Main

In atopic patients, allergic responses are characterized by high serum levels of allergen-specific IgE. It is generally believed that a spectrum of T helper 2 (Th2) type cytokines, including IL 4 (IL-4), IL-5 and IL-13, are responsible for the precipitation of these responses (1). However, the majority of neonatal T cells are of naive phenotype and produce far less cytokines as compared with memory type adult T cells. Consequently, it is not expected that these T cells provide efficient help for immunoglobulin production by B cells (2–5). Therefore, it is plausible to assume that allergic inflammatory reaction requires the cooperative effect of multiple components of inflammatory cells to initiate IgE responses within the innate immune environment. It has been recently shown that basophils and mast cells exert not only their classical granule releasing function, but they also function as efficient helper cells for B cell differentiation into IgE-secreting plasma cells. The latter function is mediated through the production of IL-4 and contact-mediated signals, including CD40 ligand expression (6–11). Therefore, the local induction of a Th2-like microenvironment by basophils or mast cells may play an important role in the first step of the cascade leading finally to efficient IgE production. These initial reactions evoked by basophils/mast cells may have profound effects on the future development of allergic inflammation by driving the host immune response toward a Th2-dominant reaction to antigens. IgE may play a key role in this scenario by amplifying the interaction between the innate and professional Th2 responses. Here, we analyzed the effect of IgE on the expression of high affinity receptors for IgE (FcεRI) on the basophil surface in children.

METHODS

Subjects and materials.

We studied 86 atopic children aged 1 mo to 17 y with allergic disorders such as bronchial asthma and atopic dermatitis. Each patient was diagnosed based on typical clinical symptoms, family history, and laboratory data. Every patient had positive IgE responses to either house dust or mite antigen. Control subjects were composed of 82 healthy age-matched nonatopic children. One milliliter of peripheral blood containing ethylenedinitrilotetraacetic acid was withdrawn from these children after parental informed consent was obtained. Neonatal cord blood samples were collected from the placental end of the umbilical cord immediately after the birth of full-term newborn babies. Cord blood leukocytes were prepared by dextran sedimentation, and basophil-rich fractions were separated by Percoll gradient centrifugation (12). Basophil-rich fractions contained 5 to 10% basophils as confirmed by May-Gründwald-Giemza staining. All samples were collected after written informed consents were obtained from the patients or from the parents. The study was approved by the institutional review board.

MAb and reagents.

Anti-FcεRI (AER37) MAb was prepared as previously described (13). FITC-conjugated anti-human IgE was purchased from Tago Inc. (Burlingame, CA). Phycoerythrin (PE)-conjugated anti-CD3 (Leu4), anti-CD4 (Leu3a), anti-CD16 (Leu11c), anti-CD20 (Leu16), and anti-CD69 (Leu23) MAbs were obtained from Becton Dickinson Immunocytometry Systems (San Jose, CA). FITC-conjugated anti-CD14 (OKM14) was from Ortho Diagnostic Systems (Tokyo, Japan). FITC-conjugated rabbit anti-mouse antibody (Zymed Laboratories, San Francisco, CA) was used as the second antibody for FcεRI staining. FITC-conjugated anti-CD17 was the product of Pharmingen (San Diego, CA). AER37 has no inhibitory activity on IgE-binding to basophils.

IgE measurement.

Serum IgE was measured by a sandwich ELISA (14). Anti-IgE MAbs (7.12 and 4.15) were obtained from A. Saxon (UCLA, School of Medicine). Alkaline phosphatase-conjugated goat anti-human IgE was the product of Kirkegaard & Perry Laboratories (Gaithersburg, MD). Purified human IgE (Chemicon International Inc., Temecula, CA) was used as a standard. The sensitivity of this IgE assay was 5 ng/mL.

Flowcytometric Analysis.

Basophil-rich fractions separated by Percoll gradient centrifugation were located exclusively within the lymphocyte region by a flow cytometry. IgE-bearing cells isolated by electronic sorting with an Epics Elite (Coulter Electronics Inc., Hialeah, FL) were morphologically determined to be >95% basophils, and IgE-negative cells had <0.05% contaminating basophils by May-Gründwald-Giemsa staining. The cells within the lymphocyte region contained <0.1% CD14+ monocytes. Preliminary experiments indicated that circulating basophils can be detected as IgE+ CD3− CD16− CD20− cells within the lymphocyte region of peripheral whole blood by flow cytometry. For the patient analysis, the whole blood samples were stained with FITC-conjugated anti-human IgE, together with PE-conjugated anti-CD3, anti-CD16 and anti-CD20 MAbs. For the analysis of FcεRI expression, the samples were first treated with AER37 and stained further with FITC-conjugated anti-mouse antibody. Unbound sites were blocked by adding normal mouse serum and the cells were stained with PE-conjugated anti-CD3, anti-CD16 and anti-CD20 MAbs. After erythrocyte lysis and washing, the cells were analyzed with a Cytoron Absolute flow cytometer (Ortho Diagnostic Systems). FITC-brightly positive and PE-negative cells within the lymphocyte region were gated. FITC mean fluorescence intensity (MFI) within the gate was determined. Representative flowcytometric analysis of basophil-bound IgE is shown in Figure 1.

Figure 1
figure 1

Flowcytometric evaluation of basophil-bound IgE. Peripheral whole blood was washed in PBS and stained with FITC-conjugated anti-human IgE, together with PE-conjugated anti-CD3, anti-CD16, and anti-CD20 MAbs. Mononuclear cell region (A) was gated for the fluorescence analysis. Furthermore, FITC-positive cell regions (B) were selected for the measurement of FITC mean fluorescence intensity (MFI), as shown in (C).

Full size image

Cell cultures.

The culture medium consisted of Roswell Park Memorial Institute 1640 medium containing 10% FCS, 20 mM HEPES, 2 × 10−5 M 2-mercaptoethanol, 0.3 mg/mL L-glutamine, 200 U/mL penicillin G and 10 mg/mL gentamicin. The basophil-enriched cord leukocytes were prepared by Percoll gradient centrifugation and seeded at 1 × 106/well in 24-well flat-bottomed plates (Corning Glass Works, Corning, NY). They were incubated with various concentrations of IgE (1 ng/mL to 10 μg/mL) at 37°C in a humidified atmosphere of 5% CO2 in air. After 3 or 48 h of culture, the cells were extensively washed in PBS to remove unbound IgE and the cell-bound IgE and FcεRI expression on neonatal basophils was analyzed by flow cytometry. In some experiments, cycloheximide (CHX; Sigma Chemical Co.) (0 to 10 μg/mL) was added to the medium at the start of culture to inhibit de novo protein synthesis.

Statistical Analysis.

Data represent the means ± SD. The statistical significance was assessed using paired t test. Correlations were analyzed by Pearson's correlation coefficient. Differences were considered significant when the p value was < 0.05.

RESULTS

Serum IgE and basophil-bound IgE.

Consistent with previous observations (15), mean concentrations of serum IgE in allergic patients were higher than in normal controls among various age groups, although atopic infants at 1 y of age or under had relatively low serum IgE concentrations. But the levels increased progressively with advancing age. In contrast, serum IgE levels in normal controls increased gradually with age and remained low at all ages (Fig. 2A). Basophil-bound IgE in allergic patients increased rapidly after birth and reached the adult maximum level (MFI 160–180) before 6 mo of age. In normal controls, basophil-bound IgE levels were low during infancy and inclined to increase with advancing age in parallel with serum IgE (Fig. 2B). There was a good correlation between basophil-bound IgE and serum IgE levels, but basophil-bound IgE plateaued at an MFI level of 160 to 180 when IgE level exceeded 300 ng/mL (Fig. 3). However, it is noteworthy that there exist relatively wide variations in basophil-bound IgE levels at given serum IgE concentrations.

Figure 2
figure 2

Comparison of serum IgE concentrations and basophil-bound IgE levels between atopic patients and normal controls from different age groups. (A) Serum IgE concentration was measured by an ELISA. (B) The level of basophil-bound IgE was expressed as MFI. Closed columns and open columns represent atopic patients and normal controls, respectively. Numbers in parentheses indicate the numbers of subjects within each age group. Data are geometric means ± SD; *p< 0.05, **p< 0.01, ***p< 0.001 vs values for age-matched controls. m, month; y, year.

Full size image
Figure 3
figure 3

Correlation between serum IgE concentration and basophil-bound IgE level. Closed circles denote the data from atopic patients and open circles, those from normal controls.

Full size image

Surface FcεRI expression on basophils.

The ability of circulating basophils to bind serum IgE is presumably determined by the level of expression of high-affinity receptors. Therefore, we examined surface FcεRIα expression on basophils to see if IgE-binding is tightly related to surface receptor expression. Indeed a good correlation was found between FcεRI expression and basophil-bound IgE levels (Fig. 4). As expected from these results, the level of surface FcεRI on basophils correlated relatively well with serum IgE levels and it plateaued when the IgE concentration exceeded 300 ng/mL (data not shown).

Figure 4
figure 4

Correlation between basophil-bound IgE levels and FcεRI expression. FcεRI expression was examined using AER37 by flow cytometry. Correlation coefficient was calculated.

Full size image

Enhancement of basophil-bound IgE and FcεRI after incubation withIgE.

Serum IgE concentration and basophil-bound IgE levels were extremely low or undetectable in cord blood. However IgE-binding cells were detectable within cord blood after 1 h of incubation with IgE. When the cells were cultured for 3 h, the MFI increased and plateaued at the level of 120 to 130 with 100 ng/mL or more of IgE (Fig. 5A). The MFI level never reached the adult level of 160 to 180 even when the added IgE concentration was increased up to 10 μg/mL. The enhancement of FcεRI expression was negligible (Fig. 5B). It is intriguing, however, that IgE-binding of basophils continued to increase dose-dependently and reached the adult level when the cells were cultured for 48 h (Fig. 5A). FcεRI expression on neonatal basophils also increased dose-dependently (Fig. 5B). After the culture, we confirmed that >95% of the cells were viable as determined by forward light scatter and Annexin-V binding analysis (16) (data not shown).

Figure 5
figure 5

IgE-induced enhancement of IgE-binding and FcεRI expression on basophils. Basophil-enriched fractions separated from neonatal cord blood were cultured in the presence of various concentrations of IgE for 3 h (open circle) or 48 h (closed circle). (A) Basophil-bound IgE levels and (B) FcεRI expression were determined by a flow cytometry. Data represent the means ± SD.

Full size image

To clarify whether the enhancement of FcεRI expression was dependent on de novo protein synthesis, we evaluated the effect of CHX on surface FcεRI expression. Enhancement of FcεRI after incubation with IgE was inhibited almost completely with 1 μg/mL or more of CHX, although CD17 expression on basophils (17) was not affected by CHX (Fig. 6).

Figure 6
figure 6

Effect of CHX on FcεRI expression. Basophil-enriched fractions from neonatal cord blood were cultured with 1 μg/mL of IgE for 48 h. CHX was added to the culture at different concentrations. The difference in MFI for FcεRI (open circle) and CD17 (closed circle) expression after culture was compared with the value obtained without added CHX and expressed as percent expression. Data represent means ± SD.

Full size image

DISCUSSION

Previous studies using acid elution of IgE have indicated that cell-bound IgE and FcεRI expression on basophils correlate with serum IgE in adult (18, 19). It is recently shown that FcεRI expression on basophils is significantly higher among atopic subjects than nonatopic controls (20, 21). In addition, there is increasing evidence suggesting that FcεRI expression on mast cells or basophils is positively regulated by IgE itself in mouse and human systems (22–25). It is therefore intriguing to study the developmental changes of cell-bound IgE and FcεRI expression on the basophil surface during early childhood, when comparisons can be made between atopic patients and nonatopic controls from different age groups.

In this study, we found that cell-bound IgE and FcεRI expression on basophils increased with advancing age in parallel with serum IgE and plateaued when the serum IgE level was greater than 300–500 ng/mL in normal children. In atopic children, the level of cell-bound IgE and FcεRI on basophils, as well as serum IgE, were higher than normal controls at all ages. In particular, basophil-bound IgE levels were greatly elevated and reached maximum levels before 6 mo of age in atopic children when serum IgE remained at relatively low levels. Consistent with previous reports (18, 19), basophil-bound IgE levels correlated closely with serum IgE concentrations. However, there existed a relatively wide range of basophil-bound IgE levels even with similar serum IgE concentration. These results suggest the possibility that basophil-bound IgE levels are influenced by the level of FcεRI expression on the basophil surface. Age-dependent increase of IgE-binding and the marked difference in basophil-bound IgE levels between atopic patients and normal controls may indicate that serum IgE has a direct effect on the regulation of FcεRI expression on circulating basophils.

Neonatal cord basophils could not be detected as IgE-bearing cells by flow cytometry because of the extremely low serum IgE levels in cord blood. However, they express low, but significant levels of FcεRI. Incubation of neonatal leukocytes with high concentration of IgE rapidly increased basophil-bound IgE levels. This increase simply reflected the saturation of the preexisting IgE receptors because no enhancement of FcεRI expression was seen after 3 h of culture. Maleveaux et al. (18) have demonstrated that after passive sensitization with IgE-rich serum, the receptors on human basophils were saturated at about 3 μg/mL. In our study, the apparent saturation of IgE binding capacity was achieved with greater than 100 ng/mL of IgE. We also showed that incubation with IgE at higher concentrations for 48 h enhanced FcεRI expression on neonatal basophils in a dose-dependent manner. Therefore, basophil-bound IgE levels after 48 h of culture were higher than after 3 h of culture. The saturation of FcεRI with monomeric IgE led to up-regulation of FcεRI without adding exogenous cytokines to the culture. Cross-linking of the receptor by antigens through cell-bound IgE was not necessary, either.

It was recently reported that IgE upregulates FcεRI expression on human basophils in vitro (25). These basophils were obtained from adults and first cultured for approximately 21 d to down-regulate FcεRI in the presence of IL-3, because the starting FcεRI density is a strong determinant of up-regulation. In this report, we used freshly isolated neonatal basophils, which expressed much less FcεRI than adult ones did. In fact, FcεRI expression on freshly isolated adult basophils was not enhanced after 48 h of culture with IgE (data not shown). It is plausible that the receptors occupied by IgE binding become more stable, and unoccupied receptors may degrade rapidly due to continuous turnover of the surface receptors. The up-regulation of FcεRI expression was completely inhibited by CHX, indicating that the increased receptor levels were dependent on de novo protein synthesis. Taken together, there exist two different mechanisms of IgE-induced up-regulation of basophil-bound IgE levels after 48 h of culture. One is through the saturation of preexisting FcεRI and the other is through the active induction of the surface receptors for IgE.

The cell preparation used in this study is composed not only of basophils but also of B cells, T cells, and trace contamination of monocytes and eosinophils, some of which may express FcεRII on the surface (20, 26). Although we could not detect FcεRI expression on monocytes in neonatal cord blood, it was recently reported that peripheral blood monocytes can express high affinity IgE receptors (21, 27). It is possible that FcεRI enhancement is the result of the cellular interaction between different types of cells within the neonatal leukocyte population. Experiments using highly purified neonatal basophils should be applied to delineate the regulatory mechanism of FcεRI expression. Further studies will be necessary to determine whether the up-regulation of FcεRI occurs at the gene transcriptional level.

Current immunologic understanding suggests that in the first step of the cascade leading finally to IgE production, the local induction of Th2 microenvironment by basophils, results in recruitment of professional Th2 type T cells and the effective production of IL-4 and IL-5 (1, 28). In this regard, it is interesting that FcεRI expression is detected at the very early stages of in vitro cultures of bone marrow cells in both mouse and human systems (29). The findings in this report also indicate that IgE-dependent enhancement of FcεRI expression plays an important role in triggering and enhancing this cascade. In this regard, it is important to regulate IgE-mediated allergic reaction by reducing allergen exposures. Furthermore, early therapeutic intervention is critical to abrogate the amplification of allergic inflammation, thus preventing excessive IgE production during early childhood.