Nasal polyp tissues were obtained at surgery done as a part of the treatment for the removal of nasal polyps (polypectomy/functional endoscopic sinus surgery). Nasal mucosal specimens were obtained at surgery (conchotomy) performed for the treatment of hypertrophied turbinates. Both nasal polyp specimens and nasal mucosal tissue specimens (3×4 mm) were rinsed in phosphate-buffered saline (PBS; pH 7.6), and then processed for immunohistochemistry as described below.

The nasal polyp and nasal mucosa tissue specimens were first fixed in periodate lysine paraformaldehyde (PLP), and then incubated for 4 hours each in gradients of sucrose that varied from 10% to 20% in PBS. Subsequently, the specimens were embedded in Tissue Tech OCT compound, frozen in liquid nitrogen, and stored at -80℃ until further use. To investigate the expression of FOXP3 and to localize the protein in the nasal polyps and nasal mucosa, immunohistochemistry was performed.

For immunohistochemistry, frozen sections of PLP-fixed specimens were cut at 5-µm thickness using a cryostat (Sakura Finetek, Tokyo, Japan), air-dried for 30 minutes, and then fixed in acetone for 10 minutes. Immunohistochemistry was performed using the peroxidase-based avidin-biotin complex (ABC) method (Vectastain ABC kit; Vector Laboratories, Burlingame, CA, USA). Briefly, acetone-fixed sections were rehydrated by incubation in Tris-buffered saline (TBS; pH 7.4). Thereafter, the specimens were incubated in 10% H₂O₂ for 30 minutes to block endogeneous peroxidase. After two rinses in TBS, the specimens were incubated for 45 minutes in Protein Block Serum Free Ready to Use (Dako, Glostrup, Denmark). After two rinses in TBS, the sections were incubated overnight at 4℃ in the optimal concentration of primary mouse monoclonal antibody against FOXP3 (5 µg/mL) (clone 236A/E7, sc-56680, lot # K2707; Santa Cruz Biotechnology, Santa Cruz, CA, USA). IHC staining was then performed using the HRP method according to the manufacturer's instructions. After rinsing in TBS that contained 0.1% Tween-20 (TBS-t; pH 7.6), the sections were incubated in the ABC. The sections were then rinsed in TBS-t. Specific immunoreactivity of the anti-FOXP3 antibodies was visualized by incubating the tissues in multiple-adsorption biotinylated secondary horse anti-mouse IgG antibody (1:200 dilution, Vectastain Elite ABC Reagent; Vector Laboratories) and the 3-amino-9-ethyl carbazole (AEC) substrate chromogen (Ready to Use kit; Dako). All rinses were carried out in TBS-t (pH 7.6) at room temperature. The reaction was stopped by immersing in distilled water. The sections were counterstained with Mayer's hematoxylin and mounted in glycergel mounting medium (C0563; Dako). As a negative control, the primary antibodies were substituted with an irrelevant isotype-matched mouse IgG (IgG1; Dako).

Although FOXP3 is a biomarker of Tregs, it also expressed in some non-CD4+ cells.16 To confirm that the FOXP3+ cells are Tregs, we assessed FOXP3 and CD4 expression by double-immunohistochemistry using the labeled streptavidin biotin (LSAB) and alkaline phosphatase methods, so as to identify FOXP3+ CD4+ cells.15 Processing of the tissue samples for double-immunohistochemistry was performed as previously described for single staining for FOXP3. For double-immunohistochemistry, we performed the first immunostaining with HRP, as in the previous single immunohistochemistry, using HRP label detection with 3,3'-diaminobenzidine (DAB-brown) for the first primary FOXP3 mouse mAb (clone 236/E7; AbD Serotec, Kidlington, UK). Subsequently, the tissue sections were incubated in an elution buffer of 10 mM citrate and 0.05% Tween-20 (pH6.0), which had been heated to 60℃ for 10 minutes before the second immunohistochemistry, which involved the alkaline phosphatase-labeled detection system with Fast Red Substrate Systems (Dako) for the anti-CD4 mouse mAb (clone MT 310, sc-19641; Santa Cruz Biotechnology). For the negative control, the second primary antibodies were substituted with an irrelevant isotype-matched mouse IgG (IgG1) (Dako).

The numbers of positively stained FOXP3+ cells and FoxP3+ CD4+ cells were counted under an Olympus microscope in six randomly selected visual fields using an objective micrometer in an area of 0.202 mm² and at a magnification of 400 HPF. The average for the six fields was taken as the cell count for that section.

Epithelia and lamina propria were evaluated separately. The epithelium, defined as the area between the cilia and basal membrane, was evaluated by counting positive cells per 0.0625 mm². The lamina propria, defined as the area immediately underneath the basal membrane, was evaluated by counting positively stained cells in six randomly selected areas per high-power field (area: one high-power field at 400× is equivalent to 0.0625 mm²). The results are expressed as the number of cells per 0.0625 mm² in the epithelial and subepithelial areas and compared between the nasal mucosa and nasal polyp tissues.

The data are presented as mean±SD. Statistical analyses were carried out using the Student's t-test. Differences were considered statistically significant only when the P-values were less than 0.05. All statistical analyses were performed using the SPSS ver. 15 software (SPSS, Chicago, IL, USA).

We analyzed the expression of FOXP3 in nasal polyps. FOXP3 expression was detected in the inflammatory cells in the epithelium and lamina propria (Fig. 1A). The negative control did not show any immunoreactivity for FOXP3 (Fig. 1B).

We analyzed the expression of FOXP3 in the nasal mucosa of patients with allergic rhinitis. FOXP3 expression was detected in the inflammatory cells in the epithelium and lamina propria (Fig. 2A). The negative control did not show any immunoreactivity for FOXP3 (Fig. 2B).

A large proportion of the CD4+ cells expressed FOXP3 (Fig. 3A and B). Almost all the FOXP3+ cells were CD4+. The negative control (isotype-matched IgGs) did not show immunoreactivity for either FOXP3 or CD4 (Fig. 3C).

The number of FOXP3+ cells in the epithelia of the nasal polyps group was 2.39±2.33 cells per 0.0625 mm². In the lamina propria, the number of FOXP3+ cells was 2.79±1.74 per 0.0625 mm² (Fig. 4A). No statistically significant difference was detected for the number of FOXP3+ cells between the epithelia and lamina propria.

The number of FOXP3+ cells in the epithelia of the nasal mucosa of allergic patients was 3.60±2.64 cells per 0.0625 mm². The number of FOXP3+ cells in the lamina propria of the nasal mucosa of allergic patients (5.99±4.29 cells) was significantly higher than that in the epithelial layer (P=0.009) (Fig. 4B).

The numbers of FOXP3+ cells were significantly lower in the lamina propria of the nasal polyps than in the lamina propria of the nasal mucosa of the AR patients (2.79±1.74 vs. 5.99±4.29, P=0.008) (Fig. 5A). Although the number of FOXP3+ cells was higher in the epithelia of the allergic nasal mucosa, as compared to that in the nasal polyps (3.60±2.64 vs. 2.39±2.33), the difference was not statistically significant (Fig. 5B).

There was no difference in the numbers of FOXP3+ cells in the nasal polyps obtained from atopics and non-atopics, with respect to both the epithelia (2.70±1.90 vs. 2.07±2.71) and lamina propria (3.52±2.04 vs. 2.25±1.41).

Double immunohistochemistry for FOXP3 and CD4 was performed to detect cells that stained positively for both FOXP3 and CD4. The number of FOXP3+CD4+ cells was significantly higher in the nasal mucosa of the AR patients than in the nasal polyps (84.83%±5.73% vs. 63.18%±15.59%, P=0.002) (Fig. 6).