCs from cat allergic patients than cat allergens. Abbreviations: IgE, immunoglobulin E; IL‐10, interleukin 10; OD, optical density; PBMCs, peripheral blood mononuclear cells; sIgG, specific IgG.

1. Introduction

Cat is one of the most important and wide‐spread respiratory allergen sources responsible for rhinitis and especially for allergic asthma in Europe, America, and Asia [1]. According to the International Union of Immunological Societies allergen database (https://allergen.org/ accessed July 1, 2024), eight cat allergen molecules have been described. In terms of IgE recognition frequencies in different populations and allergenic activity, Fel d 1, Fel d 4, and Fel d 7 represent the clinically most relevant allergen molecules in cats [2, 3, 4]. Molecular diagnosis based on purified, mainly recombinant allergen molecules in combination with a medical case history and provocation testing has advantages for identifying cat allergic patients for allergen‐specific immunotherapy (AIT) [5, 6, 7]. AIT is a highly effective allergy treatment based on the administration of disease‐causing allergens to sensitized patients to induce allergen‐specific clinical tolerance. Major mechanisms of AIT include the induction of allergen‐specific IgG antibodies that block allergen‐specific IgE recognition and thus the induction of IgE‐mediated activation of inflammatory immune cells such as mast cells, basophils, T cells, and eosinophils [8, 9]. In regard to cat allergy, recombinant monoclonal IgG₄ antibodies blocking IgE binding to the major cat allergen Fel d 1 are effective in reducing IgE‐mediated allergic inflammation in clinical trials, which emphasizes the importance of blocking antibodies for successful treatment [10, 11]. Furthermore, AIT has been reported to induce tolerogenic cytokines such as interleukin 10 (IL‐10) originating from T‐ and B‐regulatory cells capable of inducing tolerance [12, 13, 14, 15, 16]. In this context, it should be mentioned that treatment with tolerogenic T cell epitope‐containing peptides of Fel d 1 was found to reduce symptoms of cat allergy in clinical trials [17, 18, 19].

Currently, AIT for cat allergy is performed with allergen extracts. Difficulties in standardizing cat allergen extracts and treatment‐induced side effects represent major limitations for the treatment of cat allergy by AIT [20, 21]. Accordingly, new forms of AIT are urgently required. Molecular forms of AIT have the advantage that different immunological mechanisms can be precisely targeted [22]. Furthermore, molecular approaches allow us to combine the clinically relevant allergen molecules in the AIT vaccines. In this study, we report the construction, immunological, and preclinical characterization of a recombinant hypoallergenic and tolerogenic vaccine for the treatment of cat allergy. For the vaccine construction, we used a platform that is based on recombinantly expressed fusion proteins consisting of the hepatitis B virus (HBV)‐derived PreS [23, 24] as a carrier protein and hypoallergenic allergen peptides from the IgE binding sites of the allergens [25]. We have previously used this approach to construct a peptide‐carrier vaccine based on peptides of the major cat allergen Fel d 1 [26]. However, it turned out that in addition to Fel d 1, Fel d 4 and Fel d 7 also represent important cat allergens [3, 4, 27]. In this study, the construction and preclinical characterization of peptide‐carrier‐based vaccines combining all three important cat allergens (Fel d 1, Fel d 4, and Fel d 7) within single vaccine antigens, termed PreS‐Cat 1‐PreS‐Cat 5, is reported.

2. Materials and Methods

2.1. Cat‐Allergic Patients

Serum samples from patients with cat allergy (n = 80) were obtained at the Department of Pathophysiology and Allergy Research, Medical University of Vienna, with the approval of the Ethics committee of the Medical University of Vienna (EK 1641/2014) and in the Moscow City Center of Allergy and Immunology, Clinical City Hospital #52, 123,182 Moscow, Russia (Ethics vote 34‐20, 2020) upon having received the informed consent from the patients or their legal guardians. Demographic and clinical data were collected with an adapted ISAAC questionnaire that included detailed questions regarding the occurrence of symptoms upon contact with cats, other allergies, and medication. Specific IgE sensitization to cats was confirmed by the measurement of cat allergen‐specific IgE antibodies (i.e., Fel d 1, Fel d 4, and Fel d 7) by ImmunoCAP (Thermofisher, Uppsala, Sweden). IgE levels ≥ 0.1 kU_A/L were considered positive. Sera from three non‐cat‐allergic subjects were used for control purposes. Table S1 shows the demographic, clinical, and serological characteristics, in particular IgE levels specific for Fel d 1, Fel d 4, and Fel d 7 of patients and control subjects.

2.2. Design of PreS‐Cat 1—PreS‐Cat 5 Fusion Proteins Containing Allergen‐Derived Peptides Fused to PreS

Hypoallergenic Fel d 4 and Fel d 7‐derived peptides, capable of inducing upon immunization of rabbits IgG antibodies strongly inhibiting allergic patients IgE binding to the allergens (Table S2, Figures S1–, S5), were highlighted in the three‐dimensional structures of Fel d 4 (PDB: 8AMC) and Fel d 7 (PDB: 8EPV) using the Pymol 3.0 software (https://pymol.org/) (Figure 1A, Table S2). Fel d 1‐derived peptides capable of inducing IgE‐blocking IgG have been described earlier [26] and were also visualized in the three‐dimensional structure of Fel d 1 (PDB: 1PUO) (Figure 1A). Five recombinant PreS‐fusion proteins were designed, each containing two copies of the identified hypoallergenic allergen peptides (Figure 1B, Table S2) in a different order. In some of the constructs, identical peptides were inserted as twins (e.g., PreS‐Cat 1, PreS‐Cat 2, PreS‐Cat 3, PreS‐Cat 5), whereas in PreS‐Cat 4, twin positions were avoided (Figure 1B). No linker sequences were inserted. The rationale behind the selection of the constructs was that twin constructs may eventually show higher immunogenicity, but this may eventually be at the cost of a less reduced IgE reactivity and allergenic activity.

FIGURE 1.

Hypoallergenic peptides (Table S2) used for the construction of the cat vaccine candidates are highlighted in the ribbon (left) and surface representation (right) of the three dimensional structures of Fel d 1 (PDB: 1PUO), Fel d 4 (PDB: 8AMC), and Fel d 7 (PDB: 8EPV) using the PyMOL software (A). NT and CT mean the N‐terminus and the C‐terminus, respectively. Remaining parts of chain 1 and chain 2 are indicated in dark and light gray, respectively. (B) Schemes of the PreS fusion proteins (PreS‐Cat 1—PreS‐Cat 5) containing allergen‐derived peptides in different order and location.

2.3. Expression, Purification, and Biochemical Characterization of Hexa‐Histidine‐Tagged Recombinant PreS‐Cat 1–PreS‐Cat 5 Proteins

The amino acid sequence of PreS (GenBank: AAT28735.1) was used for the construction of PreS‐Cat 1–PreS‐Cat 5. Synthetic DNA encoding the PreS‐Cat 1–5 proteins was optimized for expression in E. coli (ATG: biosynthetic GmbH, Merzhausen, Germany) and cloned into the pET‐27b vector (Novagen, Maddison, USA). DNA sequences of constructs were verified by double‐strand DNA sequencing. Recombinant PreS‐Cat 1–5 were expressed in E. coli BL‐21 GOLD (DE3) (Agilent Technologies, Santa Clara, USA) as described in the Supplement for rEqu c 1 and purified using an inclusion body preparation protocol (Qiagen, Hilden, Germany). Urea was removed from the elution fractions by step‐wise dialysis decreasing urea concentration from 8 to 1 M Urea in PBS (pH 7.4), and then, dialysis against PBS (pH 7.4) was done. The purity of recombinant PreS‐Cat 1–5 proteins was analyzed by SDS‐PAGE. The concentration of the recombinant fusion proteins was measured using the Micro BCA assay kit (ThermoFisher, Pierce, Waltham, MA, USA). Contents of the secondary structure of PreS‐Cat 1–5 were assessed on a JASCO J‐810 spectropolarimeter (Tokyo, Japan). CD spectra were recorded in the wavelength range from 250 to 190 nm with a resolution of 0.5 nm, with three scans for each sample. The results obtained after subtracting the buffer are presented as the molar ellipticity (Θ).

Potential aggregation of PreS‐Cat 1, PreS‐Cat 3, and PreS‐Cat 5 proteins was analyzed by size exclusion chromatography (SEC) on a 5 μM SEC‐s3000 LC column 300 × 7.8 mm (Phenomenex, Torrance, CA), performed on a Nexera series HPLC system (Shimadzu Corporation, Japan). Initial equilibration was performed with the corresponding PBS buffer (pH 7.4). Then, purified PreS‐Cat proteins in concentrations of 0.45 mg/mL (PreS‐Cat 3) and 0.3 mg/mL (PreS‐Cat 1 and 5) were injected into Verex Vial Kit vials (Phenomenex, Torrance, CA) and run at a flow rate of 1 mL/min. For the control, BSA (Molecular mass 66.5 kDa) and amylase inhibitor (Molecular mass 210 kDa) in PBS were used. Matrix‐assisted laser desorption/ionization time‐of‐flight mass spectrometry (MALDI‐TOF mass spectrometer, Shimadzu, Japan) was used for molecular mass determination as described [28]. Endotoxin measurements were performed using chromogenic endotoxin Quant kit (Pierce, Thermo Scientific, Austria).

2.4. Synthesis and Characterization of Cat Allergen‐Derived Peptides

Information regarding the selection, synthesis, purification, and characterization of cat allergen‐derived peptides can be found in the Supplement of this article.

2.5. Quantification of Allergen‐Specific IgE by ImmunoCAP

IgE levels specific for cat allergens (Fel d 1, Fel d 4, and Fel d 7) as well as to a mix of Fel d 1, Fel d 4, and Fel d 7 and to PreS‐Cat 1–5 proteins were measured by ImmunoCAP technology (Thermofisher). For the measurements of specific IgE levels to Fel d 1, Fel d 4, and Fel d 7, commercial ImmunoCAPs were bought, whereas Streptavidin ImmunoCAPs (o212 ImmunoCAP, Thermo Fisher) were used to prepare ImmunoCAPs containing PreS‐Cat 1–Cat 5 proteins or an equimolar mix of Fel d 1, Fel d 4, and Fel d 7 allergens as described [4]. According to Table S1 the following sera were tested: PreS‐Cat 1: 1–48; PreS‐Cat 2: 1–19; PreS‐Cat 3: 1–4; PreS‐Cat 4: 1–48; PreS‐Cat 5: 1–48. Measurements of specific IgE levels were performed according to the manufacturer's instructions on an ImmunoCAP100 instrument (Thermo Fisher Scientific/Phadia). IgE levels equal to and > 0.1 kU_A/L were considered positive.

2.6. Assessment of the Allergenic Activity of Allergens, Peptides, and Cat 1‐Cat 5

Basophil activation tests were performed using rat basophil leukemia (RBL) cells expressing the human high‐affinity IgE receptor FcϵRI [29]. Measurements were performed in duplicates (deviation < 5%), and the results are expressed as percentages of total β‐hexosaminidase contents of cells after the addition of 1% Triton X‐100 (100% release). The cut‐off levels of β‐hexosaminidase releases correspond to the percentages of releases observed with medium alone (negative controls).

2.7. Direct and Indirect Mapping of IgE Epitopes of Fel d 4 and Fel d 7

As detailed in the Supplement of this article, we performed direct IgE epitope mapping by testing cat allergen‐derived synthetic peptides for IgE binding, whereas indirect IgE epitope mapping was performed by inhibiting allergic patients IgE binding to the allergens by peptide‐specific rabbit antibodies.

2.8. Comparison of Aluminum‐Hydroxide‐Adsorbed PreS‐Cat 1‐Cat 5 With Licensed Allergen Extract‐Based AIT Vaccines Regarding Induction of Allergen‐Specific IgG Responses in Rabbits

Three registered allergen extract‐based AIT vaccines Alutard‐Cat SQ L555 (batch: 00872548700I; expiration date: October 2022; Alutard SQ, ALK‐Abelló, Hørsholm, Denmark), Tyro‐SIT Bencard‐Cat (batch: 2021/050/016895; expiration date: 13 September 2022; Bencard Allergie GmbH, Munich, Germany), and Clustoid Roxall‐Cat (batch: 174821; expiration date: March 2023; Roxall Medizin GmbH, Hamburg, Germany) were purchased for immunization. Alutard SQ is a cat natural hair allergen extract adsorbed onto Aluminum hydroxide. Bencard Tyro‐SIT vaccine is a modified cat hair extract adsorbed on microcrystalline L‐tyrosine. Roxall Clustoid vaccine is a glutaraldehyde‐modified highly polymerized allergen extract adsorbed on Aluminum hydroxide. The products were stored at +4°C, following the manufacturers' recommendations and used for immunization of rabbits according to the manufacturer's instructions for use in allergic patients. Injections were performed as prime and boost schedules like in allergic patients.

Molecular PreS‐Cat 1‐PreS‐Cat 5 vaccines were formulated by adsorbing sterile‐filtrated PreS‐Cat 1–5 proteins (200 μg/mL final concentration, 25% PBS, 0.9% NaCl) onto aluminum hydroxide (1.92 mg/mL final concentration) (Alhydrogel, Brenntag Biosector, Ballerup, Denmark). The degree of adsorption was confirmed by centrifugation of the suspension at 16,000 g for 1 min and measurement of non‐adsorbed proteins in the supernatant by a microBCA assay Kit (ThermoFisher, Pierce, Waltham, MA, USA).

Allergen extract‐ and PreS‐Cat vaccines were tested for the induction of antibodies in rabbits because rabbits are not inbred and hence better reflect variability of immunogenicity due to genetic variations than inbred mice. Furthermore, we found for our PreS‐based vaccine platform that results regarding doses and injection intervals obtained in rabbits transfer well into the human situation [30, 31]. Rabbit immunizations were performed at Charles River Laboratories (Chatillon‐sur‐Chalaronne, France) following European guidelines for animal welfare with ethics permission (https://www.criver.com/) (https://www.criver.com/about‐us/about‐us‐overview/animals‐research?region=3696). Pre‐immune sera were collected shortly before the first injections, and post‐immune sera were obtained 4 weeks after the last injection for each of the vaccines. Rabbits (n = 2 per PreS‐Cat vaccine) were immunized two times subcutaneously at monthly intervals with 0.5 mL containing 100 μg of aluminum hydroxide‐adsorbed PreS‐Cat 1–5 (in total 2 injections). In addition, two rabbits each were immunized with Alutard SQ, Tyro‐SIT, or CLUSTOID according to the manufacturer's instructions for AIT of allergic patients. For Tyro‐SIT, six injections of increasing strength were administered at weekly intervals, followed by one maintenance injection after 1 week and two maintenance injections at 2 weeks intervals (in total nine injections). Alutard was administered in 15 weekly injections with increasing doses within 14 weeks (in total 15 injections). For Clustoid, two injections, the latter with increasing dose, were given in 15‐min intervals on the same day followed by 4 monthly maintenance injections (in total 6 injections).

The development of allergen‐specific IgG antibodies in the rabbits immunized with PreS‐Cat 1–5 and allergen extract‐based vaccines was measured by ELISA as described for the peptide‐immunized rabbits.

2.9. Ability of PreS‐Cat‐ and Allergen Extract Specific Rabbit Antisera to Inhibit Allergic Patients' Allergen‐Specific IgE Binding and Basophil Activation

Methods used for studying the ability of PreS‐Cat‐ and allergen extract‐specific rabbit antibodies to inhibit allergic patients IgE binding to cat allergens and cat allergen‐induced basophil activation can be found in the Supporting Information.

2.10. Induction of CD4 ⁺ and CD8 ⁺ Specific T Cell Activation and Cytokine Release in PBMCs Stimulated With PreS‐Cat 1, PreS‐Cat 5, Allergens, or PreS

PBMCs from 20 cat‐allergic patients (#61–80) (triplicate cultures) were labeled with carboxyfluorescein succinimidyl ester (CFSE) and stimulated with 0.27 nM PreS‐Cat 1, PreS‐Cat 5, PreS, or an equimolar mix of Fel d 1, Fel d 4, Fel d 7, and PreS. For control, an antigen‐free medium was used. The percentages of proliferated CD3⁺CD4⁺ and CD3⁺CD8⁺ cells were assessed after 7 days of culture [24]. The levels of IL‐2, IL‐12, IFN gamma, IL‐5, GM‐CSF, TNF‐alpha, and IL‐10 were determined in PBMC supernatants using the Luminex multiplex assay (Bio‐Plex Pro Human Cytokine 9‐plex Assay; Bio‐Rad) [24]. Endotoxins in the protein samples were measured using Endosafe‐PTS (Charles River) with Limulus amebocyte lysate (LAL) test cartridges.

2.11. Statistical Analysis

The statistical program IBM SPSS Statistics 20.0 (New York, USA) and GraphPad Prism 5 software (GraphPad Software, La Jolla, Calif) were used for statistical analysis and figure preparation. Cytokine levels were log‐transformed as distribution analysis revealed the best fit by a log‐normal distribution. Differences between paired groups were tested by Student's t‐test for paired samples. p‐values were Sidak‐Holm corrected [32]. p‐values < 0.05 were considered significant.

3. Results

3.1. Molecular Typing of Clinically Defined Cat Allergic Patients

Serum samples from 80 clinically well‐defined cat allergic patients were available (Table S1). Ninety‐eight percent of the patients were sensitized to Fel d 1 (IgE levels from 0.1 to 100 kU_A/L, mean IgE level 19.25 kU_A/L); 63% was sensitized to Fel d 4 (IgE levels from 0.1 to 100 kU_A/L, mean IgE level 6.2 kU_A/L) and 65% of the patients was sensitized to Fel d 7 (IgE levels from 0.1 to 100 kU_A/L, mean IgE level 6.3 kU_A/L) (Table S1). The frequency of allergic symptoms was 89% rhinitis, 71% conjunctivitis, 64% asthma, and 31% dermatitis (Table S1). We noted that patients with asthma were typically sensitized to more than one of the three cat allergens as compared to patients with rhinitis/rhinoconjunctivitis only, who were mainly sensitized to Fel d 1 (Table S1).

3.2. Fel d 4‐ and Fel d 7‐Derived Peptides Are Hypoallergenic

In a previous study, we identified two hypoallergenic surface‐exposed peptides from the major cat allergen Fel d 1, which gave rise to a vaccine inducing upon immunization of rabbits IgG antibodies blocking allergic patients IgE binding to Fel d 1 [26] (Table S2). Since Fel d 4 and Fel d 7 have been identified as important additional cat allergens [1, 4], Fel d 4 (Figure S1, left: P1, P2, P3, P4) and Fel d 7 peptides (Figure S1, right: P1, P2, P3, P4, P5) with predicted high surface probability were identified and synthesized. The peptides had molecular weights between 2848 and 4791 Da and contained Cysteine residues for chemical coupling to KLH (Table S2). Isolated and KLH‐coupled Fel d 4 and Fel d 7 peptides showed strongly reduced IgE reactivity as compared to the complete allergens (Figure S2, Table S3).

Basophil activation testing showed that neither Fel d 4‐derived KLH‐coupled peptides nor Fel d 7‐derived KLH‐coupled peptides induced IgE‐dependent basophil degranulation in sensitized patients (Figure S3).

3.3. Fel d 4 and Fel d 7 Contain Defined Major IgE Binding Regions

Indirect mapping of conformational IgE epitopes of Fel d 4 and Fel d 7 was performed by raising IgG antibodies against peptides that are surface exposed on the folded allergen. The peptide‐specific antibodies were then used to block allergic patients IgE reactivity to the intact allergen, as reported [33]. Fel d 4‐specific antibody levels were highest in the anti‐Fel d 4 antiserum, lower in anti‐P1, anti‐P3, and anti‐P4 antisera, and lowest in the anti‐P2 antiserum (Figure S4A). Regarding Fel d 7, we found that IgG levels against Fel d 7 were highest for the anti‐P1 > anti‐P3 > anti‐P5 antisera, followed by the anti‐Fel d 7 antiserum and the anti‐P2 antiserum, whereas the lowest anti‐Fel d 7 IgG levels were found in the anti‐P4 antiserum (Figures S4B). No binding to the allergens was observed with the corresponding pre‐immune sera.

Figure S5A shows that allergic patients' IgE reactivity to Fel d 4 can be blocked to a varying degree with anti‐Fel d 4 or anti‐Fel d 4 peptide antisera. The inhibition of IgE binding obtained with anti‐Fel d 4 antibodies ranged between 12% and 86% (mean inhibition 75.4%) (Figures S5A, Table S4). The best inhibition of IgE binding by anti‐peptide antibodies was obtained with anti‐P4 antibodies (42%–76%, mean inhibition 63%), whereas inhibition of IgE binding by the other anti‐peptide antibodies was low (anti‐P1: 0%–78%, mean 23%; anti‐P2: 0%–29%, mean 6%; anti‐P3: 0%–37%, mean 14%) (Figure S5A, Table S4).

The inhibition of IgE binding to Fel d 7 obtained with anti‐Fel d 7 antibodies ranged between 40% and 99% (mean inhibition 82%) (Figures S5B, Table S5). The best inhibition of IgE binding by anti‐peptide antibodies was obtained with anti‐P5 antibodies (31%–88%, mean inhibition 49%) followed by anti‐P3 antibodies (0%–51%, mean 32%). The inhibition of IgE binding by the other anti‐peptide antibodies was low (anti‐P1: 0%–36%, mean 17%; anti‐P2: 0%–45%, mean 19%; anti‐P4: 0%–26%, mean 7%) (Figure S5B, Table S5). Regarding Fel d 7, we found that the combination of anti‐P3 and anti‐P5 yielded a mean inhibition of IgE binding of 73% which was almost comparable to the result obtained with anti‐Fel d 7 antibodies (i.e., 88%) (data not shown). The results from the indirect mapping of IgE epitopes of Fel d 4 and Fel d 7 by antibody competition experiments thus revealed that Fel d 4‐derived peptide 4 is part of a major IgE epitope‐containing region of Fel d 4 and that peptides 3 and 5 are part of a major IgE epitope‐containing area of Fel d 7 (Figure 1A).

3.4. PreS‐Cat 1–5 Combine Peptides of Fel d 1, Fel d 4, and Fel d 7 With HBV‐Derived PreS Within Single Fusion Proteins

HBV‐derived PreS has been previously identified as a suitable carrier for the construction of molecular allergy vaccines because it renders allergen‐derived peptides immunogenic by providing T cell help when incorporated into the vaccine antigen. Additionally, it induces PreS‐specific antibodies which can protect against HBV infection in vitro [23, 24, 34, 35]. Accordingly, we designed the cat allergy vaccines, PreS‐Cat 1—PreS‐Cat 5 using PreS as an immunogenic carrier in the middle, flanked by two copies of each of the hypoallergenic Fel d 1‐, Fel d 4‐, and Fel d 7‐derived peptides, which had induced blocking IgG antibodies. The PreS‐Cat 1–5 constructs only differed regarding the order of the allergen‐derived peptides (Figure 1B). PreS‐Cat 1–5 could be expressed in E. coli and purified to homogeneity (Figure S6A). Without optimization, more than 10 mg protein per liter culture was obtained. They migrated in SDS‐PAGE as defined bands of approximately 60 kDa (Figure S6A). Recombinant fusion proteins were monomeric in physiological buffers and did not form aggregates, as exemplified for PreS‐Cat 1, PreS‐Cat 3, and PreS‐Cat 5 by size exclusion chromatography (Figure S6B). Only PreS‐Cat 1, PreS‐Cat 3, and PreS‐Cat 5 were analyzed because they best induced blocking antibodies, as indicated in the IgE inhibition experiments performed with PreS‐Cat‐specific antisera. The analysis by circular dichroism spectroscopy showed that PreS‐Cat 1–5 exhibited a minimum at 200 nm, indicating mainly unordered (random coil) secondary structures (Figure S7). The molecular masses determined by mass spectrometry for PreS‐Cat 1–5 ranged between 56.4 and 56.5 kDa, which corresponded to the molecular mass calculated according to their amino acid sequences (i.e., 56,021 Da) (Figure S8). Endotoxin levels per mg were 1209 EU for PreS‐Cat 1, 474 EU for PreS‐Cat 2, 108 EU for PreS‐Cat 3, 295 EU for PreS‐Cat 4, and 360 EU for PreS‐Cat 5 and thus considerably lower than what reported for allergen extracts used for AIT [36].

3.5. PreS‐Cat 1–5 Show Strongly Reduced IgE Reactivity and Lack Allergenic Activity

IgE levels specific for PreS‐Cat 1–PreS‐Cat 5 were quantified by ImmunoCAP measurements and compared to those directed against a mix of Fel d 1, Fel d 4, and Fel d 7. Figure 2A shows the percentage reductions of PreS‐Cat 1‐5‐specific IgE levels, which for the majority of patients was greater than 95%. However, the most relevant parameter for immediate allergenic activity is the ability of a given antigen to induce basophil activation. We therefore compared the allergenic activity of PreS‐Cat 1–5 with an equimolar mix of Fel d 1, Fel d 4, and Fel d 7 (i.e., amounts of allergens were equimolar to the peptides included in the PreS‐Cat antigens) in a first set of basophil activation experiments (Figure 2B). In each of the tested patients, we found an induction of basophil activation by the allergen mix, while none of the PreS‐Cat molecules induced relevant mediator release (Figure 2B).

FIGURE 2.

Lower IgE binding capacity of PreS‐Cat 1–5 as compared to Fel d 1 + Fel d 4 + Fel d 7 (Α). Shown are the percentages of reduction of IgE (y‐axes) when quantitative IgE levels measured by ImmunoCAP specific for PreS‐Cat 1—PreS‐Cat 5 (x‐axes) were compared with IgE levels specific for a mix of Fel d 1, Fel d 4, and Fel d 7 in the numbers of patients indicated below. (B) Comparison of the allergenic activity of different concentrations of PreS‐Cat 1–5 (x‐axes) with an equimolar mix of Fel d 1 + Fel d 4 + Fel d 7 regarding basophil activation in cat‐allergic patients (patients 3, 6, 10, 12, 14, 15, 16, 17, 18, or 19). β‐hexosaminidase releases are expressed as percentages of total mediator contents of cells (y‐axes). The horizontal dashed lines correspond to the percentages of releases observed with medium alone (negative controls).

3.6. Two Immunizations With PreS‐Cat 1–5 Induce IgG Antibodies Blocking Allergic Patients' IgE Binding to Fel d 1, Fel d 4, and Fel d 7

Table S6 shows that it is possible to formulate stable vaccines containing 200 μg/mL of PreS‐Cat 1–PreS‐Cat 5 maintaining the vast majority of protein adsorbed (i.e., PreS‐Cat 1: 95.78%; PreS‐Cat 2: 87.93%; PreS‐Cat 3: 96.62%; PreS‐Cat 4: 90.86%; PreS‐Cat 5: 96.65%). Next, we immunized rabbits with two subcutaneous injections containing 100 μg of each of the cat fusion proteins (PreS‐Cat 1—PreS‐Cat 5) at monthly intervals. Likewise, rabbits were immunized with registered allergen extract‐based vaccines from three companies (Alutard, Tyro‐SIT, Clustoid) as recommended by the manufacturer for AIT of patients (Figure 3). Rabbits received either two injections of cat fusion proteins PreS‐Cat 1–5, 15 injections of Alutard, 9 injections of Tyro‐SIT, or 6 injections of Clustoid. Allergen‐specific IgG antibody levels were compared in serum samples obtained from rabbits 4 weeks after the last injection performed with each of the vaccines. Figure 4A shows allergen‐specific IgG levels induced by PreS‐Cat 1–5 (left part) and by allergen extract‐based vaccines (right part). PreS‐Cat 1 and PreS‐Cat 5 induced the highest levels of Fel d 1‐specific IgG which were comparable to those induced by Alutard and Clustoid (Figure 4A). PreS‐Cat 1 induced the highest levels of Fel d 4‐specific IgG followed by PreS‐Cat 2, PreS‐Cat 3, and PreS‐Cat 5, which induced comparable levels of Fel d 4‐specific IgG. PreS‐Cat 4 induced the lowest Fel d 4‐specific IgG levels (Figure 4A, left). Alutard induced low levels of Fel d 4‐specific IgG, whereas Tyro‐SIT and Clustoid failed to induce Fel d 4‐specific IgG (Figure 4A, right). The highest levels of Fel d 7‐specific IgG were induced by PreS‐Cat 1, PreS‐Cat 3, and PreS‐Cat 5, whereas PreS‐Cat 2 and PreS‐Cat 4 induced lower Fel d 7‐specific IgG levels (Figure 4A, left). Only very low levels of Fel d 7‐specific IgG were induced by allergen extract‐based vaccines (Figure 4A, right). Taking all three allergens (Fel d 1, Fel d 4 and Fel d 7) into consideration, PreS‐Cat 1, PreS‐Cat 3, and PreS‐Cat 5 seemed to have advantages over PreS‐Cat 2 and PreS‐Cat 4 regarding the induction of allergen‐specific IgG. This advantage was further confirmed by IgE inhibition experiments, which showed that IgG antibodies induced by PreS‐Cat 1, PreS‐Cat 3, and PreS‐Cat 5 inhibited allergic patients' IgE binding to Fel d 1 better than those induced by PreS‐Cat 2 and PreS‐Cat 4 (Figure 4B).

FIGURE 3.

Schedules of immunization of rabbits with PreS‐Cat 1–5 or allergen extract‐based vaccines (Alutard, Tyro‐SIT, and Clustoid). Time points of injections and blood sampling are indicated (x‐axis), and the total number of injections are displayed for each vaccine.

FIGURE 4.

Allergen‐specific IgG levels induced by immunization with PreS‐Cat 1–5 (left panels) or allergen extract‐based vaccines (Alutard, Tyro‐SIT, and Clustoid) (right panels). (A) Shown are OD values corresponding to IgG levels specific for Fel d 1, Fel d 4, and Fel d 7 determined by ELISA for different dilutions of immune sera obtained 4 weeks after the last injection or of a 1:500 dilution of the pre‐immune serum (y‐axes). (B) Comparison of allergic patients' (n = 11) IgE binding to Fel d 1 (OD log₁₀) (y‐axis) after pre‐incubation of Fel d 1 with rabbit antibodies obtained after immunization with PreS‐Cat 1–5 or corresponding pre‐immune sera (x‐axis). p‐values < 0.05 were considered significant. *p‐value < 0.05, **p‐value ≤ 0.001, ns—not significant.

Accordingly, we performed extensive IgE inhibition studies with PreS‐Cat 1‐, PreS‐Cat 3‐, and PreS‐Cat 5‐specific antisera (Table S7, Figure 5). The inhibition of allergic patients' IgE binding to Fel d 1 by PreS‐Cat 1‐specific antibodies ranged from 28%–75% (mean 49.5%), by PreS‐Cat 3‐specific antibodies, it was 38%–74% (mean 53.1%), and by PreS‐Cat 5‐specific antibodies, it was 51%–85% (mean 65%) (Table S7, Figure 5). Inhibition of allergic patients IgE binding to Fel d 4 was 30%–67% (mean 53.9%) by PreS‐Cat 1‐specific antibodies, 29%–60% (mean 44.3%) by PreS‐Cat 3‐specific antibodies, and 42%–57% (mean 51.9%) by PreS‐Cat 5‐specific antibodies (Table S7, Figure 5). Inhibition of allergic patients IgE binding to Fel d 7 was 43%–72% (mean 59.3%) for PreS‐Cat 1‐specific antibodies, 39%–69% (mean 55.1%) for PreS‐Cat 3‐specific antibodies, and 56%–80% (mean 62.1%) for PreS‐Cat 5‐specific antibodies (Table S7, Figure 5).

FIGURE 5.

Comparison of allergic patients' IgE binding to (A) Fel d 1 (patients n = 22), (B) Fel d 4 (patients n = 6), or (C) Fel d 7 (patients [n = 7]) (OD log₁₀) (y‐axis) after pre‐incubation of allergens with rabbit antibodies obtained after immunization with PreS‐Cat 1–5, Alutard, Tyro‐SIT, Clustoid, or corresponding pre‐immune sera (x‐axis). Mean percentages of inhibition of IgE binding are indicated. p‐values < 0.05 were considered significant. *p‐value < 0.05, **p‐value ≤ 0.001, ***p‐value ≤ 0.0001, ns—not significant.

Inhibition of allergic patients' IgE binding to Fel d 1 with antibodies induced by allergen extract‐based vaccines was as follows: Alutard, 68%–86% (mean 76%), Tyro‐SIT, 0%–11% (mean 1.9%), Clustoid, 67%–88% (mean 76.3%). For Fel d 4, it was: Alutard, 0%–62% (mean 36.5%), Tyro‐SIT, 0%–6% (mean 1.5%), Clustoid, 0%–26% (mean 8.2%). For Fel d 7: Alutard, 0%–50% (mean 11.4%), Tyro‐SIT, 0%–17% (mean 3.3%), and Clustoid, 0%–57% (mean 12.4%).

In summary, PreS‐Cat 5 was better than PreS‐Cat 1 and PreS‐Cat 3 regarding the induction of IgG antibodies blocking allergic patients' IgE binding to Fel d 1 and comparable to Alutard and Clustoid. Tyro‐SIT failed to induce antibodies blocking allergic patients' IgE binding to Fel d 1. Only PreS‐Cat proteins, but not allergen extract‐based vaccines, induced antibodies blocking allergic patients' IgE binding to Fel d 4 and Fel d 7 in a relevant manner.

3.7. PreS‐Cat Vaccine‐Specific IgG Antibodies Partially Inhibit Allergic Patients' IgE Binding to Can f 1 and Equ c 1

Fel d 7 and Fel d 4 have shown sequence and structural similarity to the major dog allergens Can f 1 and Can f 6, respectively [37, 38]. Furthermore, there is a similarity between Fel d 4 and the major horse allergen Equ c 1 [39]. Indeed, we found that antibodies induced by PreS‐Cat 1, PreS‐Cat 3, and PreS‐Cat 5 inhibited patients' IgE binding to Can f 1 and Equ c 1 to some extent (approximately 30% mean inhibition), whereas inhibition was low (< 10%) with antibodies obtained by immunization with allergen extract‐based vaccines (Figure S9).

3.8. PreS‐Cat Vaccine‐Specific IgG Antibodies Inhibit Basophil Activation Induced by Fel d 1, Fel d 4, and Fel d 7

Table S8 shows the induction of basophil activation for different doses of an equimolar mix of Fel d 1, Fel d 4, and Fel d 7, which had been pre‐incubated with normal rabbit antibodies or rabbit antibodies from PreS‐Cat 1‐ or PreS‐Cat 5‐immunized rabbits. For most of the 20 tested patients, the percentages of beta‐hexosaminidase releases were shifted to a 100‐fold or higher allergen concentration, when allergens had been pre‐incubated with PreS‐Cat 1‐ or PreS‐Cat 5‐specific rabbit antibodies (Table S8). For four patients (i.e., patients 7, 8, 48, and 51), basophil release was completely suppressed by PreS‐Cat 1‐ or PreS‐Cat 5‐specific antibodies (Table S8). For patients 52, 56, 53, and 16, at least a 10‐fold shift of the dose response curve was noted. For patient 6, the tested concentrations appeared to represent the plateau of the bell‐shaped release curve so that a shift could not be calculated. However, the release in the plateau was approximately 30% lower in the presence of PreS‐Cat 1 or PreS‐Cat 5‐specific antibodies as compared to normal rabbit antibodies (Table S8).

3.9. PreS‐Cat 1 and PreS‐Cat 5 Induce Significantly Higher Levels of the Tolerogenic Cytokine IL‐10 in Cultured PBMCs From Cat Allergic Patients Than the Allergen Mix

In the next set of experiments, we studied the effects of the mix of Fel d 1, Fel d 4, Fel d 7, and PreS, PreS‐Cat 1, PreS‐Cat 5, or PreS alone on in vitro cultured PBMCs from 16 cat‐allergic subjects and 3 subjects without cat allergy. Figure S10A shows that PreS‐Cat 1 and PreS‐Cat 5 induced a slightly higher proliferation of CD4⁺ and CD8⁺ T cells in PBMC cultures from cat allergic patients than the allergen mix containing PreS, while PreS alone did not induce proliferation. A similar but less pronounced effect was observed for PBMCs from non‐allergic subjects (Figure S10B). Also here, PreS‐Cat 1 and PreS‐Cat 5 induced a higher proliferation of CD4⁺ and CD8⁺ T cells than the allergen mix including PreS. The measurement of cytokines in the supernatants of cultured and stimulated PBMCs from cat allergic patients showed a complex pattern of cytokine secretion with some consistent findings for the PreS‐Cat molecules (Figure 6). The PreS‐Cat‐specific effects were that both PreS‐Cat 1 and PreS‐Cat 5 induced significantly higher levels of IL‐2 than the cat allergen mix containing PreS (Figure 6, Table S9). Likewise, the production of IL‐10 and IL‐13 was significantly higher when allergic patients PBMCs were exposed to PreS‐Cat 1 or PreS‐Cat 5 as compared to the cat allergen mix containing PreS (Figure 6, Table S9). In contrast, the effects of PreS‐Cat molecules on Th2 cytokines varied: PreS‐Cat 5 induced significantly higher IL‐4 levels than the allergen mix; PreS‐Cat 1 induced significantly lower levels of IL‐5 and GM‐CSF than the allergen mix, whereas PreS‐Cat 5 induced significantly higher levels of TNF‐alpha than the allergen mix (Figure 6, Table S9). PreS alone did not induce any relevant secretion of cytokines (Figure 6).

FIGURE 6.

Induction of cytokine production in PBMCs of cat allergic patients (n = 20) (y‐axes: Cytokine concentrations; box‐and‐whisker plots showing minimum, quartiles, median, and maximum values) by allergen mix (Fel d 1, Fel d 4, Fel d 7, and PreS), PreS‐Cat 1, PreS‐Cat 5, PreS, or medium alone (x‐axes). Results are presented as bar diagrams with individual data points. p‐values < 0.05 were considered significant. *p‐value < 0.05, **p‐value ≤ 0.001, ***p‐value ≤ 0.0001.

4. Discussion

Cat allergy affects more than 150 million people in the world who suffer from severe forms of respiratory allergy, especially asthma. Fel d 1 is the major cat allergen [40], but patients with cat‐related asthma have been found to be sensitized also to additional cat allergens, especially to Fel d 4 and Fel d 7 [3, 4]. We report the design and characterization of molecular vaccines for AIT treatment of cat allergy, PreS‐Cat 1–5, which combine hypoallergenic peptides from the IgE binding sites of the three major cat allergens, Fel d 1, Fel d 4, and Fel d 7 [3, 4] fused to an immunological carrier protein, HBV‐PreS. PreS‐Cat antigens showed a strong reduction of IgE reactivity as well as of allergenic activity and induced in PBMC cultures the secretion of the tolerogenic cytokine IL‐10. Only two injections of PreS‐Cat vaccines induced upon immunization of rabbits better blocking antibodies to the three important cat allergens, Fel d 1, Fel d 4, and Fel d 7 than three licensed allergen extract‐based vaccines.

The carrier protein PreS in the PreS‐Cat antigens serves the purpose of increasing the immunogenicity of the fused peptides to increase allergen‐specific protective IgG responses and to reduce side effects induced by allergen‐specific T cell epitope‐containing peptides. The aforementioned principle has already been used for the development of the grass pollen allergy vaccine BM32, which has been studied up to phase II immunotherapy trials for the treatment of grass pollen‐induced rhinitis and achieved a reduction of approximately 25% of the combined symptom medication score over placebo [41]. The carrier‐peptide vaccine platform is designed to induce blocking IgG antibodies against IgE binding sites of allergens and therefore differs from the strategy of using tolerogenic allergen‐derived peptides to induce T cell tolerance and hence suppress downstream allergic inflammation. The strategy of using T cell epitope‐containing peptides has been advanced by the group of Mark Larché, who found that antigen‐specific non‐responsiveness could be induced by the injection of tolerogenic T cell epitope‐containing peptides of Fel d 1 [42]. The T cell peptide vaccine for cat allergy was shown to generate IL‐10‐dependent immunological tolerance and provided linked epitope suppression [43]. Immunotherapy with Fel d 1‐derived tolerogenic peptides in clinical trials showed a reduction of late phase skin reactions [17, 44] and nasal symptoms in cat allergic patients [19, 45].

Due to the importance of cat allergy, several other strategies for AIT based on Fel d 1 have been developed. For example, recombinant Fel d 1 bound to carbohydrate‐based particles has been shown to induce Fel d 1‐specific IgG responses in mice [46]. A hypoallergenic recombinant derivative of Fel d 1 reduced cat allergic symptoms in a mouse model [47], and Fel d 1 displayed on virus‐like particles was found to exhibit reduced allergenic activity but enhanced immunogenicity [48].

The PreS‐Cat 1–5 vaccine antigens described in our study represent a major further development of the PreS‐based fusion protein approach described for the grass pollen allergy vaccine, BM32. In BM32, four vaccine antigens covering the epitopes of the four major grass pollen allergens Phl p 1, Phl p 2, Phl p 5, and Phl p 6 were included. Each of these vaccine antigens needed to be produced separately and formulated as a mix of aluminum hydroxide‐adsorbed proteins. By contrast, we succeeded in including the critical peptides from the IgE binding sites of Fel d 1, Fel d 4, and Fel d 7 in single fusion proteins to reach the goal of obtaining AIT vaccines containing only one vaccine antigen comprising all three major cat allergens. PreS‐Cat 1–5 includes peptides from chain 1 and chain 2 of the major cat allergen Fel d 1, which have shown to potently induce, upon immunization, IgG antibodies blocking allergic patients' IgE binding to Fel d 1 [26]. Interestingly, Fel d 1‐derived peptides contain also the binding sites of a monoclonal antibody, REGN1908, which has been used for the treatment of cat allergy by passive immunization [10]. We were able to map the IgE binding sites for Fel d 4 and Fel d 7 with antisera raised against allergen‐derived peptides for identifying those Fel d 4 and Fel d 7 peptides, which give rise to antibodies strongly blocking allergic patients' IgE binding to the allergens. For Fel d 4, peptide 4 from the C‐terminus of the allergen was identified, whereas for Fel d 7, a combination of peptides 3 and 5, the latter also derived from the C‐terminus of the allergen, was found to induce the best IgG antibodies blocking allergic patients' IgE.

PreS‐Cat 1–PreS‐Cat 5 were constructed by assembling the aforementioned peptides (P1 and P5, Fel d 1), (P4, Fel d 4), and (P3 and P5, Fel d 7) in different order considering that the assembly may affect the allergenic activity of the fusion proteins and their ability to induce blocking antibodies upon immunization. We found that the allergenic activity, as tested by IgE binding and, most importantly, by the induction of allergen‐specific basophil activation, was strongly reduced. Each of the five PreS‐Cat antigens could be expressed in a large quantity in E. coli , and they were purified as defined proteins with correct molecular mass. CD analysis showed that PreS‐Cat 1–5 were unfolded, which is considered to be the reason for their strongly reduced allergenic activity because IgE epitopes of Fel d 1, Fel d 4, and Fel d 7 are mainly of the conformational type. In fact, no relevant IgE binding to the allergen‐derived synthetic peptides representing linear epitopes was found by us. When we compared PreS‐Cat 1–5 with allergen extract‐based vaccines, we found that allergen extract‐based vaccines induced a highly variable induction of IgE‐blocking antibodies, and there was a very poor induction of blocking antibodies for Fel d 4 and Fel d 7, which would indicate that the allergen‐extract‐based vaccines may not or only poorly protect patients who are also sensitized to Fel d 4 and Fel d 7. In this context, it should be mentioned that PreS‐Cat vaccines induced blocking antibodies not only to Fel d 4 and Fel d 7 but also against cross‐reactive major dog and horse allergens, Can f 1 and Equ c 1, respectively, so that one may perhaps expect a beneficial effect for patients who are cross‐sensitized to dog and/or horse [37, 38, 39]. The induction of IgE‐blocking antibodies to Fel d 1 was comparable between allergen extract‐based vaccines and PreS‐Cat 1–PreS‐Cat 5, but PreS‐Cat 5 seemed to induce blocking Fel d 1‐specific antibodies more consistently than the other PreS‐Cat proteins.

An important finding of the immunization studies was that only two immunizations with PreS‐Cat antigens were needed for the induction of blocking antibodies, whereas 6–15 injections of allergen extract‐based vaccines were required. Similar results were also observed for a PreS‐based recombinant vaccine for the treatment of birch pollen and related oral allergy syndrome caused by apple [24]. One may therefore expect that AIT schedules with PreS‐Cat antigens will be simpler and more convenient than those for allergen extract‐based vaccines, which in turn may increase the treatment compliance of patients. A comparison of the doses of the PreS‐Cat antigens with allergen extract‐based vaccines is impossible because manufacturers of allergen extracts cannot provide exact amounts of allergens present in their AIT vaccines. Only for the grass pollen extract‐based vaccine Alutard was 20 μg of one of the major allergens (i.e., Phl p 5) reported for the highest dose [49]. The dose of 100 μg of the PreS‐Cat‐based vaccines seems therefore quite realistic to be used in clinical trials, given the strong reduction of allergenic activity of the PreS‐Cat antigens.

The molecular vaccines PreS‐Cat 1–5 are mainly constructed to induce blocking polyclonal IgG antibody responses against the IgE epitopes of allergens and thus to block allergen‐induced activation of mast cells, basophils, T cells, and IgE production by memory B cells via inhibition of IgE‐mediated mechanisms. However, we found another interesting feature of the PreS‐Cat antigens when we exposed allergic patients' PBMCs to a mix of cat allergens and the PreS‐Cat antigens. In fact, PreS‐Cat antigens induced significantly more of the tolerogenic cytokine IL‐10 than cat allergens [12]. IL‐10 has been identified as a crucial cytokine produced by regulatory T cells as well as by regulatory B cells and we thus speculate that PreS‐Cat proteins will also exhibit a tolerogenic effect on T cells and B cells when used for AIT [8, 12, 13, 16, 50, 51]. It is quite possible that the induction of IL‐10 is due to the fact that PreS‐Cat proteins expanded Treg cells by induction of IL‐2 [52]. At the same time, one may expect that PreS‐Cat vaccines can induce high levels of allergen‐specific IgG₄ which may be driven by IL‐10 [53].

We are aware that clinical studies will be needed to confirm the results of our preclinical data in order to investigate the safety of the recombinant vaccine and to demonstrate its clinical efficacy. However, it has been shown that grass pollen allergic patients tolerated up to 160 micrograms of BM32 proteins in the previously conducted clinical trials [31, 41]. Our preclinical study has limitations: For example, we have immunized only a limited number of rabbits with the molecular cat vaccines and the allergen extract‐based vaccines, and results were obtained with a mix of the antisera of the two animals only. Nevertheless, results were very consistent, but additional dose‐finding studies in rabbits and in first clinical studies may be needed.

There are currently only very limited data from clinical studies available for the allergen extract‐based vaccines studied by us. We found only for Alutard a few placebo‐controlled studies performed in limited numbers of patients indicating a significant clinical improvement in actively versus placebo treated patients [54, 55, 56]. Since PreS‐Cat vaccines induced blocking antibodies to Fel d 1 in a comparable manner to Alutard and better blocking antibodies to Fel d 4 and Fel d 7, we hope that PreS‐Cat vaccines may prove effective in clinical trials as well. Another possibility is to compare preclinical results obtained by us with those obtained by Orengo et al., who have investigated the blocking effects of Fel d 1‐specific antibodies and clinical effects obtained by passive immunization with these antibodies [10]. They found that each of the antibodies (REGN 1908, REGN 1909) achieved a maximal 50% inhibition of patients' IgE binding to Fel d 1 and when combined a maximal inhibition of 83%, however, with a large variation of blocking effects. These data correspond to the IgE inhibition data which we obtained with PreS‐Cat 5‐induced IgG antibodies, but the variation of inhibition observed for PreS‐Cat 5 was low. Furthermore, the passive immunization studies conducted by Orengo et al. did not include Fel d 4 or Fel d 7‐specific antibodies. Nevertheless, Orengo et al. achieved a highly significant reduction of total nasal symptom scores (TNSS) in their clinical trials for cat allergen induced allergic rhinitis [10, 11], and their antibodies significantly prevented reductions of FEV1 in cat allergic asthma patients in cat exposure chambers [57]. Based on the comparison of the preclinical data obtained by us for PreS‐Cat 5 and the mix of REGN 1908 and REGN 1909, we consider PreS‐Cat 5 as a highly promising vaccine candidate for the treatment of cat allergy warranting further clinical studies.

Author Contributions

D.T.: design and perform experiments, investigation, validation, data analysis, visualization, writing – original draft preparation, and review and editing; M.C.: design and perform experiments, investigation, validation, data analysis, and review and editing; M.F.T., Z.L., K.B., P.G., M.v.H., H.G., R.K., H.J.H., W.K., K.R., A.K., M.K., I.T., and D.F.: investigation, visualization, validation, formal analysis, and review and editing; A.K. (Alexander Karaulov) project administration, data curation, funding acquisition, and review and editing; R.V.: conceptualization, design and supervision of experiments, project administration, data curation, funding acquisition, and writing – original draft preparation.

Conflicts of Interest

Rudolf Valenta has received research grants from WORG Pharmaceutical (Hangzhou, China) and HVD Life Biotech (Vienna, Austria). He serves as a consultant for HVD and WORG Pharmaceuticals. D.T., M.C., K.R., A.K. (Antonina Karsonova), M.v.H., A.K. (Alexander Karaulov) and R.V. are authors of the patent application “Vaccine for treating allergies” application number: PCT/CN2022/130414, reference: P20222370. The authors with a Russian affiliation declare that they have prepared the article in their “personal capacity” and/or that they are employed at an academic/research institution where research or education is the primary function of the entity. M.v.H. has received lecture fees from Astra Zeneca and Thermofisher scientific outside the submitted work. The other authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Supporting information

Figure S1. Surface accessibility plot of (A) Fel d 4 and (B) Fel d 7. Shown are the percentages of surface accessibility according to Expasy ProtScale (y‐axes) for the amino acid sequences of Fel d 4 (left) and Fel d 7 (right) (x‐axes). Synthetic allergen‐derived peptides with high surface accessibility were identified and are indicated at the bottom (Fel d 4: P1, P2, P3, and P4; Fel d 7: P1, P2, P3, P4, and P5). Cysteine residues labeled in red were added to facilitate coupling.

Figure S2. IgE reactivity (y‐axes) of Fel d 4, KLH, Fel d 4‐derived coupled and uncoupled peptides (A, x‐axis) and of Fel d 7, KLH, Fel d 7‐derived coupled and uncoupled peptides (B, x‐axis). Average OD values (405/492 nm, y‐axes) measured by ELISA corresponding to bound IgE are shown for 16 patients for Fel d 4 and for 12 patients for Fel d 7. The cut‐off levels are shown as dotted horizontal lines.

Figure S3. Comparison of the allergenic activity of Fel d 4 and Fel d 7 with allergen‐derived coupled peptides. RBL cells expressing human FcεRI were loaded with serum IgE from four Fel d 4‐sensitized patients (A) or from four Fel d 7‐sensitized patients (B) and then exposed to different concentrations (x‐axis: 100, 10, 1, and 0.1 ng/mL) of the allergens or KLH‐coupled peptides. β‐hexosaminidase releases are expressed as percentages of total releases at the y‐axes. Horizontal broken lines represent the cut‐offs as determined with HSA.

Figure S4. IgG reactivity of rabbits immunized with Fel d 4 or Fel d 4‐derived coupled peptides (A) or with Fel d 7 or Fel d 7‐derived coupled peptides (B) with the corresponding allergens. Shown are average OD levels (< 5% standard deviation) corresponding to bound IgG (y‐axes) for the pre‐immune serum (dilution 1:500) and for different dilutions of the antisera obtained 4 weeks after the last immunization (x‐axes).

Figure S5. Inhibition of patients’ IgE binding to Fel d 4 (A) or Fel d 7 (B) with allergen or peptide‐specific rabbit antisera (x‐axes). Shown are the mean percentages ± SD of IgE inhibitions as histograms (y‐axes) for 12 Fel d 4‐ and for 12 Fel d 7‐positive cat allergic patients and Fel d 7‐derived peptides (B) (n = 12). Dots show the individual values for each patient.

Figure S6. Coomassie brilliant blue‐stained SDS‐polyacrylamide gel electrophoresis containing purified PreS‐Cat 1‐5 proteins (A). Molecular weight markers are indicated in kilo Dalton (kDa). (B) Size exclusion chromatography showing the molecular masses of the recombinant PreS‐Cat 1, PreS‐Cat 3, and PreS‐Cat 5 proteins in relation to gel filtration standards Phl p 6 (11.79 kDa), BSA (66.5 kDa), and amylase inhibitor (210 kDa) and thyreoglobulin (669 kDa). The y‐axis shows the peak intensities and the x‐axis the elution times. Position of markers is indicated by the arrows.

Figure S7. CD spectra of recombinant PreS‐Cat 1‐5 proteins. Molar ellipticities (Θ) of PreS‐Cat 1‐5 (y‐axes) were recorded at different wave lengths (x‐axes).

Figure S8. Analysis of PreS‐Cat 1‐5 by mass spectrometry. Molecular mass was determined by matrix‐assisted laser desorption/ionization time‐of‐flight mass spectrometry. Shown are signal intensities (y‐axes) for the range of masses (x‐axes).

Figure S9. Percentages reduction of allergic patients’ IgE binding to Can f 1 (patients n = 4) (left) or to Equ c 1 (patients n = 3) (right) (OD log10) (y‐axis) after pre‐incubation of allergens with rabbit antibodies obtained after immunization with PreS‐Cat 1‐5, Alutard, Tyro‐SIT, and Clustoid (x‐axis). Average percentages of inhibition of IgE binding are indicated.

Figure S10. Induction of CD4⁺ and CD8⁺ T cell proliferation in cat allergic patients (A) or in subjects without cat allergy (B). CD4⁺ and CD8⁺ T cell proliferation (y‐axes: percentages of proliferated cells; box‐and‐whisker plots showing minimum, quartiles, median, and maximum values) by allergen mix (Fel d 1, Fel d 4, Fel d 7, and PreS), PreS‐Cat 1, PreS‐Cat 5, or PreS (x‐axes). Results are presented as bar diagrams with individual data points for 16 cat allergic patients (A) and 3 subjects without cat allergy (B). p‐values < 0.05 were considered significant. *p‐value < 0.05.

Table S1. Demographic, clinical, and serological characteristics of allergic patients and control subjects. Shown are sex, age, cat‐related symptoms (A: Asthma; R: Rhinitis; C: Conjunctivitis; D: Dermatitis) and allergen‐specific IgE levels (ImmunoCAP: kU_A/L).

Table S2. Amino acid sequences, molecular masses, and numbers of amino acids for peptides derived from Fel d 1, Fel d 4, and Fel d 7 are shown. Cysteine residues added to facilitate coupling of the peptides are indicated in red. Amino acids involved in binding of the monoclonal antibody REGN 1908 are indicated by yellow background. The amino acid sequence of the HBV‐derived PreS sequence used for the construction of fusion proteins is shown below the Table.

Table S3. Original average OD values (standard deviation < 5%) corresponding to IgE levels specific for Fel d 4, KLH, Fel d 4‐derived coupled, and uncoupled peptides (upper part) and specific for Fel d 7, KLH, Fel d 7‐derived coupled, and uncoupled peptides (lower part) for Figure S2 are displayed.

Table S4. Average OD values (< 5% standard deviation) corresponding to IgE levels specific for Fel d 4 after pre‐incubation of the allergen with pre‐immune or sera from rabbits after immunization with Fel d 4 or Fel d 4‐derived peptides and corresponding percentages of IgE inhibition for the 12 patients in Figure S5A. Mean inhibitions for the 12 patients are shown in the bottom line.

Table S5. Average OD values (< 5% standard deviation) corresponding to IgE levels specific for Fel d 7 after pre‐incubation of the allergen with pre‐immune or sera from rabbits after immunization with Fel d 7 or Fel d 7‐derived peptides and corresponding percentages of IgE inhibition for the 12 patients in Figure S5B. Mean inhibitions for the 12 patients are shown in the bottom line.

Table S6. Adsorption of recombinant PreS‐Cat 1–5 constructs to aluminum hydroxide.

Table S7. Comparison of allergic patients’ IgE binding to Fel d 1, Fel d 4, or Fel d 7 after pre‐incubation of allergens with rabbit antibodies obtained after immunization with PreS‐Cat 1–5, Alutard, Tyro‐SIT, Clustoid, or corresponding pre‐immune sera. Shown are OD values corresponding to bound IgE antibodies and percentages of IgE inhibition. Significance of differences in IgE binding was calculated by Students t‐test.

Table S8. Allergen‐specific inhibition of activation of basophils by rabbit anti‐PreS‐Cat 1 or anti‐PreS‐Cat 5 antibodies. Shown are ß‐hexosaminidase releases (averages with < 10% deviation) expressed as percentages of total release for different allergen concentrations and different patients after pre‐incubation of allergens with normal rabbit serum or rabbit anti‐PreS‐Cat 1 or anti‐PreS‐Cat 5 antibodies. Allergen‐specific IgE levels are shown as kU_A/L or ISU for each patient.

Table S9. Detailed statistical analysis of the differences of cytokine levels in PBMCs of cat allergic patients induced by allergen mix (Fel d 1, Fel d 4, Fel d 7, and PreS), PreS‐Cat 1, PreS‐Cat 5, PreS, or medium alone. Differences between groups were compared by Sidak‐Holm‐corrected t‐test for multiple comparisons.

Appendix S1.

Data Availability Statement

The data that supports the findings of this study are available in the Supporting Information of this article.