Antibodies and superantibodies in patients with chronic rhinosinusitis with nasal polyps

Abstract

Background: Chronic rhinosinusitis with nasal polyps is associated with local immunoglobulin hyperproduction and the presence of IgE antibodies against Staphylococcus aureus enterotoxins (SAEs). Aspirin-exacerbated respiratory disease is a severe form of chronic rhinosinusitis with nasal polyps in which nearly all patients express anti-SAEs.

Objectives: We aimed to understand antibodies reactive to SAEs and determine whether they recognize SAEs through their complementarity-determining regions (CDRs) or framework regions.

Methods: Labeled staphylococcal enterotoxin (SE) A, SED, and SEE were used to isolate single SAE-specific B cells from the nasal polyps of 3 patients with aspirin-exacerbated respiratory disease by using fluorescence-activated cell sorting. Recombinant antibodies with "matched" heavy and light chains were cloned as IgG₁, and those of high affinity for specific SAEs, assayed by means of ELISA and surface plasmon resonance, were recloned as IgE and antigen-binding fragments. IgE activities were tested in basophil degranulation assays.

Results: Thirty-seven SAE-specific, IgG- or IgA-expressing B cells were isolated and yielded 6 anti-SAE clones, 2 each for SEA, SED, and SEE. Competition binding assays revealed that the anti-SEE antibodies recognize nonoverlapping epitopes in SEE. Unexpectedly, each anti-SEE mediated SEE-induced basophil degranulation, and IgG₁ or antigen-binding fragments of each anti-SEE enhanced degranulation by the other anti-SEE.

Conclusions: SEEs can activate basophils by simultaneously binding as antigens in the conventional manner to CDRs and as superantigens to framework regions of anti-SEE IgE in anti-SEE IgE-FcεRI complexes. Anti-SEE IgG₁s can enhance the activity of anti-SEE IgEs as conventional antibodies through CDRs or simultaneously as conventional antibodies and as "superantibodies" through CDRs and framework regions to SEEs in SEE-anti-SEE IgE-FcεRI complexes.

Keywords: Chronic rhinosinusitis with nasal polyps; Staphylococcus aureus enterotoxin; aspirin-exacerbated respiratory disease; basophil; superantibody; superantigen.

Fig E1

Cloning strategy for constructing recombinant IgG₁ expression vector. Paired VH and VL regions from sorted single SAE-specific B cells were amplified by means of RT-PCR and sequenced as described in the Methods section in this article's Online Repository. The 15-bp homologous arm was introduced by means of PCR to permit cloning into pIgG₁(κ) or pIgG₁(λ) vectors by means of homologous recombination with the Seamless method, as described in the Methods section in this article's Online Repository. C_L, Constant region of the light chain; J_H, J region of the heavy chain; J_L, J region of the light chain; L, leader; SL, seamless.

Fig E2

Purity of recombinant SAE-specific IgG₁, IgE, and Fab antibodies. All recombinant SAE-specific IgG₁(A), IgE (B), and Fab (C) antibodies were transiently expressed in Expi293F cells and purified by means of protein A, anti-IgE, or LambdaFabSelect affinity chromatography, respectively. Five micrograms of each antibody was fractionated by using 12% SDS-PAGE under reducing conditions before Coomassie Blue staining.

Fig E3

Amino acid sequence alignment for 6 cloned SAE-specific antibodies against germline sequences. The V genes encoding the VH (A) and VL (B) of SAE-specific antibodies were identified by using IMGT/V-Quest. Somatic hypermutations that resulted in amino acid changes are indicated.

Fig E4

1G2 and 203 bind to distinct epitopes on SEE. Each well of a 96-well ELISA plate was coated with 1G2 Fab or IgE antibodies at 1 μg/mL (50 μL/well), followed by saturating 1G2 antibody binding sites with SEE at 1 μg/mL (100 μL/well). The 1F3, 1G2, and 203 IgG₁ antibodies were added to each well, followed by incubation with peroxidase-conjugated goat anti-human IgG/Fc.

Fig E5

Expressed single-cell IgG₁ clones with IgE relatives identified by using next-generation sequencing (NGS) of B-cell repertoire. The immunoglobulin VH gene repertoire in the nasal polyps from patient HPK-014 was captured on the Illumina 2x300 MiSeq System and bioinformatically analyzed (manuscript in preparation). The sequences in the HPK-014 NGS data set were aligned with the recombinant IgG₁ antibodies expressed from 18 FACS-sorted SAE⁺ B cells (SAE⁺CD19⁺CD138⁻7-AAD⁻), and their clonal relatedness was determined by using clone-specific CDR3 DNA motifs. Two clonal families were identified, showing 2 expressed recombinant IgG₁ antibodies (A: HPK-014_1A6 and B: HPK-014_2A8; Sequence ID in blue) related to IgE (sequence ID in red) and other IgG mutants in the HPK-014 NGS data set. For illustration purposes, not all members of the clones isolated by using NGS are shown.

Fig 1

Isolation of SAE⁺ B cells from nasal polyps by means of FACS. Cells were labeled with a mixture of biotinylated SEA, SED, and SEE and phycoerythrin-coupled anti-CD138, allophycocyanin-coupled anti-CD19, and 7-AAD. A, The lymphocyte population was selected for size and granularity. FSC, Forward scatter; SSC, side scatter. B, Live B cells were identified as CD19⁺7-AAD⁻ cells. C-E, Fluorescence minus one (FMO) control (Fig 1, C) and no SAE control (Fig 1, D) were used to select SAE⁺CD19⁺CD138⁻7-AAD⁻ cells, which constituted 0.25% of the total B-cell population (Fig 1, E). The No SAE control indicates that the cells were stained with all fluorescent antibodies and streptavidin–fluorescein isothiocyanate (FITC) but without biotinylated SAEs, whereas FMO control indicates cells stained with all antibodies but without streptavidin-FITC.

Fig 2

Six of the expressed antibodies exhibited SAE binding. The cleared supernatants collected from 293T cells, separately transfected with 40 IgG₁ expression plasmids, were tested for reactivity with SEA, SEB, SEC1, SED, SEE, TSST-1, and ovalbumin (OVA) by means of ELISA. An antibody with specificity to a particular SAE was identified with an absolute OD value greater than that of the than control IgG. The antibodies selected for further studies are shown in boldface. Control IgG is a human IgG₁ with VH4 and Vκ. We expected to select only high-affinity binders by using ELISA because of the low sensitivity of this assay.

Fig 3

Specificity and high affinity of anti-SEE and anti-SEA antibodies. A-C, Recombinant IgG₁ antibodies 1G2, 203, and 1F3 at indicated concentrations were tested for binding to SEA, SEB, SEC1, SED, SEE, and TSST-1 by using ELISA. D-F, Binding kinetics were characterized by means of SPR with biotin-labeled SEE or SEA immobilized on a streptavidin-coated sensor chip. Purified antibodies were injected at the indicated concentrations, followed by dissociation.

Fig 4

Anti-SAE IgEs mediate SAE-induced basophil degranulation. A-C, RBL SX-38 cells were sensitized with anti-SEA IgE 1B6 or 1F3 (Fig 4, A), anti-SED IgE 1A4 or 2D6 (Fig 4, B), or anti-SEE IgE 1G2 or 203 (Fig 4, C) and stimulated by the cognate SAE or anti-IgE as a positive control. D, RBL SX-38 cells were sensitized with the indicated anti-SAE IgEs and stimulated with the indicated SAEs. Degranulation was assayed based on β-hexosaminidase release.

Fig 5

Anti-SEE antibodies bind to distinct nonoverlapping epitopes. A, 1G2 IgG₁ (250 nmol/L) was injected onto a sensor chip coated with SEE for 3 minutes, immediately followed by a 3-minute injection of either running buffer (black), 500 nmol/L 1G2 IgG₁(blue), or 100 nmol/L 203 IgG₁(red). B, After regeneration of the chip by using glycine-HCl (pH 2.5), the experiment was performed in reverse, with a 3-minute binding of 100 nmol/L 203 IgG₁, followed by a 3-minute injection of either buffer (black), 500 nmol/L 203 IgG₁(red), or 100 nmol/L 1G2 (blue).

Fig 6

The unrelated anti-SEE IgG₁ or Fab enhances anti-SEE IgE–mediated basophil degranulation. RBL SX-38 cells were sensitized with anti-SEE or anti-SEA IgE antibodies and stimulated with SEE in the presence of anti-SEE (+1G2/+203) or anti-SEA (+1F3) IgG₁(A) or Fab (B). Dotted lines represent positive (upper; anti-IgE) and negative (media only) controls. Degranulation was assayed based on β-hexosaminidase release.

Fig 7

Schematic representation of proposed antibody and “superantibody” activities. A, A monomeric SEE molecule cross-links 2 FcεRI-bound IgE molecules on the surface of a basophil (receptor not shown) through 2 epitopes: the green site is recognized conventionally by the CDRs of the IgE antibody, and the red site interacts with the framework regions of the IgE V region in a superantigen manner. B, An IgG₁ antibody that recognizes a nonoverlapping blue epitope cross-links 2 FcεRI-bound IgE molecules that recognize the green epitope; both blue and green epitopes are recognized conventionally by the CDRs. C, Because the IgG₁ Fab (highlighted in darker blue) can also cross-link FcεRI-bound IgE, we propose that 2 monomeric SEE molecules can affect this: one is recognized conventionally through the 2 nonoverlapping (green and blue) epitopes by 2 unrelated anti-SEEs, and the other involves the superantigen site (red). Although the cross-linking ability of the Fab reveals the interactions that underpin this proposed mechanism, the whole antibody can act not only in the manner depicted in Fig 7, B, but also that shown in Fig 7, C. This antibody, exhibiting the ability to recognize 2 SEE molecules simultaneously through both CDRs and framework regions, we call a “superantibody.”