Production and secretion of IgE by B cells, plasmablasts, and plasma cells is a central step in the development and maintenance of allergic diseases. IgE can bind to one of its receptors, the low-affinity IgE receptor CD23, which is expressed on activated B cells. As a result, most B cells bind IgE through CD23 on their surface. This makes the identification of IgE producing cells challenging. In this study, we report an approach to clearly identify live IgE+ plasmablasts in peripheral blood for application by both flow cytometry analysis and in vitro assay. These IgE+ plasmablasts readily secrete IgE, upregulate specific mRNA transcripts (BLIMP-1 IRF4, XBP1, CD138, and TACI), and exhibit highly differentiated morphology all consistent with plasmablast differentiation. Most notably, we compared the presence of IgE+ plasmablasts in peripheral blood of allergic and healthy individuals using a horse model of naturally occurring seasonal allergy, Culicoides hypersensitivity. The model allows the comparison of immune cells both during periods of clinical allergy and when in remission and clinically healthy. Allergic horses had significantly higher percentages of IgE+ plasmablasts and IgE secretion while experiencing clinical allergy compared with healthy horses. Allergy severity and IgE secretion were both positively correlated to the frequency of IgE+ plasmablasts in peripheral blood. These results provide strong evidence for the identification and quantification of peripheral IgE-secreting plasmablasts and provide a missing cellular link in the mechanism of IgE secretion and upregulation during allergy.

Immunoglobulin E–mediated allergic diseases arise when allergen-specific B cells class switch to produce IgE, which then binds to effector mast cells and basophils to promote clinical allergy (1). B cells differentiate into Ab-secreting peripheral plasmablasts before becoming plasma cells (2, 3). In fact, recent single-cell RNA sequencing suggested that the majority of human peripheral IgE+ B cells may actually be differentiating plasmablasts (4). Although this is well accepted in allergy biology, the study of IgE+ B cells and plasmablasts prove difficult in large part due to the expression of CD23, the low-affinity IgE receptor, and binding of IgE by CD23 on the cell surface of most activated B cells. Typically, identification of IgE+ B cells has required either exclusion of all other B cell classes, where IgM/IgD/IgG/IgA B cells are assumed to be IgE+ (57), or intracellular staining of IgE production (8), which prevents live cell functional analysis. The quantification of IgE-expressing B cells in these studies likely overestimated their frequency (7). An anti-human IgE mAb that can distinguish between membrane-expressed and receptor-bound IgE has also been used to label IgE+ B cells (9). IgE+ B cells measured in these ways are increased in humans with atopic dermatitis and food allergy, compared with healthy controls (5, 6), and also may increase following allergen exposure (9). However, due to the difficulty of characterizing these cells for analysis, the relationship between allergen exposure and the presence of IgE+ B cells in peripheral blood is incompletely understood. Likewise, rare peripheral IgE+ plasmablasts and plasma cells have been identified, but their relationship to clinical allergy is less defined (4, 10).

CD23 is a C-type lectin that is expressed on activated B cells in humans and other mammals, including horses (11, 12). CD23 has different roles in B cell signaling and is most often assumed to decrease IgE production and subsequent allergy (1, 13, 14). IgE binding to CD23 decreases the availability of IgE for sensitization on mast cells and basophils and also downregulates IgE production (11, 1517). However, CD23 on B cells is also capable of facilitated Ag presentation (FAP), in which allergen/IgE complexes are internalized and presented on B cell MHC class II (MHC II) molecules, or transferred to dendritic cells, to promote epitope spreading and T cell activation, thereby promoting allergic responses (1820).

Although CD23 is expressed on most activated peripheral B cells, recent single-cell sequencing of B cell surface proteins identified that some human individuals have two distinct CD23+ B cell populations (CD23lo and CD23hi) (21). This suggests heterogeneity in CD23 function on B cells. However, the cause and clinical relevance of these two populations were not explored. Similarly, another recent study noticed that allergic humans have CD23hi class-switched B cells that are absent in healthy controls (22). However, due to CD23-bound IgE, this unique CD23hi B cell population was not further characterized. In addition, increased CD23 surface density on B cells has been positively correlated with allergen-specific IgE levels in allergic individuals (23), further supporting that variations in B cell CD23 expression may occur in the context of allergy. Together, these studies suggest that CD23hi cells may have clinical relevance during allergic diseases.

Different mammalian species can experience naturally occurring IgE-mediated allergic diseases, including humans, horses, and dogs. The horse has been used as a model for human allergy due to many similarities in the immune mechanism of disease (12, 2431). Horses also allow the ability to control key variables, such as uniform environment and allergen exposure, identical living conditions for allergic and control animals, and identical treatment for allergic animals, if any. The most common allergic disease in horses is Culicoides hypersensitivity, in which horses exhibit an IgE-mediated skin allergy, characterized by alopecia, pruritus, and allergic dermatitis. The allergy develops in response to salivary allergens of Culicoides midges during environmental exposure to these biting insects in the summer (26, 3240). Horses express CD23 on peripheral B cells, and previous work has identified that most equine CD23+ B cells are either IgM+ or IgG+ B cells (12). However, this prior study only examined B cells in healthy horses without any history of Culicoides hypersensitivity. Therefore, we asked whether allergic horses express similar populations of CD23hi or IgE+ B cells as seen in humans and how this population correlates to clinical allergy. In addition, we investigated the differentiation state of CD23hi and IgE+ B cells. We hypothesized that these populations could be better defined as plasmablasts, which circulate in peripheral blood and exhibit an intermediate phenotype between B cells and plasma cells. Plasmablasts are characterized by both Ig surface expression and secretion (3), as well as upregulation of transcription factors B lymphocyte-induced maturation protein 1 (BLIMP-1), IFN regulating factor 4 (IRF4), and X-box binding protein 1 (XBP1) (41, 42).

In this study, we used horses with and without Culicoides hypersensitivity to develop an approach to clearly identify IgE+ plasmablasts without any interference by CD23-bound IgE. Additionally, we compared IgE+ plasmablasts during allergen exposure and allergy remission in allergic and healthy horses to show that IgE+ plasmablasts directly correlate to clinical allergy severity and may even precede and/or initiate the clinical phase of the allergic disease. Finally, we characterized the functional role of IgE+ plasmablasts as IgE-secreting cells and confirmed that these cells express genes indicative of peripheral plasmablasts.

Heparinized blood samples were obtained from the jugular vein using the BD Vacutainer system (BD Biosciences, Franklin Lakes, NJ). All animals studied were Icelandic horses living together in the same environment with natural exposure to Culicoides from mid-May to mid-October. Culicoides were absent from the environment of the horses for the remainder of the year. Vaccination and deworming were synchronized. All horses were annually vaccinated against rabies, tetanus, West Nile virus, and Eastern and Western encephalitis virus, as well as dewormed with moxidectin and praziquantel (Zoetis, Parsippany, NJ) once a year in December. All horses were on the same diet. They were kept full time on large pastures with run-in sheds, free access to water, mineral salt blocks, and were grazing in the summer and fed grass hay in the winter. All horses studied had either naturally occurring Culicoides hypersensitivity or were clinically healthy. Allergic horses (n = 7) included five mares (11–17 y, median 16 y), one gelding (age 9), and one stallion (age 8). Healthy horses (n = 6) included four mares (7–9 y, median 8 y) and two geldings (age 8). For CD23+ cell phenotyping experiments (Figs. 13), three allergic mares, one allergic gelding, one stallion, and four healthy mares were used. For FACS sorting and PCR analysis, two allergic mares and one allergic gelding were used. For longitudinal analysis of ex vivo PBMC (Fig. 5B), all seven allergic and six healthy horses were compared at each time point except for March, which included seven allergic and four healthy horses.

As previously described (40), all horses were given a clinical allergy score every 2–4 wk for the duration of the study. Scores were assigned based on pruritis (0–3), alopecia (0–4), and dermatitis (0–3) and total scores ≥3 were considered allergic. All allergic horses displayed allergic signs with scores >3 for at least 1 y before and for the duration of this study (Table I). Culicoides-specific hypersensitivity was further confirmed by intradermal skin testing with Culicoides whole-body extract (Stallergenes Greer, Cambridge, MA) in comparison with saline and histamine controls as previously described (27, 28). Allergic horses developed an immediate skin reaction to Culicoides whole-body extract. Nonallergic horses never exhibited clinical allergy or scoring >3 and did not react to intradermal Culicoides injections.

Table I.

Clinical scores of Culicoides hypersensitivity of allergic and healthy horses living together in the same environment

MonthaCulicoides ExposureAllergy Score, Median (Range)b
Allergic Horses (n = 7)Healthy Horses (n = 6)
February None 0 (0–0.5) 0 (0) 
March None 0 (0) 0 (0) 
April None 0 (0) 0 (0) 
May Onsetc 0 (0–1) 0 (0) 
June Constantd 5.5 (2.5–6.5) 0.75 (0–1) 
July Constantd 6 (4–9) 0.5 (0–2) 
August Constantd 5 (4.5–6) 1 (0.5–2.5) 
September Constantd 3 (3–4) 0.5 (0–1) 
October Declinee 3 (2–3) 1 (0–1) 
November None 1.5 (1–2) 0.25 (0–0.5) 
December None 1 (0–1) 0 (0–0) 
MonthaCulicoides ExposureAllergy Score, Median (Range)b
Allergic Horses (n = 7)Healthy Horses (n = 6)
February None 0 (0–0.5) 0 (0) 
March None 0 (0) 0 (0) 
April None 0 (0) 0 (0) 
May Onsetc 0 (0–1) 0 (0) 
June Constantd 5.5 (2.5–6.5) 0.75 (0–1) 
July Constantd 6 (4–9) 0.5 (0–2) 
August Constantd 5 (4.5–6) 1 (0.5–2.5) 
September Constantd 3 (3–4) 0.5 (0–1) 
October Declinee 3 (2–3) 1 (0–1) 
November None 1.5 (1–2) 0.25 (0–0.5) 
December None 1 (0–1) 0 (0–0) 
a

Boldface rows (May to October) denote months with Culicoides allergen exposure. There was no allergen exposure from February to April and November to December.

b

Clinical allergy scores ≥3 are considered allergic.

c

Culicoides exposure began in late May and clinical signs began to develop in allergic horses afterward.

d

Culicoides exposure was constant in June to September, and chronic allergy severity plateaued in allergic horses.

e

Culicoides exposure began to subside in October, and clinical signs also began to go into remission in allergic horses.

All animal procedures were carried out in accordance with the recommendation in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The animal protocol was approved by the Institutional Animal Care and Use Committee at Cornell University (protocol 2011–0011). The study also followed the Guide for Care and Use of Animals in Agricultural Research and Teaching.

Heparinized peripheral blood was left in collection tubes for ∼1 h at room temperature to allow the erythrocytes to settle. The leukocyte-rich plasma was layered on Ficoll (Ficoll Plaque Plus; GE Healthcare, Chicago, IL) at a 2:1 (v/v) ratio, followed by centrifugation at 2000 × g for 20 min at 10°C without brakes. The interphase containing PBMC was collected into a 50-ml conical tube, prefilled with PBS (Thermo Fisher Scientific, Waltham, MA). The pellet was recovered by centrifugation at 500 × g for 15 min at room temperature and suspended in PBS. Cells were washed once more in PBS to remove platelets and recovered by centrifugation at 150 × g for 5 min at room temperature. All subsequent centrifugations were carried out at 150 × g for 5 min at 4°C or room temperature.

A total of 3 × 106 PBMC were washed in 1 ml ice-cold wash solution (130 mM NaCl and 5 mM KCl; pH 6). Cells were then resuspended in ice-cold acid wash solution (10 mM lactic acid [Alfa Aesar; Thermo Fisher Scientific, Lancashire, U.K.], 130 mM NaCl, and 5 mM KCl; pH 2.8–3) and incubated for 5 min on ice. After incubation, 0.2 ml of 1 M Tris HCl (pH 8; Thermo Fisher Scientific) was added to neutralize the acid before pelleting the cells by centrifugation. All wash steps were performed at 150 × g for 5 min at 4°C. Cells were washed once with 1 ml PBS and either used immediately for live cell sorting or fixed in 2% (v/v) paraformaldehyde solution (Sigma-Aldrich, St. Louis, MO) for 20 min at room temperature. An aliquot of PBMC without acid washing was fixed in 2% paraformaldehyde as a control.

For extracellular staining and analysis, PBMC were incubated for 15 min with different Ab master mixes (Supplemental Table I) in PBS-BSA (0.5% [w/v] BSA and 0.02% [w/v] NaN3; all from Sigma-Aldrich). mAbs against different horse cell surface markers or Ig isotypes were conjugated to Alexa Fluor 647 (AF647), Alexa Fluor 488 (AF488), or biotin according to the manufacturer’s protocols (Thermo Fisher Scientific). Master mixes 1–10 (Supplemental Table I) were used to phenotype the different CD23+ populations in equine PBMC. Master mixes 11–15 were used to measure intracellular IgE production. Master mix 16 was used for cell sorting of CD23hi/IgE+ cells and IgE B cells. Master mix 17 was used to measure IgE+ plasmablasts in allergic and healthy horses at different time points. Streptavidin-PE (Jackson ImmunoResearch Laboratories) was used to label biotinylated mAbs.

For intracellular staining and analysis, fixed PBMC were incubated first with master mix 11 (Supplemental Table I) in PBS-BSA to label cell surface IgE and CD23. Cells were subsequently incubated with streptavidin-PE to label biotinylated CD23. Cells were then washed once in saponin buffer (0.5% saponin, 0.5% BSA, and 0.02% NaN3 in PBS; all from Sigma-Aldrich) and incubated with master mix 12 in saponin buffer to label intracellular IgE. As a control, one aliquot of cells was incubated with master mix 13, in PBS-BSA instead of saponin buffer. Isotype controls were included in master mixes 14 and 15, which were subsequently stained on an additional PBMC aliquot to set the IgE+ gates.

Samples were recorded on a BD FACSCanto II flow cytometer, and data analysis was performed with FlowJo version 10.4 (Tree Star, Ashland, OR). A total of 100,000 events/sample were recorded for all master mixes except mixes 2 and 3, which recorded 50,000 events/sample. All flow cytometry images were gated first to exclude doublets and second to gate on PBMC by forward and side scatter characteristics. IgE+ cells were analyzed quantitatively.

Following PBMC isolation and removal of surface-bound IgE, 1 × 107 cells were incubated with mAb master mix 16 (Supplemental Table I), including a viability marker (Thermo Fisher Scientific), to isolate IgE+ plasmablasts by FACS. We used IgE mAb 176 for cell sorting due to its inability to induce crosslinking of receptor-bound IgE, as described previously (2529, 31). All cell sorting was performed at 4°C to minimize activation of B cells. Cells were sorted on a BD FACSAria Fusion Sorter at the Cornell Institute of Biotechnology’s flow cytometry core facility. Sorting was performed through a 100-µm nozzle at 20 pounds per square inch in a sterile hood. Compensation was calculated with input from single-stained UltraComp beads and amine-reactive beads (Thermo Fisher Scientific). Live CD23hi/IgE+ cells were collected into cell culture medium (DMEM supplemented with 1% [v/v] nonessential amino acids, 2 mM l-glutamine, 50 µM 2-ME, 50 µg/ml gentamicin, 100 U/ml penicillin, and 100 µg/ml streptomycin; Thermo Fisher Scientific) and 10% FCS (Atlanta Biologicals, Flowery Branch, GA) at 4°C. Live CD23lo/IgE cells were simultaneously sorted and collected for comparison. Sorted fractions were counted and tested for viability using trypan blue exclusion. All samples were >88% viable (average 93% live cells). To confirm morphology of sorted fractions, cytospin smears were prepared in a cytocentrifuge (Thermo Shandon Cytospin 4; Thermo Fisher Scientific) at 176 × g for 4 min at room temperature, and then stained with Wright’s stain. Morphology descriptions were performed by a single observer who was blinded to the origin of the cells and the sorting procedure.

Sorted CD23hi/IgE+ cells and IgE B cells (1 × 104 cells each) from three allergic horses in October were frozen in 3:1 TRIzol LS cell culture medium (Thermo Fisher Scientific) at −80°C until processed for RNA extraction. RNA was extracted in the aqueous layer with chloroform (Thermo Fisher Scientific) in Phase Lock Gel Heavy tubes (VWR International, Radnor, PA). RNA pellets were coprecipitated with GlycoBlue (Ambion Incorporated, Austin, TX) and washed once with isopropanol and three times with ethanol. Pellets were air dried and resuspended in 20 µl nuclease-free water (Growcells, Irvine, CA). RNA concentration and quality were measured via nanodrop. Equal amounts of extracted RNA were converted to cDNA using SuperScript VILO Reverse Transcriptase and No–Reverse Transcriptase control Master Mixes (Thermo Fisher Scientific).

Primers specific or the equine IgE H-chain C region (IGHE) were synthesized by Integrated DNA Technologies (Coralville, IA) as previously described (43). The transcript was amplified from 0.8 µl cDNA from FACS-sorted CD23hi/IgE+ cells, with 500 nM primers (Table II) (forward primer: 5′-GTCTCCAAGCAAGCCCCATTA-3′; reverse primer: 5′-TCGCAAGCTTTACCAGGGTCTTTGGACACCTC-3′) and Platinum SuperFi PCR Master Mix (Invitrogen, Carlsbad, CA). The PCR reaction was run with an initial denaturing of 94°C for 10 s, then 30 cycles of 94°C for 10 s, 57°C for 10 s, and 72°C for 30 s, followed by a final elongation step at 72°C 5 min and cooling to 4°C. PCR products were extracted from a 1% agarose gel (Invitrogen), cloned into a pCR4-TOPO plasmid (Invitrogen), and sequenced by Illumina sequencing.

Table II.

Primer pairs to differentiate memory B cells, plasmablasts, and plasma cells in IgE+ peripheral blood cells

Gene TypeGene NameLineage StageaReferenceAccession NumberForward and Reverse Primerb
MBCPBPC
Reference gene ACTB Tomlinson et al. (44), Bogaert et al. (70XM_023655002 F: CCAGCACGATGAAGATCAAG, R: GTGGACAATGAGGCCAGAAT 
 B2M  NM_001082502.3 F: TCTTTCAGCAAGGACTGGTCTTT, R: CATCCACACCATTGGGAGTAAA 
TFs BCL6 − − Nutt et al. (41), Jourdan et al. (42), Holmes et al. (71XM_003363354.4 F: ATTCCAGCTGTGAGAATGGG, R: CGGTCACACTTGTAGGGTTT 
 PAX5 − −  JQ044379.1 F: ACGCGTGTGTGACAATGA, R: CACTATGCTGTGGCTGGAA 
 BLIMP-1 − ++  XM_001501824.5 F: AGGAGCTTCTTGTGTGGTATTG, R: CTTAGGATGGCTCTGTGTTTGT 
 IRF4 − ++ ++  XM_023624331.1 F: CCCATAGAGCCAAGCATAAGG, R: TTGACGTGGTCAGCTCTTTC 
 XBP1 − ++ ++  XM_014742035 F: CTGGAACAGCAAGTGGTAGAT, R: CAGGCGCTGTCTTAACTCTT 
Cell surface proteins CD19 lo/− − Jacobi et al. (56), Qian et al. (72), Sanz et al. (57NM_001267799.2 F: AGTCCCTGCTGCAACTTTAG, R: GGGATCAGTCATTCGCTTTCT 
 CD138c − −/+ ++ Horst et al. (10), Qian et al. (72), Sanz et al. (57), Jourdan et al. (73XM_023619466.1 F: TCCTGGACAGGAAGGAAGT, R: GAGTAGCTGCCCTCATCTTTC 
 TACId − − Tellier and Nutt (58), Chihara et al. (45XM_005598001.2 F: CCATCATCTGCTGCTTCCT, R: GCTTCCATCCAGTGATCCTT 
Ig IGHE +e Wagner et al. (43AJ305047.1 F: GTCTCCAAGCAAGCCCCATTA, R: TCGCAAGCTTTACCAGGGTCTTTGGACACCTC 
Gene TypeGene NameLineage StageaReferenceAccession NumberForward and Reverse Primerb
MBCPBPC
Reference gene ACTB Tomlinson et al. (44), Bogaert et al. (70XM_023655002 F: CCAGCACGATGAAGATCAAG, R: GTGGACAATGAGGCCAGAAT 
 B2M  NM_001082502.3 F: TCTTTCAGCAAGGACTGGTCTTT, R: CATCCACACCATTGGGAGTAAA 
TFs BCL6 − − Nutt et al. (41), Jourdan et al. (42), Holmes et al. (71XM_003363354.4 F: ATTCCAGCTGTGAGAATGGG, R: CGGTCACACTTGTAGGGTTT 
 PAX5 − −  JQ044379.1 F: ACGCGTGTGTGACAATGA, R: CACTATGCTGTGGCTGGAA 
 BLIMP-1 − ++  XM_001501824.5 F: AGGAGCTTCTTGTGTGGTATTG, R: CTTAGGATGGCTCTGTGTTTGT 
 IRF4 − ++ ++  XM_023624331.1 F: CCCATAGAGCCAAGCATAAGG, R: TTGACGTGGTCAGCTCTTTC 
 XBP1 − ++ ++  XM_014742035 F: CTGGAACAGCAAGTGGTAGAT, R: CAGGCGCTGTCTTAACTCTT 
Cell surface proteins CD19 lo/− − Jacobi et al. (56), Qian et al. (72), Sanz et al. (57NM_001267799.2 F: AGTCCCTGCTGCAACTTTAG, R: GGGATCAGTCATTCGCTTTCT 
 CD138c − −/+ ++ Horst et al. (10), Qian et al. (72), Sanz et al. (57), Jourdan et al. (73XM_023619466.1 F: TCCTGGACAGGAAGGAAGT, R: GAGTAGCTGCCCTCATCTTTC 
 TACId − − Tellier and Nutt (58), Chihara et al. (45XM_005598001.2 F: CCATCATCTGCTGCTTCCT, R: GCTTCCATCCAGTGATCCTT 
Ig IGHE +e Wagner et al. (43AJ305047.1 F: GTCTCCAAGCAAGCCCCATTA, R: TCGCAAGCTTTACCAGGGTCTTTGGACACCTC 
a

B cell differentiation lineage stages.

b

Primers presented in 5′ to 3′ direction.

c

Also known as SDC1.

d

Also known as TNFRSF13B.

e

Expressed in IgE+ MBC, PB, and PC.

F, forward; MBC, memory B cell; PB, plasmablast; PC, plasma cell; R, reverse primer; TF, transcription factor.

Additional gene-specific primers (Table II) were designed to span intron-exon boundaries with Integrated DNA Technologies’ PrimerQuest tool and were synthesized by Integrated DNA Technologies. Primers were also aligned with the equine genome and did not have any predicted off-target amplicons of similar product size. Therefore, to confirm primer specificity, PCR product size was confirmed on a 2% agarose gel (Invitrogen) with a 10-bp ladder (Thermo Fisher Scientific). Primers for β2-microglobulin (B2M) were synthesized by Eurofins MWG Operon (Huntsville, AL) as previously described (44). B2M and β-actin (ACTB) were used as reference genes.

Quantitative real-time PCR (qPCR) was performed with 10-µl reactions each containing a cDNA equivalent obtained from a reverse transcriptase reaction of 3.4 ng mRNA, 500-nM forward and reverse primers, and SsoAdvanced Universal SYBR Green Master Mix (Bio-Rad Laboratories, Raleigh, NC). No–reverse transcriptase controls were run simultaneously for all samples and primer pairs. The qPCR reaction was run on a QuantStudio 5 thermocycler (Thermo Fisher Scientific) with an initial dissociation at 95°C for 30 s, then 40 cycles of 95C° for 15 s and 60°C for 15 s, followed by melt curve analysis. For qPCR results, Δ threshold cycle (dCt) values were calculated for all samples where dCt = Ct[target gene] − Ct[reference gene]. dCt values were calculated using both reference genes, ACTB or B2M. In general, the smaller the dCt, the greater the amount of mRNA in the sample. The mean difference in gene expression (ddCt) was calculated in which ddCt is the average of dCt[IgE B cells] − dCt[CD23hi/IgE+ cells] from each individual. Positive ddCt values represent increased gene expression in CD23hi/IgE+ cells compared with IgE B cells.

Isolated PBMC (6 × 105 cells/well, containing an average calculated 5.2 × 102–3.5 × 103 IgE+ plasmablasts), sorted CD23hi/IgE+ (IgE+ plasmablasts), and CD23lo/IgE (IgE B cells) cells (3 × 103 sorted cells/well) were incubated without stimulation in cell culture medium. Cells were incubated in 96-well flat-bottom plates (Corning, Corning, NY) at 37°C, 5% CO2, for 72 h before cell-free supernatants were collected and stored at 4°C until analysis. Secreted equine IgE was measured in undiluted supernatants using a fluorescent bead-based Luminex assay. The assay was set up as previously described with a standard curve of six 5-fold dilutions (3.2 ng/ml–10 µg/ml IgE) (26). Total IgE in serum was measured with the same assay design but instead used a standard curve of eight 2-fold dilutions (78 ng/ml–10 µg/ml) to measure the higher concentration of IgE in serum.

D’Agostino and Pearson tests were performed on all data sets and confirmed that the data were not normally distributed. Thus, nonparametric tests were used for data analysis. To compare IgE+ plasmablast frequencies between allergic and healthy horses at different time points, a Holm–Sidak multiple-comparisons test was used. A nonparametric Spearman rank correlation was calculated for all horses to compare each individual’s clinical allergy score or concentration of secreted IgE to the peripheral IgE+ plasmablast percentage. A Mann–Whitney U test was used to compare secreted IgE concentrations between allergic and healthy groups. A paired t test was used to compare dCt values between sorted CD23hi/IgE+ cells and IgE B cells. Gene expression graphs plot individual and average dCt values for each gene and cell fraction. A paired t test was also used to compare IgE secretion between IgE B cells and IgE+ plasmablast FACS-sorted samples. qPCR gene expression and IgE secretion were considered normally distributed due to the comparison of two cell types within each individual. All graphs plot median and range unless specified otherwise, and p values <0.05 were considered significant. Analysis was performed with Prism software version 8 (GraphPad Software, La Jolla, CA).

PBMC were stained with different mAbs and analyzed by flow cytometry to further characterize the CD23+ cell population in peripheral blood. CD23+ cells lacked T cell (CD4 and CD8) and monocyte (CD14) surface proteins but expressed high levels of MHC II (Fig. 1A–E). In allergic horses with clinical allergy (Table I), two distinct subpopulations could be separated with low and high CD23 expression (Fig. 1F). Although the frequency of CD23lo and CD23hi subpopulations varied at different time points, the distribution of Ig isotypes expressed in each CD23 subpopulation was similar. CD23lo cells were comprised of 7.0% IgM+/IgG1+ cells, 36.7% IgM+ cells, 27.0% IgG1+ cells, and 29.4% IgM/IgG1 cells (Fig. 1G, 1I, average of 4 individuals). In contrast, CD23hi cells were comprised of 1.4% IgM+/IgG1+ cells, 6.7% IgM+ cells, 10.0% IgG1+ cells, and 81.9% IgM/IgG1 (Fig. 1H, 1I, average of 4 individuals). Circulating CD23+ cells were also IgE+ (Fig. 2A). We then developed an approach to further characterize the CD23hi/IgM/IgG1 cell population.

FIGURE 1.

Equine CD23+ cells include a CD23hi/IgM/IgG1 population. PBMC were stained for CD23 and different cell surface markers. All cells were first gated for doublet exclusion (A) and cells with PBMC morphology (B). (C) CD23 expression on PBMC-gated cells. Expression of different cell proteins was measured on CD23+ gated cells: CD4 and CD8 (D) and MHC II and CD14 (E). (F) CD23+ cells can be separated into a CD23lo and CD23hi population. IgM and IgG1 expression on CD23lo cells (G) and CD23hi cells (H). (I) The distribution of IgM+/IgG1+ (gray striped bars), IgM+ (gray bars), IgG1+ (white bars), and IgG1/IgM (black bars) expression within the CD23lo and CD23hi cell fractions. Graph shows the average distribution from four allergic individuals. FACS images are representative of one of eight individuals.

FIGURE 1.

Equine CD23+ cells include a CD23hi/IgM/IgG1 population. PBMC were stained for CD23 and different cell surface markers. All cells were first gated for doublet exclusion (A) and cells with PBMC morphology (B). (C) CD23 expression on PBMC-gated cells. Expression of different cell proteins was measured on CD23+ gated cells: CD4 and CD8 (D) and MHC II and CD14 (E). (F) CD23+ cells can be separated into a CD23lo and CD23hi population. IgM and IgG1 expression on CD23lo cells (G) and CD23hi cells (H). (I) The distribution of IgM+/IgG1+ (gray striped bars), IgM+ (gray bars), IgG1+ (white bars), and IgG1/IgM (black bars) expression within the CD23lo and CD23hi cell fractions. Graph shows the average distribution from four allergic individuals. FACS images are representative of one of eight individuals.

Close modal
FIGURE 2.

Removal of CD23-bound IgE reveals a population that expresses IgE and no other isotypes. PBMC were treated with a lactic acid wash solution to remove surface CD23 receptor-bound IgE. Afterwards, cells were stained for CD23 and the presence of cell surface Ig as part of their B cell receptors and measured by flow cytometry. (A) CD23 and IgE staining on ex vivo PBMC prior to the acid wash. Gates distinguish CD23hi and CD23lo cells. (B) CD23 and IgE expression on PBMC after treatment with a lactic acid wash solution to remove CD23 receptor-bound IgE. Gates distinguish CD23hi and CD23lo cells. Gated cells from (B) are analyzed in (C) and (D). (C) Expression of IgE and IgM, IgD, or IgG1 on CD23lo cells. (D) Expression of IgE and IgM, IgD, IgG1, IgG3/5, IgG4/7, or IgG6 on CD23hi cells. (E) CD23 expression on acid-washed PBMC. (F) Mouse IgG1 isotype controls on CD23+ gated cells. FACS images are representative of one of four individuals.

FIGURE 2.

Removal of CD23-bound IgE reveals a population that expresses IgE and no other isotypes. PBMC were treated with a lactic acid wash solution to remove surface CD23 receptor-bound IgE. Afterwards, cells were stained for CD23 and the presence of cell surface Ig as part of their B cell receptors and measured by flow cytometry. (A) CD23 and IgE staining on ex vivo PBMC prior to the acid wash. Gates distinguish CD23hi and CD23lo cells. (B) CD23 and IgE expression on PBMC after treatment with a lactic acid wash solution to remove CD23 receptor-bound IgE. Gates distinguish CD23hi and CD23lo cells. Gated cells from (B) are analyzed in (C) and (D). (C) Expression of IgE and IgM, IgD, or IgG1 on CD23lo cells. (D) Expression of IgE and IgM, IgD, IgG1, IgG3/5, IgG4/7, or IgG6 on CD23hi cells. (E) CD23 expression on acid-washed PBMC. (F) Mouse IgG1 isotype controls on CD23+ gated cells. FACS images are representative of one of four individuals.

Close modal

We assumed that many of the CD23+ cells had bound IgE from circulation to their cell surface via the low-affinity IgE receptor CD23. Most of the IgE could be removed from CD23+ cells by a short lactic acid wash, leaving two populations, CD23lo/IgE and CD23hi/IgE+ (Fig. 2B) cells. In addition, the lactic acid wash removed IgE from FcεRI+ cells, such as CD23/CD14+ monocytes and CD23/CD14 basophils (Supplemental Fig. 1A–D). After the acid wash, CD23lo cells lost all surface IgE, but retained IgM, IgD, or IgG1 (Fig. 2C) on the cell surface. This suggested that these cells were B cells expressing an IgM, IgD, or IgG1 BCR. Hereafter, we refer to the CD23lo population as IgE B cells. In contrast, CD23hi cells were still IgE+ after acid wash. The majority of CD23hi cells did not express any other isotype and were IgM (99.7%), IgD (98.0%), IgG1 (95.1%), IgG3/5 (98.3%), IgG4/7 (98.9%), and IgG6 (99.0%) (Fig. 2D, average of 4 allergic individuals). This demonstrated that these cells do not express any other isotype except IgE. Acid-washed cells were also stained with mouse IgG1 isotype controls (Fig. 2I, 2J). Together, these data support that CD23hi/IgE+ cells are IgE-expressing B cells with an IgE BCR on their cell surface.

To further confirm if CD23hi/IgE+ cells produce IgE, CD23-bound IgE was again removed from PBMC by acid wash (Fig. 3A). Acid-washed cells did not bind isotype controls extracellularly or intracellularly (Fig. 3B). The cells were stained with subsequent Ab master mixes to label them first for cell surface IgE (IgE mAb 176 coupled to AF647), followed by intracellular IgE staining (IgE mAb 176 coupled to AF488). The majority of CD23hi/IgE+ cells (92.9%; average of 4 allergic horses) expressed both cell surface (Fig. 3C, 3E) and intracellular IgE (Fig. 3D, 3E). A small fraction (6.8%; average of 4 allergic horses) of CD23hi/IgE+ cells expressed surface IgE only and no intracellular IgE. Less than 0.1% of CD23hi/IgE+ cells expressed intracellular IgE with no surface IgE.

FIGURE 3.

CD23hi/IgE+ cells produce intracellular IgE. PBMC were first treated with acid wash solution to remove CD23-bound IgE. Cells were fixed and stained for surface IgE expression, followed by intracellular IgE staining and flow cytometric analysis. (A) CD23 and cell surface IgE expression on PBMC after acid wash. Gates show CD23hi/IgE+ cells and IgE B cells (BC). (B) Mouse IgG1 isotype controls did not bind acid-washed cells during extracellular incubation (AF647) or intracellular incubation (AF488). (C)–(H) show CD23hi/IgE+ gated cells. Surface labeling with IgE mAb-AF647 (C) was followed by intracellular labeling with IgE mAb-AF488 (D), and expression of both surface and intracellular IgE (E) is shown. Surface labeling with IgE mAb-AF647 (F) was followed by a second surface labeling with IgE mAb-AF488 (G), and labeling of both surface antibodies (H) is shown. (I)–(K) show gated IgE BC. Surface labeling with IgE mAb-AF647 (I) was followed by intracellular labeling with IgE mAb-AF488 (J), and expression of both surface and intracellular IgE (K) is shown. FACS images are representative of one of four horses.

FIGURE 3.

CD23hi/IgE+ cells produce intracellular IgE. PBMC were first treated with acid wash solution to remove CD23-bound IgE. Cells were fixed and stained for surface IgE expression, followed by intracellular IgE staining and flow cytometric analysis. (A) CD23 and cell surface IgE expression on PBMC after acid wash. Gates show CD23hi/IgE+ cells and IgE B cells (BC). (B) Mouse IgG1 isotype controls did not bind acid-washed cells during extracellular incubation (AF647) or intracellular incubation (AF488). (C)–(H) show CD23hi/IgE+ gated cells. Surface labeling with IgE mAb-AF647 (C) was followed by intracellular labeling with IgE mAb-AF488 (D), and expression of both surface and intracellular IgE (E) is shown. Surface labeling with IgE mAb-AF647 (F) was followed by a second surface labeling with IgE mAb-AF488 (G), and labeling of both surface antibodies (H) is shown. (I)–(K) show gated IgE BC. Surface labeling with IgE mAb-AF647 (I) was followed by intracellular labeling with IgE mAb-AF488 (J), and expression of both surface and intracellular IgE (K) is shown. FACS images are representative of one of four horses.

Close modal

The surface labeling of IgE saturated all surface IgE mAb binding sites, and therefore, intracellular staining revealed only intracellular IgE. This was confirmed by two sequential extracellular IgE staining steps. The majority of the CD23hi/IgE+ cells were only positive for the first IgE label (Fig. 3F, 3H), and the second labeling step did not label any additional surface IgE (Fig. 3G, 3H). Finally, IgE B cells were negative for all surface (Fig. 3I, 3K) and intracellular IgE labeling (Fig. 3J, 3K). Together, these data further confirm that CD23hi/IgE+ cells have undergone class switching to IgE and consequently express IgE intracellularly and as part of their cell surface BCR.

Before becoming plasma cells, B cells differentiate into plasmablasts, which secrete Ab and travel from the lymph node through peripheral blood (2, 3). To find additional evidence for the differentiation stage of peripheral CD23hi/IgE+ cells as Ab-secreting cells, we used the RNA extracted from FACS-sorted CD23hi/IgE+ cells and RNA extracted from IgE B cells sorted simultaneously as controls (Supplemental Fig. 2). Gene expression for B cell, plasmablast, and plasma cell differentiation was determined with gene-specific primers (Table II) by quantitative RT-PCR. Each sample was normalized to ACTB (Fig. 4A) or B2M (Supplemental Fig. 3A), and normalized gene expression (dCt) was compared between CD23hi/IgE+ cells and IgE B cells. Normalization with each reference gene resulted in similar trends in gene expression.

FIGURE 4.

CD23hi/IgE+ cells have upregulated plasmablast differentiation marker transcripts and exhibit plasmablast morphology. Gene expression was compared in IgE B cells (BC; □; n = 3) and CD23hi/IgE+ cells (▪; n = 3). RNA was extracted, equal RNA concentrations were converted into cDNA, and cDNA was amplified by quantitative RT-PCR with gene-specific primers. Data were normalized to ACTB (dCt). (A) Individual dCt values and means are graphed for the transcription factors BCL6, PAX5, BLIMP-1, IRF4, and XBP1 and surface proteins CD138, TACI, and CD19. Graph plots individual values with medians. (B) Gel image showing PCR products from IGHE primers run on CD23hi/IgE+ sorted cells. Lane 1 shows annotated ladder, and lanes 2–4 show the PCR product from three different allergic individuals. Sorted IgE BC (C) and sorted CD23hi/IgE+ cells (D) were prepared by cytospin. Perinuclear clear zone (black arrow) and cytoplasmic clear vacuoles (gray arrow) are labeled in two CD23hi/IgE+ cells. Cells were differentially stained and imaged under ×50 magnification. *p < 0.05, **p < 0.01.

FIGURE 4.

CD23hi/IgE+ cells have upregulated plasmablast differentiation marker transcripts and exhibit plasmablast morphology. Gene expression was compared in IgE B cells (BC; □; n = 3) and CD23hi/IgE+ cells (▪; n = 3). RNA was extracted, equal RNA concentrations were converted into cDNA, and cDNA was amplified by quantitative RT-PCR with gene-specific primers. Data were normalized to ACTB (dCt). (A) Individual dCt values and means are graphed for the transcription factors BCL6, PAX5, BLIMP-1, IRF4, and XBP1 and surface proteins CD138, TACI, and CD19. Graph plots individual values with medians. (B) Gel image showing PCR products from IGHE primers run on CD23hi/IgE+ sorted cells. Lane 1 shows annotated ladder, and lanes 2–4 show the PCR product from three different allergic individuals. Sorted IgE BC (C) and sorted CD23hi/IgE+ cells (D) were prepared by cytospin. Perinuclear clear zone (black arrow) and cytoplasmic clear vacuoles (gray arrow) are labeled in two CD23hi/IgE+ cells. Cells were differentially stained and imaged under ×50 magnification. *p < 0.05, **p < 0.01.

Close modal

Plasmablasts and plasma cells downregulate transcription factor paired box protein 5 (PAX5) and B cell lymphoma 6 (BCL6) and upregulate BLIMP-1, IRF4, and XBP1 (Table II). We detected an expression pattern similar to this while comparing the ddCt in CD23hi/IgE+ cells and IgE B cells. Positive ddCt values represent increased expression in CD23hi/IgE+ cells and vice versa. CD23hi/IgE+ cells had decreased expression of PAX5 (ddCt = −3.67; p = 0.144) and BCL6 (ddCt = −4.63; p = 0.118) compared with IgE B cells (Fig. 4A). In contrast, CD23hi/IgE+ cells had increased expression of BLIMP-1 (ddCt = 2.833; p = 0.322), IRF4 (ddCt = 3.417; p = 0.002), and XBP1 (ddCt = 4.877; p = 0.012) compared with IgE B cells (Fig. 4A). CD23hi/IgE+ cells also had increased expression of transcripts for surface proteins CD138 (ddCt = 4.177; p = 0.005) and transactivator and CAML interactor (TACI; ddCt = 2.117; p = 2.117), compared with IgE B cells (Fig. 4A). CD138 and TACI both begin to be expressed as B cells differentiate to plasmablasts and plasma cells. CD23hi/IgE+ cells had similar expression of B cell marker CD19 compared with IgE B cells (ddCt = −0.907; p = 0.057) (Fig. 4A). Overall, these gene expression profile trends further support peripheral CD23hi/IgE+ cells as plasmablasts compared with IgE B cells.

To verify that CD23hi/IgE+ cells had undergone class switching and were truly expressing IgE, expression of the IGHE RNA transcript was measured by PCR using IGHE-specific primers (Table II). The PCR product from CD23hi/IgE+ cell cDNA was 1272 bp (Fig. 4B) and was sequenced and confirmed to be the expressed IGHE cDNA sequence with 99% homology to the annotated sequence (Table II, Supplemental Fig. 3B). We therefore refer to this sequence as the IGHEc haplotype. This provides yet more proof that these cells have undergone class switching and express IgE.

In addition, morphology of CD23hi/IgE+ cells and IgE B cells were explored by differential staining of sorted cell populations. Consistent with B cell morphology, IgE B cells had characteristic round, central nuclei with coarsely clumped chromatin, a narrow cytoplasmic area, and small cell size (<10-µm diameter) (Fig. 4C, Supplemental Fig. 3C). In contrast, CD23hi/IgE+ cells were much larger (>10-µm diameter) (Fig. 4D, Supplemental Fig. 3D) with central nuclei and coarse chromatin. The CD23hi/IgE+ cells also had perinuclear clear zone (Fig. 4D, black arrow), a larger and deeply basophilic cytoplasm, and occasional small clear vacuoles (Fig. 4D, gray arrow). This is indicative of an expanded Golgi apparatus, higher cytoplasmic protein content, and secretory vacuoles, respectively, and consistent with Ab-secreting cells (45, 46). In summary, CD23hi/IgE+ cells show typical gene expression and morphology characteristics of peripheral IgE+ plasmablasts. Hereafter, we refer to CD23hi/IgE+ cells as IgE+ plasmablasts.

To determine whether IgE+ plasmablasts were elevated during clinical allergy, we compared percentages of IgE+ plasmablasts in the peripheral blood of healthy horses (n = 6) with those of horses allergic with Culicoides hypersensitivity (n = 7). Culicoides hypersensitivity is a seasonal allergic disease, and therefore, allergic horses only develop clinical signs in the summer during Culicoides allergen exposure (Fig. 5A, Table I). CD23hi B cells were first measured in PBMC ex vivo, which still had CD23-bound IgE. As shown above, CD23hi/IgM/IgG1 cells include, but do not entirely distinguish, IgE+ plasmablasts. We compared the frequency of CD23hi/IgM/IgG1 cells between healthy and allergic horse groups for 1 y and determined that during the chronic phase of allergy, in late summer and fall, allergic horses had significantly increased percentages of CD23hi/IgM/IgG1 B cells compared with healthy controls (Fig. 5B).

FIGURE 5.

Peripheral IgE+ plasmablast percentages correlate with severity of clinical allergy. Percentages of peripheral blood IgE+ plasmablasts were compared in allergic (●, black bars; n = 7) and nonallergic (○, white bars; n = 6) horses by flow cytometry. (A) Allergy scores of both horse groups during year of study. (B) CD23hi/IgM/IgG1 cells were analyzed over 1 y. Percentages of CD23hi/IgM/IgG1 cells out of total CD23+ cells are shown. (C) Percentages of IgE+ plasmablasts out of total CD23+ cells in July and December. PBMC were treated with acid wash solution before antibody staining to remove CD23-bound IgE. (D) Spearman rank correlation of the percentage of IgE+ plasmablasts in peripheral blood and allergy scores for all horses in July, during the peak of allergen exposure and clinical allergy. (A and B) The shaded boxes represent months when Culicoides were present in the environment. (A and D) The dotted line represents the threshold where horses with scores ≥3 have clinical allergy. (A)–(C) Asterisks denote months when IgE+ plasmablast frequency was significantly increased in allergic horses compared with nonallergic horses. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 5.

Peripheral IgE+ plasmablast percentages correlate with severity of clinical allergy. Percentages of peripheral blood IgE+ plasmablasts were compared in allergic (●, black bars; n = 7) and nonallergic (○, white bars; n = 6) horses by flow cytometry. (A) Allergy scores of both horse groups during year of study. (B) CD23hi/IgM/IgG1 cells were analyzed over 1 y. Percentages of CD23hi/IgM/IgG1 cells out of total CD23+ cells are shown. (C) Percentages of IgE+ plasmablasts out of total CD23+ cells in July and December. PBMC were treated with acid wash solution before antibody staining to remove CD23-bound IgE. (D) Spearman rank correlation of the percentage of IgE+ plasmablasts in peripheral blood and allergy scores for all horses in July, during the peak of allergen exposure and clinical allergy. (A and B) The shaded boxes represent months when Culicoides were present in the environment. (A and D) The dotted line represents the threshold where horses with scores ≥3 have clinical allergy. (A)–(C) Asterisks denote months when IgE+ plasmablast frequency was significantly increased in allergic horses compared with nonallergic horses. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

However, acid wash treatment of PBMC to remove CD23-bound IgE allows more accurate gating and identification of IgE+ plasmablasts. IgE+ plasmablasts were compared in acid-washed PBMC. Allergic horses had significantly increased frequencies of IgE+ plasmablasts compared with healthy horses in July, but not in December, when all horses appeared clinically healthy again (Fig. 5C). In fact, IgE+ plasmablasts comprised ∼50% (median 52.4%, range 26.1–68.9%) of total CD23+ cells in allergic horses in the summer. This was in clear contrast to healthy horses, in which only ∼10% of CD23+ cells were IgE+ plasmablasts (median 11.9%, range 7.6–18.2%). Clinical scores and IgE+ plasmablast percentages were highly correlated in July when Culicoides allergen exposure and clinical allergy peaked in allergic horses (rsp = 0.7828; p < 0.01) (Fig. 5D), demonstrating that IgE+ plasmablasts are indicative of allergy severity in Culicoides hypersensitivity.

To determine if IgE+ plasmablasts are secreting IgE, isolated PBMC from allergic (n = 7) and nonallergic horses (n = 7) were cultured for 3 d in cell culture medium without additional stimuli. Afterwards, secreted IgE Abs were measured in the supernatants. Although PBMC from all horses secreted IgE, PBMC from allergic horses secreted significantly more IgE (median 486.2 ng/ml, range 270.8–822.0 ng/ml) than healthy horses (median 149.8 ng/ml, range 94.2–558.5 ng/ml) (Fig. 6A). PBMC were also treated with the acid wash solution to remove all CD23-bound IgE and subsequently cultured for 3 d. The concentrations of secreted IgE in the supernatant of these acid-washed cells (Fig. 6B) (median 544.8 ng/ml, range 444.7–1428.2 ng/ml in allergic horses) were similar compared with ex vivo incubated PBMC, and allergic horses had significantly higher secreted IgE concentrations compared with healthy controls. This difference in IgE secretion was not reflected in serum. Total IgE was measured in serum samples and was similar between allergic (median 114.7 µg/ml, range 22.2–343.0 µg/ml) and healthy horses (median 92.0 µg/ml, range 38.4–203.9 µg/ml) (Fig. 6C).

FIGURE 6.

IgE+ plasmablasts readily secrete IgE proportional to cell percentage. Functional ability to secrete IgE was compared between healthy (○; n = 7) and allergic (●; n = 7) horses. PBMC were incubated for 72 h in cell culture medium, and total IgE was measured in the supernatant from allergic and healthy groups. Both ex vivo PBMC (A) and acid-washed PBMC (B) were incubated in parallel and IgE secretion was compared. (C) Total IgE in serum. (D) The frequency of IgE+ plasmablasts out of total PBMC, measured by flow cytometry after treatment with acid wash solution before incubation. (E) Spearman rank correlation of the percentage of peripheral IgE+ plasmablasts and the concentration of total IgE in supernatant of ex vivo PBMC after 72-h incubation in cell culture medium. Line shows a simple linear regression. (F) Concentration of total IgE in the supernatant of sorted IgE B cells (BC; ▵) and IgE+ plasmablasts (PB; ▴) from allergic horses (n = 3) after 72-h incubation in cell culture medium. Graphs plot median and individual values. *p < 0.05, **p < 0.01.

FIGURE 6.

IgE+ plasmablasts readily secrete IgE proportional to cell percentage. Functional ability to secrete IgE was compared between healthy (○; n = 7) and allergic (●; n = 7) horses. PBMC were incubated for 72 h in cell culture medium, and total IgE was measured in the supernatant from allergic and healthy groups. Both ex vivo PBMC (A) and acid-washed PBMC (B) were incubated in parallel and IgE secretion was compared. (C) Total IgE in serum. (D) The frequency of IgE+ plasmablasts out of total PBMC, measured by flow cytometry after treatment with acid wash solution before incubation. (E) Spearman rank correlation of the percentage of peripheral IgE+ plasmablasts and the concentration of total IgE in supernatant of ex vivo PBMC after 72-h incubation in cell culture medium. Line shows a simple linear regression. (F) Concentration of total IgE in the supernatant of sorted IgE B cells (BC; ▵) and IgE+ plasmablasts (PB; ▴) from allergic horses (n = 3) after 72-h incubation in cell culture medium. Graphs plot median and individual values. *p < 0.05, **p < 0.01.

Close modal

An aliquot of ex vivo PBMC from each horse was treated with the acid wash solution and analyzed by flow cytometry to compare the percentage of IgE+ plasmablasts between groups (Fig. 6D). The concentration of secreted IgE was proportional to the frequency IgE+ plasmablasts in total PBMC (Fig. 6E) (rsp = 0.8877; p < 0.0001). IgE+ plasmablasts from allergic horses were also FACS sorted (Supplemental Fig. 2) and incubated for 3 d without additional stimuli. Sorted IgE B cells were included as a control. IgE+ plasmablasts, but not IgE B cells, from allergic horses secreted IgE (Fig. 6F). In summary, these data indicate that IgE+ plasmablasts are actively secreting IgE and that IgE secretion is enhanced in allergic compared with healthy horses due to the increased numbers of IgE+ plasmablasts in allergic individuals.

Secreted IgE by IgE+ plasmablasts likely first binds to available IgE receptors and then contributes to soluble IgE in the cell culture supernatant. To test this, PBMC were acid washed (Supplemental Fig. 1E) and then incubated for 40 h. After incubation, all CD23+ cells rebound secreted IgE through surface CD23 (Supplemental Fig. 1F), demonstrating that CD23+ cells, including IgE+ plasmablasts, are functional after the short acid treatment and can bind secreted IgE.

IgE-secreting cells are critical mediators of type I hypersensitivities and the source of IgE. Although these cells have been identified in human peripheral blood (47, 10), bone marrow, and local tissues (9, 47), experimental methods to characterize and isolate IgE-secreting cells often require significant manipulation or analysis. In this study, we use an approach to identify and characterize IgE+ plasmablasts in peripheral blood using a brief lactic acid wash to remove cell surface receptor-bound IgE. This approach allows live-cell functional analysis of IgE-secreting cells. This approach also allowed the observation that the frequency of peripheral IgE+ plasmablasts positively correlates with both secreted IgE from these cells and with allergic disease severity in the individual.

IgE readily binds to its two primary surface receptors, FcεRI and CD23. In peripheral blood, FcεRI is expressed on basophils, dendritic cells, eosinophils, and monocytes (26, 29, 4850), whereas CD23 is expressed on the majority of activated B cells (12, 14). Acid wash of basophils has been frequently used to remove FcεRI-bound IgE and replace it with Ag-specific IgE for functional assays (51, 52). Acid wash of B cells has also been used to remove surface-bound Ig (53, 54). In the past, acetic acid was used to remove surface-bound IgE from murine cells (54) and was insufficient in specifically removing surface IgE to allow accurate labeling of all IgE+ B cells. In this study, we provide extensive evidence that using lactic acid allows the specific identification and characterization of IgE+ plasmablasts in a naturally occurring allergy model. Lactic acid has also recently been used to measure cell-bound allergen-specific IgE as a diagnostic marker for human allergy (55). Lactic acid solution preserves the degranulation ability of basophils following the acid wash (51, 52), and we therefore used a similar solution to maintain the functional ability of B cells and plasmablasts. Following acid wash, IgE+ plasmablasts were able to secrete IgE to a similar degree as ex vivo cells, supporting that this treatment does not compromise IgE secretion or IgE detection and is appropriate for functional studies of IgE+ B cells and IgE+ plasmablasts ex vivo.

After class switching, B cells undergo a series of differentiation steps to develop into Ab-secreting plasma cells. Typically, activated B cells differentiate into either memory B cells (MHC II+/CD19+/CD23+/CD138/TACI/PAX5+/BCL6+/BLIMP-1/XBP1/IRF4) or circulating plasmablasts (MHC II+/CD19lo/CD23lo/+/CD138+/−/TACI/PAX5/BCL6/BLIMP-1+/XBP1+/IRF4+) and ultimately become bone marrow–resident plasma cells (MHC II/CD19/CD23/CD138+/TACI+/PAX5/BCL6/BLIMP-1++/XBP1+/IRF4+). MHC II expression and low CD19 expression are retained on plasmablasts and can be used to distinguish them from plasma cells, which downregulate both surface proteins (5658). Recent single-cell RNA sequencing of equine PBMC identified a cluster of CD23+ IgE-secreting cells that were BLIMP-1+/XBP1+/IRF4+ (59), and single-cell RNA sequencing of human B cells determined that the majority of IgE+ cells are plasmablasts with high CD23 transcript expression (4). Single-cell sequencing of human IgE-producing cells also noted elevated CD23 transcripts in IgE+ plasmablasts (60). However, none of these studies validated the CD23hi/IgE+ plasmablast population by methods other than RNA sequencing.

The population of CD23hi/IgE+ plasmablasts identified in our study match these phenotypes and have significantly increased expression of IRF4, XBP1, and CD138, moderately increased expression of BLIMP-1 and TACI, and decreased expression of PAX5 and BCL6 compared with CD23+/IgE cells. Importantly, we also measured expression of the IGHE gene in CD23hi/IgE+ cells, which encodes for the Cε portion of IgE. CD23hi/IgE+ plasmablasts only express the spliced IGHE sequence, confirming that they have undergone class switching to express IgE. We therefore refer to these cells as IgE+ plasmablasts due to their expression of the spliced IGHE gene, ability to secrete IgE Ab, morphology, MHC II+/CD23+ surface expression, and elevated expression of BLIMP-1 IRF4, XBP1, CD138, and TACI mRNA. As described previously (45, 46), the combination of plasma cell gene transcripts simultaneously with MHC II and CD19 expression supports the identification of these cells as differentiating plasmablasts that have not yet matured into plasma cells. This is also consistent with the recent finding that the majority of peripheral IgE+ B cells are in fact plasmablasts (4).

Plasmablasts typically downregulate surface CD23 (61), but the IgE+ plasmablasts identified in this study have unusually high CD23 expression. There are numerous possible explanations for this. First, CD23hi/IgE+ plasmablasts could have very recently differentiated to plasmablasts and not yet downregulated surface CD23. Second, continuous IgE secretion by IgE+ plasmablasts could create a high concentration, soluble IgE microenvironment around the cells, which could drive CD23 upregulation (23). IgE plasmablasts secreting other Ig isotypes, such as IgG, would not have this IgE microenvironment and therefore would downregulate CD23. Finally, CD23 could be cleaved into secretory CD23 and bound in high concentrations to the surface of IgE+ plasmablasts through secretory CD23–IgE–CD21 interactions (14). This final possibility is unlikely, as we did not see any disruption of CD23 expression after acid wash treatment. We expect that the high CD23 expression on IgE+ plasmablasts is also present in allergic humans based on prior studies that have identified variable CD23 expression on human B cells, but this has yet to be confirmed.

CD23 on B cells can engage in two distinct signaling pathways. First, CD23 expression on B cells can engage in a process called FAP. During FAP, B cells present multiple Ags to T cells: Ags bound by the BCR and also Ags by CD23-bound IgE (62). Whether CD23hi/IgE+ plasmablasts are capable of enhanced FAP compared with other CD23+ B cells will be important to explore in the future. Second, CD23 expression and activation have been associated with decreased IgE production (11, 16). CD23 binding of preformed allergen–IgE immune complexes can also decrease IgE binding and allergen-specific activation of FcεRI+ cells, thereby decreasing the allergic response (14, 15). The factors that may bias CD23+ cells toward one or the other pathway are not yet understood. We propose that high CD23 expression acts to bias the cells toward one of these two pathways. In fact, increased cell surface density of CD23 has been associated with elevated serum IgE concentrations, allergy severity, and Ag presentation to T cells (23), suggesting that the network of CD23 on the cell surface, and its interaction with free IgE or immune complexes, may contribute to the downstream signaling pathways.

We have shown in this article that the majority of CD23hi cells are CD23hi/IgE+ plasmablasts. However, in some individuals, there are occasional IgE cells that also have high CD23 expression. Therefore, gating on CD23hi cells alone, without acid wash and measurement of IgE expression, is insufficient for truly identifying the IgE+ plasmablast population. This emphasizes the importance of using this approach to select true CD23hi/IgE+ plasmablasts.

We have also shown that IgE+ plasmablasts actively secrete IgE. After removing CD23-bound IgE, these plasmablasts can bind secreted IgE via their free CD23 receptors in an autocrine manner in vitro. If this also occurs in vivo, it is possible that binding of autocrine free IgE by CD23 would favor the inflammatory signaling pathway of CD23 and, as discussed above, would enhance FAP and T cell activation (15). During allergen exposure, we expect most of the secreted IgE to be allergen-specific, in this case to Culicoides midge salivary allergens, due to the correlation of IgE+ plasmablasts with the onset of clinical allergy. Therefore, CD23 on the plasmablasts could act to amplify allergen-specific IgE on the cell surface and increase the ability to bind and respond to allergen. Autocrine binding in this way could concentrate allergen-specific IgE on the cell surface and increase allergen binding, processing, and presentation. This further supports the vicious cycle of allergen-specific IgE+ plasmablast activation and IgE production during allergic responses. In contrast, CD23hi/IgE cells do not produce or secrete IgE and therefore would be more likely to bind a variety of Ag/IgE immune complexes. CD23 binding of immune complexes results in a noninflammatory signaling pathway, enhanced clearance of serum IgE, and decreased activation of FcεRI+ cells (15).

Allergic individuals typically experience temporary allergen exposure and a range of clinical severity. As a result, it can be challenging to identify the specific mechanisms of allergen exposure and disease development. We used the horse as a large animal model of naturally occurring seasonal allergy (24). Horses can naturally develop a type I IgE–mediated allergy, Culicoides hypersensitivity (3239, 6365), which is a recurrent, seasonal disease in response to salivary proteins from Culicoides midges. Allergic horses develop moderate to severe dermatitis, pruritis, and alopecia following Culicoides bites and allergen exposure (6669). Our herd of allergic and nonallergic horses all live together in the same environment and experience similar allergen exposure frequency and duration. Therefore, the horse allergy model allows for a direct comparison among allergen exposure, IgE+ plasmablast frequency, and clinical allergy severity.

Finally, our finding that frequency of IgE+ plasmablasts in blood positively correlates to both allergy severity and to the concentration of secreted IgE suggests that IgE+ plasmablasts may serve as a clinical biomarker of allergy. The presence of IgE+ plasmablasts in human PBMC, with high CD23 expression that directly correlates with allergy severity, has been suggested (46, 9) but has not yet been specifically identified. Future work is needed to confirm that our lactic acid wash and flow cytometry approach can label human IgE+ plasmablasts. We expect that monitoring the presence of horse or human IgE+ plasmablasts as a diagnostic biomarker will provide important insight into the development of allergy in an individual and will guide allergy treatment decisions. This will benefit both human and veterinary medicine.

In conclusion, we have developed an approach to easily and reliably identify an allergy-associated, IgE-secreting cell population called IgE+ plasmablasts. We have identified that IgE+ plasmablasts express the spliced IGHE gene, express high amounts of CD23 on their cell surface, have elevated transcripts for BLIMP-1, XBP1, IRF4, CD138, and TACI proteins, and retain MHC II expression. IgE+ plasmablasts in peripheral blood readily secrete IgE and are significantly elevated in allergic individuals compared with healthy counterparts. Finally, IgE+ plasmablast frequency in peripheral blood positively correlates to allergy severity, suggesting that these cells are mechanistically involved in, and may precede, the development of clinical allergy.

We thank Fahad Raza, Naya Eady, Camille Holmes, and Heather Freer for their help with horse handling, blood collection, and sample processing. We also thank the Cornell University CARE team for care of the horses and facility maintenance, the Flow Cytometry Facility of the Biotechnology Resource Center at Cornell Institute of Biotechnology for help with sorting experiments, and, finally, Sandy Sisson and Shelley Chu at the Cornell University Clinical Pathology Diagnostic Lab and John Beebe at the Cornell University Molecular Diagnostic laboratory for support with cytospin slide preparation, imaging, and qPCR, respectively.

This work was supported by the Harry M. Zweig Memorial Fund for Equine Research at Cornell University, the U.S. Department of Agriculture/National Institute of Food and Agriculture (2005-01812, 2015-67015-23072, and 2019-67015-29833), and the National Institutes of Health (1S10RR025502 through Cornell University’s Biotechnology Resource Center). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

The online version of this article contains supplemental material.

Abbreviations used in this article:

     
  • ACTB

    β-actin

  •  
  • AF488

    Alexa Fluor 488

  •  
  • AF647

    Alexa Fluor 647

  •  
  • BCL6

    B cell lymphoma 6

  •  
  • BLIMP-1

    B lymphocyte-induced maturation protein 1

  •  
  • B2M

    β2-microglobulin

  •  
  • dCt

    Δ threshold cycle

  •  
  • ddCt

    difference in Δ threshold cycle

  •  
  • FAP

    facilitated Ag presentation

  •  
  • IGHE

    IgE H-chain C region

  •  
  • IRF4

    IFN regulating factor 4

  •  
  • MHC II

    MHC class II

  •  
  • PAX5

    paired box protein 5

  •  
  • qPCR

    quantitative real-time PCR

  •  
  • TACI

    transactivator and CAML interactor

  •  
  • XBP1

    X-box binding protein 1

1.
Gould
H. J.
,
B. J.
Sutton
.
2008
.
IgE in allergy and asthma today.
Nat. Rev. Immunol.
8
:
205
217
.
2.
Fink
K.
2012
.
Origin and function of circulating plasmablasts during acute viral infections.
Front. Immunol.
3
:
78
.
3.
Takemori
T.
,
D.
Tarlinton
,
F.
Hiepe
,
A.
Radbruch
.
2015
.
Chapter 14: B cell memory and plasma cell development.
In
Molecular Biology of B Cells
, 2nd Ed.
F. W.
Alt
,
T.
Honjo
,
A.
Radbruch
,
M.
Reth
.
Elsevier
,
Philadelphia
, p.
227
249
.
4.
Croote
D.
,
S.
Darmanis
,
K. C.
Nadeau
,
S. R.
Quake
.
2018
.
High-affinity allergen-specific human antibodies cloned from single IgE B cell transcriptomes.
Science
362
:
1306
1309
.
5.
Berkowska
M. A.
,
J. J.
Heeringa
,
E.
Hajdarbegovic
,
M.
van der Burg
,
H. B.
Thio
,
P. M.
van Hagen
,
L.
Boon
,
A.
Orfao
,
J. J. M.
van Dongen
,
M. C.
van Zelm
.
2014
.
Human IgE(+) B cells are derived from T cell-dependent and T cell-independent pathways.
J. Allergy Clin. Immunol.
134
:
688
697.e6
.
6.
Heeringa
J. J.
,
L.
Rijvers
,
N. J.
Arends
,
G. J.
Driessen
,
S. G.
Pasmans
,
J. J. M.
van Dongen
,
J. C.
de Jongste
,
M. C.
van Zelm
.
2018
.
IgE-expressing memory B cells and plasmablasts are increased in blood of children with asthma, food allergy, and atopic dermatitis.
Allergy
73
:
1331
1336
.
7.
Jiménez-Saiz
R.
,
Y.
Ellenbogen
,
K.
Bruton
,
P.
Spill
,
D. D.
Sommer
,
H.
Lima
,
S.
Waserman
,
S. U.
Patil
,
W. G.
Shreffler
,
M.
Jordana
.
2019
.
Human BCR analysis of single-sorted, putative IgE+ memory B cells in food allergy.
J. Allergy Clin. Immunol.
144
:
336
339.e6
.
8.
Ramadani
F.
,
N.
Upton
,
P.
Hobson
,
Y.-C.
Chan
,
D.
Mzinza
,
H.
Bowen
,
C.
Kerridge
,
B. J.
Sutton
,
D. J.
Fear
,
H. J.
Gould
.
2015
.
Intrinsic properties of germinal center-derived B cells promote their enhanced class switching to IgE.
Allergy
70
:
1269
1277
.
9.
Eckl-Dorna
J.
,
S.
Villazala-Merino
,
N. J.
Campion
,
M.
Byazrova
,
A.
Filatov
,
D.
Kudlay
,
A.
Karsonova
,
K.
Riabova
,
M.
Khaitov
,
A.
Karaulov
, et al
2019
.
Tracing IgE-producing cells in allergic patients.
Cells
8
:
994
.
10.
Horst
A.
,
N.
Hunzelmann
,
S.
Arce
,
M.
Herber
,
R. A.
Manz
,
A.
Radbruch
,
R.
Nischt
,
J.
Schmitz
,
M.
Assenmacher
.
2002
.
Detection and characterization of plasma cells in peripheral blood: correlation of IgE+ plasma cell frequency with IgE serum titre.
Clin. Exp. Immunol.
130
:
370
378
.
11.
Fellmann
M.
,
P.
Buschor
,
S.
Röthlisberger
,
F.
Zellweger
,
M.
Vogel
.
2015
.
High affinity targeting of CD23 inhibits IgE synthesis in human B cells.
Immun. Inflamm. Dis.
3
:
339
349
.
12.
Wagner
B.
,
J. M.
Hillegas
,
S.
Babasyan
.
2012
.
Monoclonal antibodies to equine CD23 identify the low-affinity receptor for IgE on subpopulations of IgM+ and IgG1+ B-cells in horses.
Vet. Immunol. Immunopathol.
146
:
125
134
.
13.
Acharya
M.
,
G.
Borland
,
A. L.
Edkins
,
L. M.
Maclellan
,
J.
Matheson
,
B. W.
Ozanne
,
W.
Cushley
.
2010
.
CD23/FcεRII: molecular multi-tasking.
Clin. Exp. Immunol.
162
:
12
23
.
14.
Engeroff
P.
,
M.
Vogel
.
2021
.
The role of CD23 in the regulation of allergic responses.
Allergy
76
:
1981
1989
.
15.
Engeroff
P.
,
F.
Caviezel
,
D.
Mueller
,
F.
Thoms
,
M. F.
Bachmann
,
M.
Vogel
.
2020
.
CD23 provides a noninflammatory pathway for IgE-allergen complexes.
J. Allergy Clin. Immunol.
145
:
301
311.e4
.
16.
Liu
C.
,
K.
Richard
,
M.
Wiggins
,
X.
Zhu
,
D. H.
Conrad
,
W.
Song
.
2016
.
CD23 can negatively regulate B-cell receptor signaling.
Sci. Rep.
6
:
25629
.
17.
Jabs
F.
,
M.
Plum
,
N. S.
Laursen
,
R. K.
Jensen
,
B.
Mølgaard
,
M.
Miehe
,
M.
Mandolesi
,
M. M.
Rauber
,
W.
Pfützner
,
T.
Jakob
, et al
2018
.
Trapping IgE in a closed conformation by mimicking CD23 binding prevents and disrupts FcεRI interaction.
Nat. Commun.
9
:
7
.
18.
Palaniyandi
S.
,
X.
Liu
,
S.
Periasamy
,
A.
Ma
,
J.
Tang
,
M.
Jenkins
,
W.
Tuo
,
W.
Song
,
A. D.
Keegan
,
D. H.
Conrad
,
X.
Zhu
.
2015
.
Inhibition of CD23-mediated IgE transcytosis suppresses the initiation and development of allergic airway inflammation.
Mucosal Immunol.
8
:
1262
1274
.
19.
Peng
W.
,
W.
Grobe
,
G.
Walgenbach-Brünagel
,
S.
Flicker
,
C.
Yu
,
M.
Sylvester
,
J.-P.
Allam
,
J.
Oldenburg
,
N.
Garbi
,
R.
Valenta
,
N.
Novak
.
2017
.
Distinct expression and function of FcεRII in human B cells and monocytes.
J. Immunol.
198
:
3033
3044
.
20.
Engeroff
P.
,
M.
Fellmann
,
D.
Yerly
,
M. F.
Bachmann
,
M.
Vogel
.
2018
.
A novel recycling mechanism of native IgE-antigen complexes in human B cells facilitates transfer of antigen to dendritic cells for antigen presentation.
J. Allergy Clin. Immunol.
142
:
557
568.e6
.
21.
Triqueneaux
G.
,
C.
Burny
,
O.
Symmons
,
S.
Janczarski
,
H.
Gruffat
,
G.
Yvert
.
2020
.
Cell-to-cell expression dispersion of B-cell surface proteins is linked to genetic variants in humans.
Commun. Biol.
3
:
346
.
22.
Yao
Y.
,
N.
Wang
,
C.-L.
Chen
,
L.
Pan
,
Z.-C.
Wang
,
J.
Yunis
,
Z.-A.
Chen
,
Y.
Zhang
,
S.-T.
Hu
,
X.-Y.
Xu
, et al
2020
.
CD23 expression on switched memory B cells bridges T-B cell interaction in allergic rhinitis.
Allergy
75
:
2599
2612
.
23.
Selb
R.
,
J.
Eckl-Dorna
,
A.
Neunkirchner
,
K.
Schmetterer
,
K.
Marth
,
J.
Gamper
,
B.
Jahn-Schmid
,
W. F.
Pickl
,
R.
Valenta
,
V.
Niederberger
.
2017
.
CD23 surface density on B cells is associated with IgE levels and determines IgE-facilitated allergen uptake, as well as activation of allergen-specific T cells.
J. Allergy Clin. Immunol.
139
:
290
299.e4
.
24.
Larson
E. M.
,
B.
Wagner
.
2021
.
Viral infection and allergy - What equine immune responses can tell us about disease severity and protection.
Mol. Immunol.
135
:
329
341
.
25.
Larson
E. M.
,
S.
Babasyan
,
B.
Wagner
.
2021
.
IgE-binding monocytes have an enhanced ability to produce IL-8 (CXCL8) in animals with naturally occurring allergy.
J. Immunol.
206
:
2312
2321
.
26.
Larson
E. M.
,
S.
Babasyan
,
B.
Wagner
.
2020
.
Phenotype and function of IgE-binding monocytes in equine Culicoides hypersensitivity.
PLoS One
15
:
e0233537
.
27.
Wagner
B.
,
W. H.
Miller
Jr.
,
H. N.
Erb
,
D. P.
Lunn
,
D. F.
Antczak
.
2009
.
Sensitization of skin mast cells with IgE antibodies to Culicoides allergens occurs frequently in clinically healthy horses.
Vet. Immunol. Immunopathol.
132
:
53
61
.
28.
Wagner
B.
,
W. H.
Miller
,
E. E.
Morgan
,
J. M.
Hillegas
,
H. N.
Erb
,
W.
Leibold
,
D. F.
Antczak
.
2006
.
IgE and IgG antibodies in skin allergy of the horse.
Vet. Res.
37
:
813
825
.
29.
Wagner
B.
,
T.
Stokol
,
D. M.
Ainsworth
.
2010
.
Induction of interleukin-4 production in neonatal IgE+ cells after crosslinking of maternal IgE.
Dev. Comp. Immunol.
34
:
436
444
.
30.
Wagner
B.
2006
.
Immunoglobulins and immunoglobulin genes of the horse.
Dev. Comp. Immunol.
30
:
155
164
.
31.
Wagner
B.
,
B. A.
Childs
,
H. N.
Erb
.
2008
.
A histamine release assay to identify sensitization to Culicoides allergens in horses with skin hypersensitivity.
Vet. Immunol. Immunopathol.
126
:
302
308
.
32.
Braverman
Y.
1988
.
Preferred landing sites of Culicoides species (Diptera: Ceratopogonidae) on a horse in Israel and its relevance to summer seasonal recurrent dermatitis (sweet itch).
Equine Vet. J.
20
:
426
429
.
33.
Larsen
H. J.
,
S. H.
Bakke
,
R.
Mehl
.
1988
.
Intradermal challenge of Icelandic horses in Norway and Iceland with extracts of Culicoides spp.
Acta Vet. Scand.
29
:
311
314
.
34.
Greiner
E. C.
,
V. A.
Fadok
,
E. B.
Rabin
.
1990
.
Equine Culicoides hypersensitivity in Florida: biting midges aspirated from horses.
Med. Vet. Entomol.
4
:
375
381
.
35.
Anderson
G. S.
,
P.
Belton
,
N.
Kleider
.
1993
.
Hypersensitivity of horses in British Columbia to extracts of native and exotic species of Culicoides (Diptera: Ceratopogonidae).
J. Med. Entomol.
30
:
657
663
.
36.
Littlewood
J. D.
1998
.
Incidence of recurrent seasonal pruritus (‘sweet itch’) in British and German shire horses.
Vet. Rec.
142
:
66
67
.
37.
Steinman
A.
,
G.
Peer
,
E.
Klement
.
2003
.
Epidemiological study of Culicoides hypersensitivity in horses in Israel.
Vet. Rec.
152
:
748
751
.
38.
Pilsworth
R. C.
,
D. C.
Knottenbelt
.
2004
.
Equine insect hypersensitivity.
Equine Vet. Educ.
16
:
324
325
.
39.
Wagner
B.
2016
.
The immune system of horses and other equids.
In
Encyclopedia of Immunobiology.
M. J. H.
Ratcliffe
.
Academic Press
,
Oxford
, p.
549
555
.
40.
Miller
J. E.
,
S.
Mann
,
A.
Fettelschoss-Gabriel
,
B.
Wagner
.
2019
.
Comparison of three clinical scoring systems for Culicoides hypersensitivity in a herd of Icelandic horses.
Vet. Dermatol.
30
:
536
e163
.
41.
Nutt
S. L.
,
P. D.
Hodgkin
,
D. M.
Tarlinton
,
L. M.
Corcoran
.
2015
.
The generation of antibody-secreting plasma cells.
Nat. Rev. Immunol.
15
:
160
171
.
42.
Jourdan
M.
,
A.
Caraux
,
J.
De Vos
,
G.
Fiol
,
M.
Larroque
,
C.
Cognot
,
C.
Bret
,
C.
Duperray
,
D.
Hose
,
B.
Klein
.
2009
.
An in vitro model of differentiation of memory B cells into plasmablasts and plasma cells including detailed phenotypic and molecular characterization.
Blood
114
:
5173
5181
.
43.
Wagner
B.
,
G.
Siebenkotten
,
A.
Radbruch
,
W.
Leibold
.
2001
.
Nucleotide sequence and restriction fragment length polymorphisms of the equine Cvarepsilon gene.
Vet. Immunol. Immunopathol.
82
:
193
202
.
44.
Tomlinson
J. E.
,
B.
Wagner
,
M. J. B.
Felippe
,
G. R.
Van de Walle
.
2018
.
Multispectral fluorescence-activated cell sorting of B and T cell subpopulations from equine peripheral blood.
Vet. Immunol. Immunopathol.
199
:
22
31
.
45.
Chihara
N.
,
T.
Aranami
,
W.
Sato
,
Y.
Miyazaki
,
S.
Miyake
,
T.
Okamoto
,
M.
Ogawa
,
T.
Toda
,
T.
Yamamura
.
2011
.
Interleukin 6 signaling promotes anti-aquaporin 4 autoantibody production from plasmablasts in neuromyelitis optica.
Proc. Natl. Acad. Sci. USA
108
:
3701
3706
.
46.
Bortnick
A.
,
C.
Murre
.
2016
.
Cellular and chromatin dynamics of antibody-secreting plasma cells.
Wiley Interdiscip. Rev. Dev. Biol.
5
:
136
149
.
47.
Gould
H. J.
,
Y. B.
Wu
.
2018
.
IgE repertoire and immunological memory: compartmental regulation and antibody function.
Int. Immunol.
30
:
403
412
.
48.
Siracusa
M. C.
,
B. S.
Kim
,
J. M.
Spergel
,
D.
Artis
.
2013
.
Basophils and allergic inflammation.
J. Allergy Clin. Immunol.
132
:
789
801
,
quiz 788
.
49.
Shin
J.-S.
,
A. M.
Greer
.
2015
.
The role of FcεRI expressed in dendritic cells and monocytes.
Cell. Mol. Life Sci.
72
:
2349
2360
.
50.
Rosenberg
H. F.
,
K. D.
Dyer
,
P. S.
Foster
.
2013
.
Eosinophils: changing perspectives in health and disease.
Nat. Rev. Immunol.
13
:
9
22
.
51.
Yanase
Y.
,
Y.
Matsuo
,
T.
Kawaguchi
,
K.
Ishii
,
A.
Tanaka
,
K.
Iwamoto
,
S.
Takahagi
,
M.
Hide
.
2018
.
Activation of human peripheral basophils in response to high IgE antibody concentrations without antigens.
Int. J. Mol. Sci.
20
:
45
52
.
52.
Pruzansky
J. J.
,
L. C.
Grammer
,
R.
Patterson
,
M.
Roberts
.
1983
.
Dissociation of IgE from receptors on human basophils. I. Enhanced passive sensitization for histamine release.
J. Immunol.
131
:
1949
1953
.
53.
Kumagai
K.
,
T.
Abo
,
T.
Sekizawa
,
M.
Sasaki
.
1975
.
Studies of surface immunoglobulins on human B lymphocytes. I. Dissociation of cell-bound immunoglobulins with acid pH or at 37 degrees C.
J. Immunol.
115
:
982
987
.
54.
Yang
Z.
,
M. J.
Robinson
,
C. D. C.
Allen
.
2014
.
Regulatory constraints in the generation and differentiation of IgE-expressing B cells.
Curr. Opin. Immunol.
28
:
64
70
.
55.
Qiu
C.
,
L.
Zhong
,
C.
Huang
,
J.
Long
,
X.
Ye
,
J.
Wu
,
W.
Dai
,
W.
Lv
,
C.
Xie
,
J.
Zhang
.
2020
.
Cell-bound IgE and plasma IgE as a combined clinical diagnostic indicator for allergic patients.
Sci. Rep.
10
:
4700
.
56.
Jacobi
A. M.
,
H.
Mei
,
B. F.
Hoyer
,
I. M.
Mumtaz
,
K.
Thiele
,
A.
Radbruch
,
G.-R.
Burmester
,
F.
Hiepe
,
T.
Dörner
.
2010
.
HLA-DRhigh/CD27high plasmablasts indicate active disease in patients with systemic lupus erythematosus.
Ann. Rheum. Dis.
69
:
305
308
.
57.
Sanz
I.
,
C.
Wei
,
S. A.
Jenks
,
K. S.
Cashman
,
C.
Tipton
,
M. C.
Woodruff
,
J.
Hom
,
F. E.-H.
Lee
.
2019
.
Challenges and opportunities for consistent classification of human B cell and plasma cell populations.
Front. Immunol.
10
:
2458
2475
.
58.
Tellier
J.
,
S. L.
Nutt
.
2017
.
Standing out from the crowd: how to identify plasma cells.
Eur. J. Immunol.
47
:
1276
1279
.
59.
Patel
R. S.
,
J. E.
Tomlinson
,
T. J.
Divers
,
G. R.
Van de Walle
,
B. R.
Rosenberg
.
2021
.
Single-cell resolution landscape of equine peripheral blood mononuclear cells reveals diverse cell types including T-bet+ B cells.
BMC Biol.
19
:
13
.
60.
Hoof
I.
,
V.
Schulten
,
J. A.
Layhadi
,
T.
Stranzl
,
L. H.
Christensen
,
S.
Herrera de la Mata
,
G.
Seumois
,
P.
Vijayanand
,
C.
Lundegaard
,
K.
Niss
, et al
2020
.
Allergen-specific IgG+ memory B cells are temporally linked to IgE memory responses.
J. Allergy Clin. Immunol.
146
:
180
191
.
61.
Pignarre
A.
,
F.
Chatonnet
,
G.
Caron
,
M.
Haas
,
F.
Desmots
,
T.
Fest
.
2021
.
Plasmablasts derive from CD23- activated B cells after the extinction of IL-4/STAT6 signaling and IRF4 induction.
Blood
137
:
1166
1180
.
62.
Carlsson
F.
,
F.
Hjelm
,
D. H.
Conrad
,
B.
Heyman
.
2007
.
IgE enhances specific antibody and T-cell responses in mice overexpressing CD23.
Scand. J. Immunol.
66
:
261
270
.
63.
Broström
H.
,
A.
Larsson
,
M.
Troedsson
.
1987
.
Allergic dermatitis (sweet itch) of Icelandic horses in Sweden: an epidemiological study.
Equine Vet. J.
19
:
229
236
.
64.
Björnsdóttir
S.
,
J.
Sigvaldadóttir
,
H.
Broström
,
B.
Langvad
,
A.
Sigurdsson
.
2006
.
Summer eczema in exported Icelandic horses: influence of environmental and genetic factors.
Acta Vet. Scand.
48
:
3
.
65.
Schaffartzik
A.
,
E.
Hamza
,
J.
Janda
,
R.
Crameri
,
E.
Marti
,
C.
Rhyner
.
2012
.
Equine insect bite hypersensitivity: what do we know?
Vet. Immunol. Immunopathol.
147
:
113
126
.
66.
Schaffartzik
A.
,
E.
Marti
,
R.
Crameri
,
C.
Rhyner
.
2010
.
Cloning, production and characterization of antigen 5 like proteins from Simulium vittatum and Culicoides nubeculosus, the first cross-reactive allergen associated with equine insect bite hypersensitivity.
Vet. Immunol. Immunopathol.
137
:
76
83
.
67.
Schaffartzik
A.
,
E.
Marti
,
S.
Torsteinsdottir
,
P. S.
Mellor
,
R.
Crameri
,
C.
Rhyner
.
2011
.
Selective cloning, characterization, and production of the Culicoides nubeculosus salivary gland allergen repertoire associated with equine insect bite hypersensitivity.
Vet. Immunol. Immunopathol.
139
:
200
209
.
68.
van der Meide
N. M. A.
,
N.
Roders
,
M. M.
Sloet van Oldruitenborgh-Oosterbaan
,
P. J.
Schaap
,
M. M.
van Oers
,
W.
Leibold
,
H. F. J.
Savelkoul
,
E.
Tijhaar
.
2013
.
Cloning and expression of candidate allergens from Culicoides obsoletus for diagnosis of insect bite hypersensitivity in horses.
Vet. Immunol. Immunopathol.
153
:
227
239
.
69.
Novotny
E. N.
,
S. J.
White
,
A. D.
Wilson
,
S. B.
Stefánsdóttir
,
E.
Tijhaar
,
S.
Jonsdóttir
,
R.
Frey
,
D.
Reiche
,
H.
Rose
,
C.
Rhyner
, et al
2021
.
Component-resolved microarray analysis of IgE sensitization profiles to Culicoides recombinant allergens in horses with insect bite hypersensitivity.
Allergy
76
:
1147
1157
.
70.
Bogaert
L.
,
M.
Van Poucke
,
C.
De Baere
,
L.
Peelman
,
F.
Gasthuys
,
A.
Martens
.
2006
.
Selection of a set of reliable reference genes for quantitative real-time PCR in normal equine skin and in equine sarcoids.
BMC Biotechnol.
6
:
24
.
71.
Holmes
M. L.
,
C.
Pridans
,
S. L.
Nutt
.
2008
.
The regulation of the B-cell gene expression programme by Pax5.
Immunol. Cell Biol.
86
:
47
53
.
72.
Qian
Y.
,
C.
Wei
,
F.
Eun-Hyung Lee
,
J.
Campbell
,
J.
Halliley
,
J. A.
Lee
,
J.
Cai
,
Y. M.
Kong
,
E.
Sadat
,
E.
Thomson
, et al
2010
.
Elucidation of seventeen human peripheral blood B-cell subsets and quantification of the tetanus response using a density-based method for the automated identification of cell populations in multidimensional flow cytometry data.
Cytometry B Clin. Cytom.
78
(
Suppl 1
):
S69
S82
.
73.
Jourdan
M.
,
A.
Caraux
,
G.
Caron
,
N.
Robert
,
G.
Fiol
,
T.
Rème
,
K.
Bolloré
,
J.-P.
Vendrell
,
S.
Le Gallou
,
F.
Mourcin
, et al
2011
.
Characterization of a transitional preplasmablast population in the process of human B cell to plasma cell differentiation.
J. Immunol.
187
:
3931
3941
.

The authors have submitted a patent application entitled “IgE+ plasmablasts as a predictive biomarker of allergy” that uses technology described in this article.

Supplementary data