Bovine serum albumin, human serum albumin, piperazine-N,N’-bis (2-ethanesulfonic acid) (Pipes), L-glutamine, antibiotic-antimycotic solution (10,000 IU penicillin, 10 mg/mL streptomycin, and 25 µg/mL amphotericin B), collagenase (Worthington Biochemical Corp., Lakewood, NJ, USA), Hanks’ balanced salt solution, fetal calf serum (FCS) (Thermo-Fisher, Grand Island, NY, USA), pronase (Merck Millipore, Burlington, CA, USA), RPMI 1640 with 25 mM HEPES buffer, Eagle’s minimum essential medium (Fuji Film, Research Triangle Park, NC, USA), Percoll (Pharmacia Fine Chemicals, Uppsala, Sweden), CD117 MicroBeads (Miltenyi Biotech, Bologna, Italy), Iscove modified Dulbecco Medium (IMDM) (Fuji Film, Research Triangle Park, NC, USA), HClO₄ (Baker Chemical Co., Deventer, Netherlands), hyaluronidase, chymopapain, elastase type I, cysteinyl leukotriene C₄ (LTC₄), and prostaglandin D₂ (PGD₂) (Sigma Chemical Co., St. Louis, MO), deoxyribonuclease I (Merck Millipore, Burlington, CA, USA), (³H)-LTC₄ and (³H)-PGD₂ (New England Nuclear, Boston, MA) were commercially purchased. Rabbit IgG anti-IgE antibody, produced by rabbit immunization with the Fc fragment of a human IgE myeloma (patient PS) and then absorbed with the IgE Fab as previously described (37), was kindly donated by Drs. Kimishige and Teruko Ishizaka (La Jolla Institute for Allergy and Immunology, La Jolla, CA). Rabbit anti-LTC₄ and anti-PGD₂ antibodies were a gift of Dr. Lawrence M. Lichtenstein (The Johns Hopkins University, Baltimore, MD). The Pipes buffer used in these experiments was a mixture of 25 mM Pipes, 110 mM NaCl, 5 mM KCl, pH 7.37, referred to as P. P2CG contains, in addition to P, 2 mM CaCl₂ and 1 g/L dextrose (66); pH was titrated to 7.4 with NaHCO₃.

The study was approved by the Ethics Committee of the University of Naples Federico II, School of Medicine (Prot. 198/18), and informed consent was obtained from all participants prior to collection of blood according to recommendations from the Declaration of Helsinki. Serum samples from six patients with AD (aged 5 to 17 years) and six normal donors (aged 6 to 22 years) were collected and stored at -20°C. Patients with AD had similar clinical pictures, characterized by a chronic, pruritic skin eruption marked by erythema, papules, or lichenification of flexural areas of the extremities, face and neck (67). Serum samples were obtained from these patients after not taking any drug for at least one week.

IgE myeloma protein was purified from a myeloma patient (68) by gel filtration on Sepharose G-200 followed by elution through a Sepharose CL-4B column. Analysis by sodium dodecyl-sulfate polyacrylamide gel electrophoresis of purified human monoclonal IgE proteins demonstrated a single protein with a m.w. of 180,000-200,000 D. Analysis by radioimmunoassay showed no IgG, IgM, or IgA contamination (38, 69, 70).

Human IgG were purified by precipitation of human serum with 50% saturated NH₄SO₄ followed by chromatography on a DEAE-cellulose column equilibrated with 0.01 M phosphate buffer (pH 7.9), as previously described (70, 71).

Comparable levels of IgG anti-IgE antibodies were detected in serum samples from the six AD patients studied, which averaged 1,020 ng/ml (± 135 ng/ml), much higher than in nonatopic controls (< 50 ng/ml) (45). For affinity purification of these autoantibodies, sera (3 ml for each run) were passed through an immunosorbent Sepharose 4B column (1.2 x 5 cm) coated with IgE purified from ADZ (45). Immunosorbent-bound Ig with anti-IgE activity were eluted with glycine HCl buffer 0.2 M (pH 2.8), and the pH was rapidly readjusted by the addition of 2 M NaOH. The total content of immunoglobulins of the eluted fraction was measured by radioimmunoassay. Anti-IgE activity belonged to the IgG isotype. IgE content was less than 0.05 U/ml. The specificity and activity of IgG anti-IgE were tested as described elsewhere (45).

Basophils were purified from peripheral blood of healthy volunteers, aged 19-45 years, undergoing hemapheresis within the Immunohematology Unit at the University of Naples Federico II. Buffy coats were subjected to double-Percoll density centrifugation, which produced basophil-depleted and basophil-enriched cell suspensions (72). Basophils were purified from the basophil-enriched cell suspensions using the Basophil Isolation Kit II (Miltenyi, Biotec, Bologna, Italy). Basophils, with purity ≥ 95%, assessed by Alcian blue staining, were incubated in IMDM in the presence of activating stimuli for 4 hours (IL-4 secretion) or 18 hours (IL-13 secretion) at 37°C (38). At the end of these incubations, the cell-free supernatants were stored at -20°C for subsequent assay of IL-4 and IL-13.

The study was approved by the Ethics Committee of the University of Naples Federico II (Protocol: Human MC No. 7/19) and informed consent was obtained from all donors. Skin obtained from patients undergoing either mastectomy for breast cancer or elective cosmetic surgery was separated from the subcutaneous fat by blunt dissection. The tissue was finely cut into 1- to 2-mm fragments and dispersed into single-cell suspension as previously described (73). Yields with this technique ranged between 0.1 and 0.9 × 10⁶ mast cells/g of wet tissue, and purity was between 5 and 10%. Human skin mast cells (HSMCs) were further purified using a CD117 MicroBead kit cell sorting system (Miltenyi Biotec, Bologna, Italy) according to the manufacturer’s instructions. Mast cell purity using this technique ranged from 36 to 71% as assessed by Alcian blue staining.

Human lung mast cells (HLMCs) were purified from macroscopically normal lung tissue obtained from patients [hepatitis C virus (HCV−), hepatitis B surface Ag (HBsAg−), HIV−] affected by lung adenocarcinoma undergoing thoracic surgery (74, 75). Freshly resected lung tissue was obtained intraoperatively and was minced finely with scissors and washed extensively with Pipes buffer over Nytex cloth (120-μm pore size) (Tetko, Elmsford, NY, USA). The cells were suspended (10⁶ cells/mL) in RPMI 1640 with 5% FCS, 2 mM L-glutamine, and 1% antibiotic-antimycotic solution and incubated in 24-well plates (Falcon, Becton Dickinson, Milan, Italy). The enzymatic tissue dispersion yielded ≈5 × 10⁵ mast cells/gram of lung tissue and purity ranged from 4% to 19% (40). HLMCs were further purified using a CD117 MicroBead kit cell sorting system (Miltenyi Biotec, Bologna, Italy) according to the manufacturer’s instructions (40). Mast cell purity using this technique ranged from 58% to 82% as assessed by Alcian blue staining.

Whole blood samples were processed immediately after collection to obtain leukocyte-enriched preparations (76, 77). Duplicate leukocyte aliquots were incubated (45 minutes at 37°C) in P2CG buffer with increasing concentrations of rabbit IgG anti-human IgE myeloma (patient PS; anti-IgE) or human IgG anti-IgE. Cell-free supernatants were collected and stored at −20°C for subsequent assay of histamine content using an automated fluorometric technique (78). Histamine release (HR) was expressed as percent of the total content assessed in parallel samples lysed by addition of 2% HClO₄, minus the basal, or spontaneous release (77). Percent HR values were the means of duplicate determinations, differing by <5%. Basophil reactivity, that is, the maximal percent histamine release (HR_MAX), and threshold sensitivity (HR_SENS), that is, 100x the inverse of the secretagogue concentration inducing half-maximal HR (EC₅₀), were calculated as described (76, 79–81).

HSMCs or HLMCs (≈3 × 10⁴ mast cells per tube) were resuspended in P2CG. 0.3 mL of the cell suspensions were placed in 12 × 75 mm polyethylene tubes. 0.2 mL of each prewarmed releasing stimulus was added, and incubation was continued at 37°C for 45 min (40, 41). At the end of incubation, cells were centrifuged (1000× g, 4°C, 5 min) and supernatants were stored at –20°C for subsequent assay of histamine content. Histamine was measured in duplicate determinations with an automated fluorometric technique (78).

IL-4 and IL-13 were assessed in duplicate samples using ELISA kits according to manifacturer’s instructions (Quantikine ELISA Kit) (R&D Systems, Minneapolis, MN, USA). The ELISA detection range was 31-2,000 pg/ml (IL-4) and 125-4,000 pg/ml (IL-13).

LTC₄ and PGD₂ were measured in duplicate samples by radioimmunoassay (40, 82). The anti-LTC₄ and anti-PGD₂ antibodies are highly selective, with less than 1% cross-reactivity to other eicosanoids (82, 83).

Data were analyzed with the GraphPad Prism 8 software package (GraphPad Software, La Jolla, CA, USA). Values were expressed as mean ± SEM (standard error of the mean). Statistical analysis was performed using Student’s t-test or one-way analysis of variance. Values were considered significant when the probability was below the 5% confidence level (p < 0.05).

In a first group of experiments, we compared the effects of increasing concentrations of human IgG anti-IgE purified from the sera of six patients with AD, rabbit IgG anti-IgE and human polyclonal IgG on HR from human basophils. Figure 1A shows that increasing concentrations (10^-4 to 3 x 10^-2 μg/ml) of human IgG anti-IgE isolated from only one out of six AD patients, as previously described (44), induced the release of substantial amounts of histamine from basophils isolated from six different normal donors. Shown for comparison is the concentration-dependent release of histamine induced by higher concentrations of rabbit IgG anti-IgE (10^-3 to 3 x 10^-1 μg/ml) in parallel experiments with the same basophil preparations ( Figure 1B ). Similarly, in the same experiments, non-functional human IgG anti-IgE purified from the other five AD patients did not induce HR from basophils ( Figure 1C ). In these experiments, human polyclonal IgG (10^-3 to 3 μg/ml) purified from six healthy donors failed to induce mediator release from basophils ( Figure 1D ). Basophil reactivity, that is the maximal percent HR (HR_MAX) in response to human IgG anti-IgE (70.0% ± 3.80%), was similar to basophil reactivity to rabbit IgG anti-IgE (65.8% ± 3.68%). By contrast, the secretagogue concentration inducing half-maximal histamine release (EC₅₀) induced by the functionally active human anti-IgE preparation (2.4 x 10^-3 ± 5 x 10^-4 μg/ml) was significantly lower than the corresponding concentration of rabbit anti-IgE (4 x 10^-2 ± 1 x 10^-2 μg/ml), hence resulting in significantly higher HR_SENS (p < 0.05). These results indicate that one preparation of the human IgG anti-IgE preparations tested was active on human basophils. This preparation of human IgG anti-IgE is from now on referred to as “human anti-IgE”.

IgE cross-linking induced by rabbit or goat anti-IgE (57, 58, 60, 61, 72, 84–87) or superantigens (38, 39, 59) can induce the production of IL-4 and IL-13 from human basophils. In a series of parallel experiments, we compared the effects of human and rabbit anti-IgE on the release of IL-4 and IL-13 from peripheral blood basophils purified (> 95%) from healthy donors. Figure 2 shows the results of five independent experiments in which we examined the effects of increasing concentrations (10^-3 to 10^-1 μg/ml) of human and rabbit anti-IgE. In these experiments, basophils were incubated 4 hours at 37°C to evaluate IL-4 release, whereas they were incubated 18 hours at 37°C to examine IL-13 production, as previously reported (38, 39, 60, 72). Both preparations of anti-IgE induced a concentration-dependent release of IL-4 ( Figure 2A ) and IL-13 ( Figure 2B ). However, human anti-IgE, at all tested concentrations, was more effective than the corresponding concentrations of rabbit anti-IgE in inducing the release of both IL-4 and IL-13 from basophils. IgG with anti-IgE activity obtained from the other five AD patients did not cause IL-4 and IL-13 release from human basophils (data not shown). Similarly, human polyclonal IgG obtained from six normal donors did not induce cytokine release from basophils (data not shown).

The ability of human and rabbit anti-IgE to trigger basophil mediator release suggested that it might interact with basophil-bound IgE. To test this hypothesis we conducted experiments to verify whether soluble human monoclonal IgE purified from a myeloma patient (68) (70) might inhibit the mediator response to human and rabbit anti-IgE. To this end, basophils were preincubated (10 min at 37°C) with increasing concentrations of human IgE and the cells were incubated for an additional 30 min at 37°C in the presence of increasing concentrations of human or rabbit anti-IgE. Figure 3 illustrates the results of typical experiments showing that preincubation with increasing concentrations of human monoclonal IgE concentration-dependently shifted to the right effects on basophil HR of both human ( Figure 3A ) and rabbit anti-IgE ( Figure 3B ). Preincubation (10 min at 37°C) of human basophils with tenfold higher concentrations of human polyclonal IgG did not interfere with either human ( Figure 3C ) or rabbit anti-IgE effects ( Figure 3D ). Similar results were obtained in three additional experiments. The parallel shift to the right of the HR curve caused by increasing concentrations of human monoclonal IgE on both human and rabbit anti-IgE, without changes in maximal efficacy, suggested that it might act as a competitive inhibitor.

In five parallel experiments, we compared the activating properties of human and rabbit anti-IgE on HR ( Figure 4A ) and de novo synthesis of PGD₂ by HSMCs ( Figure 4B ). The maximal percent HR caused by human anti-IgE (17.8 ± 0.91%) was similar to that induced by rabbit anti-IgE (20.2 ± 2.8%). Similarly, the maximal production of PGD₂ induced by human anti-IgE (31.1 ± 3.7 ng/10⁶ cells) was comparable to that caused by rabbit anti-IgE (30.5 ± 2.6 ng/10⁶ cells). By contrast, the secretagogue concentration inducing half-maximal histamine release (EC₅₀) for histamine release was significantly lower (5 x 10^-2 ± 1 x 10^-2 μg/ml) for human anti-IgE compared to rabbit anti-IgE (2.5 x 10^-1 ± 6 x 10^-2 μg/ml) (p < 0.05), indicating a comparably higher HR_SENS. Similarly, the EC₅₀ for PGD₂ production caused by human anti-IgE (7.2 x 10^-2 ± 2.1 x 10^-3 μg/ml) was significantly lower than that of rabbit anti-IgE (2.9 x 10^-1 ± 3 x 10^-2 μg/ml) (p < 0.05).

In five experiments, we compared the effects of increasing concentrations of human and rabbit anti-IgE on HR and de novo synthesis of LTC₄ and PGD₂ from HLMCs. Increasing concentrations (10^-2 to 3x10^-1 μg/ml) of human or rabbit anti-IgE (10^-1 to 3 μg/ml) caused a concentration-dependent release of histamine from HLMCs ( Figure 5A ). The maximal percent HR in response to human anti-IgE (18.4% ± 1.8%) was similar to HLMC reactivity to rabbit anti-IgE (20.2% ± 1.2%). By contrast, the EC₅₀ was significantly lower (4.6 x 10^-2 ± 4 x 10^-3 μg/ml) for human compared to rabbit anti-IgE (3.4 x 10^-1 ± 8 x 10^-2 μg/ml) (p < 0.01). In these experiments, we also compared the effects of human and rabbit anti-IgE on the de novo synthesis of LTC₄ and PGD₂ from HLMCs. Figure 5B shows that the maximal production of LTC₄ by HLMCs exposed to human anti-IgE (40.9 ± 2.2 ng/10⁶ cells) was similar to that caused by rabbit anti-IgE (42.5 ± 2.0 ng/10⁶ cells). By contrast, the concentration of human anti-IgE inducing half-maximal LTC₄ release was significantly lower (4.0 x 10^-2 ± 4 x 10^-3 μg/ml) than the EC₅₀ for rabbit anti-IgE (2.5 x 10^-1 ± 6 x 10^-2 μg/ml) (p < 0.05). Similarly, HLMC reactivity to human anti-IgE (31.4 ± 2.6 ng/10⁶ cells) was similar to rabbit anti-IgE (38.9 ± 3.0 ng/10⁶ cells) with respect to PGD₂ production ( Figure 5C ). The EC₅₀ for PGD₂ production caused by human anti-IgE (4.2 x 10^-2 ± 1 x 10^-3 μg/ml) was significantly lower than that of rabbit anti-IgE (2.8 x 10^-1 ± 8 x 10^-2 μg/ml) (p < 0.05).