Transcytosis of IgE–Antigen Complexes by CD23a in Human Intestinal Epithelial Cells and Its Role in Food Allergy

CD23 expression in normal and food-allergic individuals. (A) Immunoblotting for CD23 in lysates obtained from primary human large IECs, T84 and Caco-2 human colonic epithelial cell lines, and B cells (positive control) and 293T cells (negative control). (B) Stool protein extracts from food-allergic patients (●) or nonatopic controls (□) were assayed for CD23, IL-4, IL-13, and TNF-α by ELISA. Units of measurement are indicated on the x-axis, and were normalized to the protein content in each stool extract. *P < .05 compared with nonatopic control. (C) Stool levels of food-specific IgE as measured by Immuno-CAP in stool extracts was plotted against stool CD23 content. r² = .76; P = .0011.

Regulation of CD23 expression in human IECs. (A) RT-PCR for CD23a or CD23b isoforms was performed. B cells were used as positive control. T84, Caco-2, and HT-29 human IEC lines were used, as well as primary isolated human large IECs from normal tissue (resections for cancer), or patients with ulcerative colitis (uc) or Crohn’s disease (cd). (B) Real-time PCR of CD23b or CD23a expression in B cells, human IEC lines, or primary human IECs treated with IL-4 or IL-13 for 6 hours. Data are expressed as percentage of expression in unstimulated controls. □, IL-4; ■, IL-13. (C) CD23 protein expression as measured by ELISA in cell lysates. Cells were stimulated with IL-4 or IL-13 overnight (18 hours) before preparation of cell lysates. Bars indicate mean ± SD. □, None; ■, IL-4; ▩, IL-13.

Confocal microscopy detection of CD23. T84 cells, either untransfected or transfected with CD23a or CD23b, were polarized and stained with ZO-1 (red) to show the level of the tight junction, DAPI to stain nuclei (blue), and CD23 was detected by immunostaining and detection with FITC (green). The XY sections were taken at the level of the ZO-1 staining, and the XZ reconstructions are shown at the bottom. The red staining indicates the tight junctions at the apical pole of the epithelial cells.

Transcytosis of IgE by T84 monolayers. T84 cells constitutively expressing CD23a were untransfected or transfected with CD23a or CD23b. Cells were grown on filter supports in transwells to polarize the cells. IgE was added to the apical (A→B) or basal (B→A) wells of the transwell, and supernatant from the opposite transwell was immunoblotted for IgE over time (30–90 minutes). The lower graph shows the ratio of optical density units of transfected/untransfected monolayers from the 90-minute timepoint. Mean ± SD of 3 experiments is shown. □, T84; ■, +CD23a; ▩, +CD23b.

IgE-facilitated antigen transcytosis. (A) BSA NIP was added to the apical chamber of polarized (untransfected) T84 cells in the presence (Ag-IgE) or absence (Ag) of anti-NIP IgE. Basolateral supernatant was immunoblotted for BSA NIP over time. (B) Detection of BSA-NIP appearance in basolateral supernatants over time after addition of BSA NIP plus anti-NIP IgE to untransfected T84 cells, or T84 cells transfected with CD23a or CD23b. The lower graph shows the ratio of optical density units of transfected/untransfected monolayers from the 90-minute timepoint. Mean ± SD of 3 experiments is shown. (C) Degranulation of human FcϵR1-transfected RBL cells as measured by β-hexosaminidase release. Basolateral supernatant from transwells incubated on the apical side with BSA NIP (antigen), anti–BSA-NIP-IgE (IgE), or both (Complex) were added to RBL cells. As positive control, cells were stimulated with phorbol myristate acetate (PMA) and ionomycin. Shown is a representative experiment of 2, with error bars indicating triplicate variation. □, 30 minutes; ▩, 60 minutes; ■, 120 minutes.

Effect of IL-4 and sCD23 on transcytosis of IgE and antigen–IgE complexes. (A) Polarized T84 monolayers were treated with IL-4 overnight before addition of IgE to the apical (A→B) or basal (B→A) wells of the transwell. Alternatively, IgE was preincubated with sCD23 before addition to the transwell. Supernatant from the opposite transwell was sampled over time and blotted for IgE. The graph on the right shows the ratio of optical density units of treated/untreated monolayers from the 90-minute timepoint. Mean ± SD of 3 experiments is shown. □, None; ■, IL-4; ▩, sCD23. (B) As described previously, but in place of IgE, BSA NIP plus anti-NIP IgE was added to the apical transwell, and blotting for BSA NIP was performed. The graph on the right shows the ratio of optical density units of treated/untreated monolayers from the 90-minute timepoint. Mean ± SD of 3 experiments is shown.

Intracellular trafficking of antigen. T84 cells were grown on coverslips and incubated with lysotracker (red) before addition of FITC–BSA–NIP (Ag) or FITC–BSA–NIP plus anti-NIP IgE (Ag + IgE). Lysosomes are stained red (top row), antigen is shown in green (second row), nuclei were stained with DAPI (blue, third row), and a merged image is shown in the bottom row with colocalization of antigen and lysosomes shown as yellow. Magnification, 1000×.

Effect of bafilomycin on transcytosis of IgE. Polarized T84 cells were pretreated with bafilomycin before addition of IgE (apical or basal) or IgE–antigen complexes (apical only). IgE or antigen were detected in the opposite transwell. The graph on the right shows optical density units from the 90-minute timepoint. Mean ± SD of 3 experiments is shown. □, Control; ■, bafilomycin.