The Fab fragment of anti-IgE Cε2 domain prevents allergic reactions through interacting with IgE-FcεRIα complex on rat mast cells

Immunoglobulin E (IgE) plays a central role in the pathogenesis of Type I hypersensitivity through interaction with a high-affinity receptor (FcεRIα). For therapeutic applications, substantial attention has been focused recently on the blockade of the IgE interaction with FcεRIα. While exploring better options for preventing allergic diseases, we found that the Fab fragment of the rat anti-murine IgE antibody (Fab-6HD5) strongly inhibited passive cutaneous anaphylaxis (PCA) in vivo, as well as spleen tyrosine kinase (Syk) activity and β-hexosaminidase release from basophilic leukemia cells in vitro. The in vivo effects of Fab-6HD5 pre-administration were maintained over a long period of time for at least 10 days. Using flow cytometry analysis, we also found that Fab-6HD5 did not recognize the IgE Cε3 domain containing specific binding sites for FcεRIα. Furthermore, deletion-mapping studies revealed that Fab-6HD5 recognized conformational epitopes on the Cε2 domain of IgE. Given that the Cε2 domain plays a key role in stabilizing the interaction of IgE with FcRIα, our results suggest that the specific binding of Fab-6HD5 to the Cε2 domain prevents allergic reactions through destabilizing the preformed IgE-FcεRIα complex on rat mast cells. Although the present study was performed using animal models, these findings support the idea that a certain antibody directed against IgE CH domains may contribute to preventing allergic diseases through interacting with IgE-FcεRIα complex.

SCIentIfIC REpORTS | (2018) 8:14237 | DOI: 10.1038/s41598-018-32200-z which suggests that the Fab fragment of IgE reduces the risk of anaphylaxis. To analyze the anaphylactic reactions elicited by IgE antibodies, we previously produced rat monoclonal antibodies that react with murine IgE. In the present study, we focused on one of these antibodies (6HD5) and analyzed its mechanism of action for preventing allergic reactions.

Results and Discussion
Inhibition of IgE-mediated anaphylactic reactions by Fab-6HD5. To explore all avenues for preventing allergic diseases using anti-IgE therapy, we focused on a Fab fragment of a monoclonal anti-IgE antibody (Fab-6HD5) and investigated its in vivo effects on IgE-mediated anaphylactic reactions using a passive cutaneous anaphylaxis (PCA) assay. First, we injected serial dilutions of anti-dinitrophenyl (DNP) IgE (SPE-7) 28 intradermally into rats. Twenty-four hours later, several dilutions of anti-IgE antibodies (6HD5, Fab-6HD5, HMK-12, and Fab-HMK-12, with rat IgG as a negative control) were injected into the same sites. Following an extravasation assay with Evans blue and DNP-BSA, the results revealed that a minimal amount of Fab-6HD5 or 6HD5 (1.25 μg/ ml) could inhibit the PCA reactions (Table 1). By contrast, 4 times the amount of anti-IgE antibodies, such as HMK-12 and Fab-HMK-12 (5 μg/ml), was needed to inhibit the PCA reactions. In addition, a significant inhibition of the PCA reaction by Fab-6HD5 was obtained for another allotype, anti-trinitrophenyl (TNP) IgE (142a). However, there was no inhibition of PCA reactions with Fab-anti-κ, which suggests that Fab-6HD5 is directed against an IgE H chain constant region. Previous studies have demonstrated that omalizumab inhibits the PCA reactions at concentration of 50 μM, but the inhibitory effect is less pronounced at 5 μM 29 . In contrast, it should be noted that a small amount of Fab-6HD5 (2 μg/ml) was sufficient to completely inhibit the PCA reactions.
Long-term prevention of allergic reactions by Fab-6HD5. To gain further insights into the role of Fab-6HD5 in allergic reactions, we next investigated how long the inhibitory effects of Fab-6HD5 on PCA reactions lasted. In the first group, SPE-7 was injected intradermally into rats on day 0. Then, serial dilutions of Fab-6HD5 were injected into the same sites on day 1. Two hours later, Evans blue and DNP-BSA were injected intravenously for the challenge. In the second group, an Evans blue extravasation assay was performed on day 10. The results demonstrated that an inhibition of PCA reactions was observed at a concentration of Fab-6HD5 ranging from 0.63 μg/ml in the day-1 group and 0.32 μg/ml in the day-10 group (Table 2). These data clearly showed that inhibition of PCA reactions by this antibody lasts at least 10 days and maintains an even higher inhibition than the challenge on day 1. These findings raise an intriguing possibility that pretreatment with Fab-6HD5 may prevent allergic reactions over a long period of time. Furthermore, strong inhibition of PCA reactions was  observed at 2 hours after the injection of Fab-6HD5, which suggests that this antibody has promising effects in the treatment of anaphylactic shock.

Inhibitory effects of Fab-6HD5 on mast cell degranulation. To confirm the results obtained in vivo,
we investigated whether Fab-6HD5 could inhibit mast cell degranulation by measuring β-hexosaminidase release via spleen tyrosine kinase (Syk) activation. For this purpose, rat basophilic leukemia cells (RBL/2H3) were incubated overnight with DNP-IgE (SPE-7). Following sensitization, cells were incubated with serial dilutions of highly purified Fab-6HD5 for 2 hours at 37 °C. After washing the cells to remove excess IgE and Fab-6HD5, the cells were then stimulated with DNP-BSA to trigger degranulation. After stimulation, Syk activity and the percentage of β-hexosaminidase released were measured as described in the Methods section. Consistent with the results obtained using the in vivo PCA assay, our results demonstrated that Fab-6HD5 inhibits Syk activity and β-hexosaminidase release from RBL/2H3 cells in a dose-dependent manner in vitro (Fig. 1a,b). Notably, the optimal concentration of Fab-6HD5 (2 μg/ml) to inhibit Syk phosphorylation and β-hexosaminidase release was much lower than that of omalizumab (2 mg/ml) needed to inhibit leukotriene release in mast cells and basophils 30 . Taken together, our results obtained from in vivo and in vitro studies raise the possibility that further development of recombinant humanized anti-IgE antibodies may contribute to preventing allergic diseases with fewer side effects.

Fab-6HD5 interacts with the IgE-FcεRIα complex on the surface of rat mast cells. According
to recent studies, several anti-IgE Fab fragments inhibit IgE-mediated serotonin release from mast cells 31,32 . These antibodies are thought to recognize the binding sites of IgE for FcεRIα. We have previously shown that the Cells were further incubated with highly purified 6HD5-Fab (2 μg/ml) or control IgG2a overnight. After washing, cells were stimulated with TNP26-BSA (100 ng/ml) for indicated periods and cell extracts were subjected to Western blotting. Proteins were detected with antiphosphorylated Syk, anti-Syk, and anti-Tpt1 antibodies followed by HRP-conjugated anti-rabbit or anti-mouse antibody.
Fab-6HD5 is directed against the IgE Cε2 domain. Based on the above findings, we next explored identifying the IgE epitopes recognized by Fab-6HD5. IgE is structurally similar to other immunoglobulin with two heavy and two light chains. It consists of four heavy chain constant region domains (Cε1, Cε2, Cε3, Cε4) and binds to FcεRIα on mast cells and eosinophils with extraordinarily high affinity 35 . To define the epitopes recognized by Fab-6HD5, we prepared GST fusion molecules comprising IgE Cε domains (Cε1-4, Cε1-2, Cε3-4, Cε1, Cε2) (Fig. 4a). These molecules were then subjected to Western blot analysis using Fab-6HD5 (Fig. 4b). Under reducing conditions, our results clearly demonstrated that only GST fusion molecules containing the Cε2 domain (Cε1-4, Cε1-2, Cε2) reacted with Fab-6HD5, but others containing the Cε1, Cε3 and Cε 4 domains (Cε1, Cε3-4) were found to be negative. These results indicate that Fab-6HD5 is directed against the IgE Cε2 domain.
It had been thought that the Cε2 domain is not involved in the interaction between IgE and FcεRIα. However, deletion of the Cε2 domain from the IgE H chain constant region increased the rate of dissociation of IgE from the receptor, which suggests that the IgE Cε2 domain plays a pivotal role in stabilizing the interaction of IgE with FcεRIα 36 . These findings support our previous observations that indicate that some anti-IgE antibodies, including

Conclusions
We have demonstrated that a small amount of anti-IgE antibody Fab-6HD5 strongly inhibits PCA reactions in vivo, as well as Syk activity and β-hexosaminidase release from rat basophilic cells in vitro. Flow cytometry analysis and deletion-mapping studies revealed that specific binding of Fab-6HD5 to the IgE Cε2 domain prevents allergic reactions over a long period of time through interacting with a preformed IgE-FcεRIα complex on the surface of rat mast cells. Given that the Cε2 domain plays a key role in stabilizing the interaction of IgE with FcεRIα, our findings suggest that pre-administration of a certain antibody directed against IgE CH domains may provide better options for preventing Type I hypersensitivity reactions. PCA reactions. PCA reactions were performed as previously described 38,39 . Several sites of freshly shaved skin of 2 rats for each group were injected intradermally with 100 ng/0.1 ml anti-DNP-IgE (SPE-7) or anti-TNP-IgE (142a). Twenty-four hours later, 1 μg/ml of Fab-6HD5, 6HD5, Fab-HMK-12, Fab-HMK-12, rat IgG and anti-κ antibody were injected into the same sites. Two hours after the second series of injections, the rats were injected intravenously with 1 ml of mixture comprised of equal quantities of 0.5% Evans blue dye in saline and 1 mg/ml of DNP-BSA or TNP-BSA. The reactions (the diameters of the blue spots) were measured 30 min later. The experiments were repeated three times. All experiments were performed in triplicate by investigators (T.H., A.K.) who were blinded to the measurements of the PCA reactions.

Animals and cells. Female
Time-course studies of PCA. Several sites of freshly shaved skin of 2 rats for each group were injected intradermally with 1 μg/ml of SPE-7 on day 0.
In the first group, serial dilutions of Fab-6HD5 were injected into the same site on day 1. Two hours later, the rats were injected intravenously with 1 ml of mixture comprised of equal quantities of 0.5% Evans blue dye in saline and DNP-BSA (1 mg/ml).
In the second group, an Evans blue extravasation assay was performed on day 10. The reactions (the diameters of the blue spots) were measured 30 min later.
Monoclonal antibodies. SPE-7 (anti-DNP murine IgE antibody) and 141a (anti-TNP murine IgE antibody) were purchased from Sigma-Aldrich Co. LLC (USA) and Pharmingen (USA), respectively. The rat monoclonal antibodies 6HD5 and HMK-12 (specific for murine IgE) have been described previously 33,37 . The rat IgG anti-murine κ antibody and phycoerythrin (PE)-labeled goat anti-rat IgG were purchased from BioLegend (CA, USA) and Jackson ImmunoResearch (PA, USA) respectively. Fab fragments of 6HD5 and HMK12 were prepared by Immune-Biological Laboratories Co. Ltd (Gunma, JAPAN). Highly purified Fab fragments of 6HD5 were prepared using Ficin, a cysteine protease isolated from fig latex. Briefly, 6HD5 (1 mg/ml) was digested with 0.5 ml of the immobilized Ficin resin slurry (Thermo scientific prod. # 44881) twice in the presence of 25 mM cysteine. The resulting Fab-6HD5 fragments were used for in vitro blocking assays of Syk phosphorylation or β-hexosaminidase release.
Antigen. The   cells/well). The adherent cells were incubated with 1 μg/ml of mouse anti DNP-IgE (SPE-7) overnight at 37 °C. Following sensitization, the cells were washed and incubated with a serial dilution of highly purified Fab-6HD5 for 2 hours at 37 °C. The cells were washed three times with HEPES-buffered Tyrode's solution, pH 7.4 (HEPES-Tyrode's solution) to remove excess IgE and anti-IgE. Then, the cells were incubated with 100 ng/ml of DNP-BSA in 0.1% BSA-HEPES-Tyrode's solution for 1 hour at 37 °C to trigger degranulation. After stimulation, aliquots (10 μl) of the supernatant from each well were incubated with 40 μl of substrate solution, p-nitrophenyl-N-acetyl-beta-D-glucosamine in 0.1 M citrate buffer, pH 4.5 (1.3 mg/ml) for 1 hour at 37 °C. The color formed by hydrolysis of the substrate to produce p-nitrophenolate was measured at 405 nm on a microplate reader (BioRad, Benchmark plus). The total cellular content of β-hexosaminidase was measured in 1% TritonX-100 cell lysates.

Preparation of GST-IgE H chain constant region fusion molecules.
Total RNA was isolated from the mouse IgE-producing hybridoma SPE-7 and reverse-transcribed using oligo dT primers. The coding gene for the IgE H chain constant region (Cε1-4) and truncated regions were synthesized from the cDNA and subcloned into pGEX-6P-3 (GE Healthcare). The PCR primer sets and primer sequence information are listed as Table S1. Expression vectors for GST-IgE H chain constant region fusion molecules were introduced into the BL21(DE3) pLysS E. coli strain, and expression of the GST-fusion molecules was induced by isopropyl-β-D -1-thiogalactopyranoside (IPTG, 0.2 mM final) for 2 hours at 32-37 °C.
Western blot analysis. Samples of GST-IgE H chain constant fusion molecules (Cε1-4) were run on 12.5% of SDS-PAGE under reducing conditions, transferred to a PVDF membrane (Millipore corp.) and probed with Fab-6HD5 followed by peroxidase-conjugated goat anti-rat IgG, F(ab')2 fragment specific (Jackson ImmunoResearch, 1:20000). Signals were developed using Pierce ECL Plus Western Blotting Substrate (Thermo 32132). Immunoreactivity was detected with an Imagequant LAS 4000 (GE Healthcare). To remove the first antibody, the membranes were soaked in stripping buffer (2% SDS, 100 mM mercaptoethanol in 62.5 mM Tris-HCl pH 6.8) at 65 °C for 30 min with occasional shaking. After washing the membranes three times with 0.05% Tween-PBS and blocking them with BlockAce (DS Pharma Biomedical), the membranes were re-probed with anti-GST-tag pAb (MBL, 1:1000) followed by HRP-conjugated anti-rabbit IgG (Jackson ImmunoResearch, 1:20000). The ECL reaction and acquisition of chemiluminescent signals from immunoblots were performed as described above. For Syk phosphorylation assay, blots were probed with respective antibodies, anti-phosphorylated Syk, anti-Syk, and anti-Tpt1 followed by HRP-conjugated anti-rabbit or anti-mouse antibody.
Statistical analyses. Data generated from the β-hexosaminidase release assay were analyzed by an unpaired t test using GraphPad PRISM to identify significant differences. P values < 0.05 were considered significant.