SEARCH

SEARCH BY CITATION

Keywords:

  • apoptosis;
  • basophils;
  • Hoxb8;
  • IL-3;
  • mast cells

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Author contributions
  8. Conflict of interest
  9. References
  10. Supporting Information

Background

Basophils constitute a rare leukocyte population known for their effector functions in inflammation and allergy, as well as more recently described immunoregulatory roles. Besides their low frequency, functional analysis of basophils is hindered by a short life span, inefficient ex vivo differentiation protocols, and lack of suitable cell models. A method to produce large quantities of basophils in vitro would facilitate basophil research and constitute a sought-after tool for diagnostic and drug testing purposes.

Methods

A method is described to massively expand bone marrow–derived basophils in vitro. Myeloid progenitors are conditionally immortalized using Hoxb8 in the presence of interleukin-3 (IL-3) and outgrowing cell lines selected for their potential to differentiate into basophils upon shutdown of Hoxb8 expression.

Results

IL-3-dependent, conditional Hoxb8-immortalized progenitor cell lines can be expanded and maintained in culture for prolonged periods. Upon shutdown of Hoxb8 expression, near-unlimited numbers of mature functional basophils can be differentiated in vitro within six days. The cells are end-differentiated and short-lived and express basophil-specific surface markers and proteases. Upon IgE- as well as C5a-mediated activation, differentiated basophils release granule enzymes and histamine and secrete Th2-type cytokines (IL-4, IL-13) and leukotriene C4. IL-3-deprivation induces apoptosis correlating with upregulation of the BH3-only proteins BCL-2-interacting mediator of cell death (BIM) and p53 upregulated modulator of apoptosis (PUMA) and downregulation of proviral integration site for Moloney murine leukemia virus 1 kinase (PIM-1).

Conclusion

A novel method is presented to generate quantitative amounts of mouse basophils in vitro, which moreover allows genetic manipulation of conditionally immortalized progenitors. This approach may represent a useful alternative method to isolating primary basophils.

Abbreviations
4-OHT

4-hydroxytamoxifen

BIM

BCL-2-interacting mediator of cell death

C5a

complement component C5a

GM-CSF

granulocyte macrophage colony-stimulating factor

IgE

immunoglobulin E

IL-3/-4/-13

interleukin-3/-4/-13

LTC4

leukotriene C4

mMCP

mouse mast cell protease

PIM-1

proviral integration site for Moloney murine leukemia virus 1

PUMA

p53 upregulated modulator of apoptosis

qPCR

quantitative RT-PCR

SCF

stem cell factor

Th2

T helper lymphocyte type 2

XIAP

X-linked inhibitor of apoptosis

Basophils develop from hematopoietic progenitors in the bone marrow and are released into the blood upon maturation. They constitute the least common granulocyte population, representing <1% of circulating leukocytes in the mouse [1]. Research on basophils is hindered by their low frequency, short expected half-life of a few days, and their phenotypic similarities to mast cells [1]. Both basophils and mast cells express the high-affinity receptor for immunoglobulin E (FcεRI) and contain granules rich in histamine and specific enzymes, which, together with de novo produced important cytokines and lipid mediators, are released upon activation [2]. However, in contrast to basophils, mast cells exit the bone marrow as progenitors and mature in peripheral tissue where they can persist for prolonged times [1]. Phenotypically, mouse basophils can be distinguished from mast cells by their lack of surface c-kit (CD117) expression and differential expression of several mouse mast cell proteases [3].

Recently, new tools for functional in vivo analysis of basophils have been developed, including basophil-depleting antibodies [4-7] and basophil-deficient mouse models [8, 9]. It has since become clear that basophils possess previously unrecognized functions in immune regulation and allergic responses. Basophils are now recognized as both initiators and effectors of chronic allergy, as important players in IgG (but not IgE)-mediated systemic anaphylaxis, and to possess a driving role in Th2 cell differentiation (reviewed in [1]). Basophils are an important source of Th2-type cytokines and, in fact, one of the first sources of interleukin-4 (IL-4), inducing and/or amplifying Th2 polarization in response to various stimuli, including signaling through FcεRI [1, 5, 10-12].

Interleukin 3 (IL-3) constitutes the best-characterized basophil growth factor and differentiation cytokine for human and mouse basophils and mouse mast cells [13, 14]. Human and mouse basophils express high levels of IL-3 receptor, and IL-3 increases the life span of human basophils [15, 16]. However, the mechanisms of IL-3-promoted basophil survival are poorly understood. Both phosphatidylinositol-3 kinase (PI3K) and the serine/threonine kinase proviral integration site for Moloney murine leukemia virus 1 (PIM-1) have been proposed to act downstream of IL-3 in human basophils [15, 16]. In the mouse, prosurvival IL-3 signaling has mainly been studied in other hematopoietic lineages [17-19].

Basophils are difficult to differentiate ex vivo in sufficient numbers and purity as mast cells predominate rapidly upon culture of bone marrow cells with IL-3 [20]. Importantly, no suitable mouse or human model cell line is available that convincingly mimics basophil function. Recently, Wang et al. described a system to generate large amounts of mouse neutrophils and macrophages ex vivo using conditional Hoxb8 [21]. The transcription factor Hoxb8 is able to promote the expansion of hematopoietic progenitors, and enforced expression of Hoxb8 blocks differentiation of stem cell factor (SCF)- or granulocyte macrophage colony-stimulating factor (GM-CSF)-dependent myeloid progenitors [22]. Here, we present a modification of the method by Wang et al. to generate IL-3-dependent, conditional Hoxb8-immortalized myeloid progenitors, which can be differentiated ‘on demand’ and in near-unlimited amounts into functional basophils. This is to our knowledge the first description of quantitative production of basophils.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Author contributions
  8. Conflict of interest
  9. References
  10. Supporting Information

Mice and reagents

C57BL/6 wild-type mice were maintained under pathogen-free conditions. Animal experiments were approved by the animal experimentation review board of the Canton of Bern (41/08 and BE31/11).

RPMI 1640/GlutaMAXTM medium was from Life Technologies (Zug, CH). Fetal calf serum (FCS, PAA Clone) was from PAA Laboratories GmbH, AT. 4-hydroxytamoxifen (4-OHT) was from Sigma-Aldrich (Buchs, CH). Q-VD-OPh was from SM Biochemicals (Anaheim CA, US). Recombinant mouse C5a was from HyCult Biotech (Uden, NL).

Generation of IL-3−condHoxb8 progenitor lines and differentiation into basophils

The coding sequence of mouse Hoxb8 was cloned into the pF-5xUAS-SV40-puro-Gev16 lentiviral vector system [23] as described elsewhere (Salmanidis et al., in revision). Hematopoietic progenitor cells were enriched from bone marrow of C57BL/6 mice by magnetic cell separation using a lineage depletion cocktail (BD IMagTM, BD Biosciences Europe) following the manufacturer's instructions. 5 × 105 freshly isolated, or cryoconserved, lineage-negative cells were incubated for 24 h or 36-48 h, respectively, in complete RPMI medium (RPMI 1640/GlutaMAXTM, 10% FCS, 1% Penicillin/Streptomycin, 50 μM 2-mercaptoethanol) in the presence of ≥200 pg/ml IL-3 (added as WEHI-3B cell conditioned medium as a source for murine IL-3; concentration quantified by ELISA, BioLegend, San Diego, US). Cells were then spin infected with pF-5xUAS-Hoxb8(mm)-SV40-puro-GEV16 lentiviral particles in the presence of 8 μg/ml polybrene at 30°C for 90 min. Cells were cultivated in complete RPMI/IL-3 and Hoxb8 expression induced by addition of 100 nM 4-OHT. 3–5 days after infection, cells were selected with 1 μg/ml puromycin for a minimum of 3 weeks. Outgrowing IL-3-dependent cell lines were selected for absence of CD117 expression for further analysis. To differentiate progenitors into mature basophils, cells were washed twice in PBS and reseeded at 7.5 × 104 cells/ml in complete RPMI medium containing IL-3 (≥200 pg/ml) for 6 days.

FcεRI crosslinking and degranulation assay

1 × 106 cells were sensitized in RPMI-1640/GlutaMAXTM medium containing 0.2% BSA (Sigma) with mouse monoclonal anti-trinitrophenyl (TNP) IgE (IgEl-b4, ATCC), added as 15% hybridoma supernatant for 90 min at 37°C. Cells were washed and activated with 100 ng/ml TNP-BSA (Biosearch Tech Inc., Novato, US) for 30 min. 140 μl of cellular supernatant was added to 60 μl of 8 mM p-nitrophenol N-acetyl-β-D-glucosaminide (Sigma) in 0.08 M citrate solution, pH 4.5, and incubated for two hours at 37°C. The reaction was stopped by addition of 0.2 M glycine, pH 10.7, and results analyzed on a spectrometer plate reader (SpectraMax M2e, Molecular Devices, λ = 405 nm).

qPCR analysis

Total RNA was extracted from 5 × 106 cells (SV total RNA isolation system, Promega AG, Dübendorf, CH), and 1 μg RNA was reverse-transcribed using oligo(dT) primer and M-MLV reverse transcriptase following the manufacturer's instructions (Promega). Quantitative RT-PCR (qPCR) analysis was performed using HOT FIREPol® EvaGreen® qPCR Mix Plus (Solis Biodyne, Tartu, EE) on a multicolor Real-Time PCR detection system (iQ5, BIO-RAD). Primer sequences specific for the various mast cell proteases were: Prss34 (mMCP-11, amplicon 149 bp) Fw 5′-GGACATTGCTCTGCTGAAAC, Rev 5′-GCAGTGGCATGTAGTTCTCG; Mcpt8 (mMCP-8, amplicon 134 bp) Fw 5′-TCATTCCTGTCAGTGAAGCC, Rev 5′-GCTCTTTGGAAGGGCAATAG; Mcpt4 (mMCP-4, amplicon 128 bp) Fw 5′-TGAGGGAGGTGAAACTGAGA, Rev 5′-TCCAGAGTCTCCCTTGTATGC; Cma1 (mMCP-5, amplicon 74 bp) Fw 5′-AGCACCAAAGCTGGAGAGAT, Rev 5′-TCCAGATAGGCCATGTAGGG. Hprt1 [24] (HPRT, amplicon 124 bp) was used as reference gene: Fw 5′-TGGATACAGGCCAGACTTTGTT, Rev 5′-CAGATTCAACTTGCGCTCATC.

Statistical analysis

Statistical analysis was performed using a Student's t-test. P < 0.05 was considered statistically significant.

Online Supporting Information

See the online Supporting Information for additional description of methods.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Author contributions
  8. Conflict of interest
  9. References
  10. Supporting Information

Generation of mouse basophils in vitro using conditional Hoxb8

Freshly isolated lineage-negative progenitors from bone marrow of C57BL/6 mice were incubated for 24 h in IL-3-containing medium prior to transduction with lentiviral particles carrying an inducible expression system for Hoxb8 (Fig. 1A). Hoxb8 overexpression was induced by addition of 100 nM 4-OHT and successfully transduced cells selected with puromycin. Outgrowing cell lines, termed IL-3−condHoxb8, were cycling and maintaining their progenitor status over prolonged periods (up to several months) in culture. IL-3−condHoxb8 progenitors displayed a bipolar surface staining for FcεRI, and lines that stained negative for c-kit (CD117) were selected for further analysis (Fig. 1C,D).

image

Figure 1. Generation of IL-3−condHoxb8 basophils. (A) Schematic overview. (B) Western blot analysis of exogenous Hoxb8 protein levels upon removal of 4-OHT. Re-probing with anti-GAPDH antibody served as loading control. Representative immunoblots of at least three independent experiments are shown. (C) H&E staining and FcεRI surface expression analysis of differentiating (-4-OHT) IL-3−condHoxb8 cells. Representative examples of at least three independent experiments are shown. Isotype controls are shown as dotted lines. (D) Representative dot plots (n ≥ 3) of surface c-kit and FcεRI expression of IL-3−condHoxb8 progenitors, differentiated basophils, differentiated SCF−condHoxb8 neutrophils, and bone marrow–derived mast cells (BMMC). (E) Flow cytometric analysis for the indicated surface markers of cells as in (D). Representative histograms of at least three independent experiments are shown; isotype controls are shown as dotted lines.

Download figure to PowerPoint

In order to drive IL-3−condHoxb8 progenitors into differentiation, 4-OHT was removed and cells reseeded at low density in the presence of IL-3. This resulted in the rapid loss of Hoxb8 protein within 2 days (Fig. 1B) and appearance of mature granulocytes within 6 days (Fig. 1C). The overall cellular expansion during differentiation was six- to eightfold, whereas cells stopped to cycle at around day 4 (not shown). The percentage of FcεRI+c-kit cells increased gradually during differentiation and was routinely above 95% at day 4 (Figs. 1C and S1). The surface marker staining profile on differentiated cells was FcεRI+c-kitCD123+CD88+Gr-1CD11b+Siglec-FlowCCR3, matching known mouse basophil markers [25] and clearly distinguishing them from bone marrow–derived mast cells (BMMC), neutrophils, or eosinophils (Figs. 1E, 4A, 5A and not shown). Terminally differentiated cells underwent spontaneous apoptosis (Fig. 2A). Of note, spontaneous apoptosis of mature IL-3-condHoxb8 basophils was significantly slower than of mature SCF−condHoxb8 neutrophils, correlating with higher levels of anti-apoptotic BCL-2 in 5-day differentiated basophils compared with 5-day differentiated neutrophils (Fig. 2A–C).

image

Figure 2. End-differentiated IL-3−condHoxb8 basophils undergo spontaneous apoptosis in culture. (A) Analysis of spontaneous cell death (in the presence of cytokines) in 6-day differentiated IL-3−condHoxb8 basophils and SCF−condHoxb8 neutrophils, respectively. Values represent means ± SD of at least three independent experiments. **P < 0.01; ***P < 0.005. Western blot analysis (representative immunoblots of ≥ 2 independent experiments) for the indicated proteins in differentiating IL-3−condHoxb8 basophils (B) and SCF−condHoxb8 neutrophils (C).

Download figure to PowerPoint

To further analyze the FcεRI bipolar staining of IL-3−condHoxb8 progenitors, we sorted FcεRI+ and FcεRI populations by flow cytometry. Both FcεRI+c-kit and FcεRIc-kit populations rapidly re-adapted a bipolar FcεRI surface staining and were able to differentiate into basophils to comparable extents, with no evidence for enhanced cell death in either differentiating population (Fig. S2A and not shown). Addition of monoclonal IgE to IL-3−condHoxb8 progenitors did not shift the percentage of FcεRI-positivity (Fig. S2B).

IL-3−condHoxb8 basophils selectively express basophil-specific mouse mast cell proteases

Ugajin et al. showed that primary mouse basophils specifically express high levels of the mouse mast cell proteases Mcpt8 (mMCP-8) and Prss34 (mMCP-11), but no (or very little of the) mast cell-specific Mcpt4 (mMCP-4) and Cma1 (mMCP-5), respectively [3]. As shown in Figure 3A, qPCR analysis confirmed that differentiated IL-3−condHoxb8 basophils express high levels of Prss34 and Mcpt8, but nondetectable (or very low) levels of Mcpt4 and Cma1. On the other hand, BMMC expressed Mcpt4 and Cma1, but no Prss34 or Mcpt8 (Fig. 3B). Analysis of mMCP expression during differentiation revealed that Prss34 and Mcpt8 are already expressed at the progenitor stage and further increase during basophil differentiation, whereas Mcpt4 and Cma1 remain low (Fig. 3C).

image

Figure 3. IL-3−condHoxb8 basophils express high levels of basophil-specific mouse mast cell proteases. Quantitative RT-PCR (qPCR) analysis of basophil-specific (Prss34, Mcpt8) and mast cell-specific (Mcpt4, CmaI) mouse mast cell proteases in mature (A) and differentiating (C) IL-3−condHoxb8 basophils and bone marrow–derived mast cells (BMMC) (B). Data represent means ± SEM of 4 (A, C) and three (B) independent experiments. *P < 0.05; **P < 0.01; ***P < 0.005. Hprt was used as reference gene.

Download figure to PowerPoint

In summary, phenotypic analyses as well as qPCR analysis of basophil and mast cell-specific mMCPs strongly indicate that in the presence of IL-3 conditional Hoxb8-immortalized progenitors can be terminally differentiated in large amounts into basophils.

Activated IL-3−condHoxb8 basophils degranulate, release histamine, and secrete Th2 cytokines and leukotriene C4

To test whether FcεRI expressed on IL-3−condHoxb8 basophils is functional, we provoked IgE-dependent degranulation and quantified the release of N-acetyl-β-D-hexosaminidase and histamine from intracellular granules. We also tested the effect of complement component C5a, as IL-3−condHoxb8 basophils expressed C5a receptor, CD88 (Fig. 4A). Both N-acetyl-β-D-hexosaminidase and histamine were specifically released after FcεRI crosslinking, as well as after stimulation with C5a (100 nM), in IL-3−condHoxb8 basophils (Fig. 4B,C). Whereas intracellular histamine levels were low in progenitors and increased during differentiation (Fig. 4D), N-acetyl-β-D-hexosaminidase was already present in progenitors, even though release of the enzyme upon activation was only detectable at later stages of differentiation (Fig. 4B and not shown).

image

Figure 4. IL-3−condHoxb8 basophils release granule contents, Th2-type cytokines and LTC4 upon activation. (A) CD88 (C5aR) is expressed on the surface of differentiated IL-3−condHoxb8 basophils but not of progenitors. SCF−condHoxb8 neutrophils served as positive control. Isotype controls are shown in gray. (B) N-acetyl-β-D-hexosaminidase release from IL-3−condHoxb8 basophils at indicated stages of differentiation upon FcεRI crosslinking or activation with C5a for 30 min. (C) Histamine release from differentiated basophils activated as in (B) for 30 min. (D) Intracellular histamine levels of untreated IL-3−condHoxb8 basophils at indicated stages of differentiation. (E) Released IL-4 from 5-day differentiated IL-3−condHoxb8 basophils upon FcεRI crosslinking or stimulation with C5a for 24 h. IL-4 (F, representative example) and IL-13 (G) release from differentiating IL-3−condHoxb8 basophils stimulated as in (E) for 24 h. (H) Released LTC4 from cells stimulated as in (E) for 30 min. Values represent means ± SEM of ≥ 4 (B, E) and three (C, D, G, H) independent experiments.

Download figure to PowerPoint

Basophils are recognized as an important source of Th2 type cytokines. As reported for human basophils, differentiated IL-3−condHoxb8 basophils secreted IL-4 and IL-13 in response to FcεRI crosslinking or C5a (Fig. 4E–G). IL-3−condHoxb8 basophils also generated the lipid mediator leukotriene C4 (LTC4) upon FcεRI but not C5aR stimulation (Fig. 4H). Taken together, we demonstrate that IL-3−condHoxb8 basophils are functional and respond to IgE-dependent activation in an identical manner as reported for primary mouse and human basophils. They also respond to the IgE-independent trigger C5a in a fashion very similar to primary human basophils [26].

Mechanisms of IL-3 withdrawal-induced apoptosis

Both IL-3−condHoxb8 progenitors and basophils expressed high levels of the IL-3 receptor CD123 on their surface and rapidly underwent apoptosis upon IL-3-withdrawal (Fig. 5A,B). Procaspase-9 was rapidly processed and effector caspases-3 and caspases-7 activated upon IL-3 removal, whereas X-linked inhibitor of apoptosis (XIAP), an anti-apoptotic protein known to directly bind and inhibit active caspase-3, caspase-7, and caspase-9 [27], was downregulated (Fig. 5D). Interestingly, the processed and partially active p19 large subunit of caspase-3 was already present in lysates from healthy cells and rapidly further processed into the highly active p17 subunit upon IL-3 removal (Fig. 5D). In contrast to cytokine withdrawal in BMMC, cytokine-deprived IL-3−condHoxb8 progenitors and basophils could only be protected from apoptosis by addition of high (100 μM) concentrations of the pan-caspase inhibitor Q-VD-OPh (Fig. 5B,C).

image

Figure 5. IL-3−condHoxb8 cells depend on IL-3 for their survival. (A) Cell surface expression of CD123 (IL-3Rα) on IL-3−condHoxb8 progenitors and basophils. Bone marrow–derived mast cells (BMMC) are shown as positive control. Representative histograms of three independent experiments are shown; isotype controls are shown in gray. Survival curves of IL-3−condHoxb8 progenitors and 6-day differentiated basophils (B) and BMMC (C) upon cytokine deprivation with or without addition of the pan-caspase inhibitor Q-VD-OPh. Values represent means ± SD of at least three independent experiments. ***P < 0.005. (D) Western blot analysis of the indicated proteins in IL-3-deprived progenitors and 6-day differentiated IL-3−condHoxb8 basophils. Presented immunoblots are representative of two independent experiments.

Download figure to PowerPoint

Interleukin-3 withdrawal resulted in a rapid increase in the pro-apoptotic BH3-only protein BCL-2-interacting mediator of cell death (BIM) in differentiated basophils, whereas BIM protein levels were low in Hoxb8-expressing progenitors (Fig. 6A). Bim upregulation was confirmed by qPCR analysis, which further revealed that the mRNA encoding for another BH3-only protein, Puma, was similarly induced by IL-3-deprivation (Fig. 6B).

image

Figure 6. Initiation of IL-3 withdrawal-induced apoptosis. (A) Western blot analysis of the indicated proteins in IL-3-deprived IL-3−condHoxb8 progenitors and 6-day differentiated basophils. Presented immunoblots are representative of two independent experiments. (B) Quantitative RT-PCR (qPCR) analysis of Bim and Puma levels upon IL-3-deprivation. Values represent means ± SEM of three independent experiments. (C) Survival curves of IL-3−condHoxb8 progenitors and basophils treated with Wortmannin or Ly294002 in the presence of IL-3. Values represent means ±SD of three independent experiments.

Download figure to PowerPoint

Prosurvival IL-3 signaling in human basophils was reported to involve the activity of PI3K. In accordance, we observed increased cell death when IL-3−condHoxb8 progenitors and basophils, in the presence of IL-3, were treated with the PI3K inhibitors LY29004 (20 μM) or high concentrations (1 μM) of Wortmannin (Fig. 6C and not shown). However, Wortmannin at 100 nM, a concentration highly specific for PI3K [28], did not induce apoptosis (Fig. 6C). LY294002, at the concentration used, is known to also inhibit proviral integration site for Moloney murine leukemia virus 1 (PIM-1), a Ser/Thr kinase downstream of IL-3 in human basophils [29]. PIM-1 kinase activity is not subjected to strong posttranslational regulation and thus correlates with PIM-1 protein levels [30]. We found that PIM-1 protein levels rapidly decreased in IL-3−condHoxb8 progenitors and basophils upon IL-3-removal (Fig. 6A).

Taken together, we show that the survival of both IL-3−condHoxb8 progenitors and basophils critically depends on IL-3. Upon IL-3-deprivation, rapid loss of the prosurvival kinase PIM-1 and transcriptional upregulation of the pro-apoptotic BH3-only genes Bim and Puma are observed, correlating with induction of caspase-dependent apoptosis.

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Author contributions
  8. Conflict of interest
  9. References
  10. Supporting Information

The characterization of mouse basophils by functional, biochemical, and genetic methods is close to impossible due to their low frequency and the excessive numbers of animals needed for even simplest experiments. Here, we describe a novel cellular model for the study of mouse basophils. The method is based on the report by Wang et al. describing the ex vivo production of neutrophils and macrophages using conditional Hoxb8 [21]. The major difference to Wang et al. is that Hoxb8-mediated immortalization was achieved in the presence of IL-3 instead of SCF, thereby favoring commitment to the basophil lineage. Also, instead of a Hoxb8-estrogen receptor fusion protein, we conditionally expressed wild-type Hoxb8 using a lentiviral system [23] that lacks any exogenous Hoxb8 expression in the absence of 4-OHT.

IL-3−condHoxb8-immortalized cell lines maintain their growth potential and progenitor status for prolonged times. Cell lines can be potentially generated from any genetically modified mouse strain, given that it contains multipotent myeloid progenitors in the bone marrow or fetal liver. The latter is of particular interest for strains presenting with severe phenotypes. Another advantage over primary cells is the possibility to genetically manipulate IL-3−condHoxb8 progenitors by standard protocols (e.g., transgenesis, RNA interference). Near-unlimited amounts of mature basophils can be generated ‘on demand’ and in high purity upon shutdown of exogenous Hoxb8 expression, enabling or facilitating experiments requiring larger cell numbers. Besides known markers [25], we show that differentiated basophils express C5aR (CD88) and Siglec-F, the closest functional paralog to human Siglec-8, which is expressed on human basophils [31].

Like primary human and mouse basophils, IL-3−condHoxb8 basophils release preformed granule enzymes and histamine, and newly generated LTC4, IL-4, and IL-13 upon IgE-mediated activation. As the cells are constantly kept in an IL-3-rich environment, we found increased IL-13 levels in the supernatants of unstimulated basophils (Fig. 4G); this is in accordance with findings in human primary basophils [26]. Interestingly, even though a high percentage of immature basophils express surface FcεRI and contain good amounts of granule enzymes, efficient IgE-mediated release was only apparent in mature cells. A similar unresponsiveness was found for the production of LTC4 and Th2 cytokines, indicating a possible lack of FcεRI signal coupling in immature basophils, requiring further investigation. We also tested the effect of the complement-derived chemotaxin C5a known to activate human basophils. Indeed, we found that mature IL-3−condHoxb8 basophils express C5a receptor and strongly respond to C5a, comparable to IgE-mediated activation. C5a not only induced degranulation but also promoted the production of IL-4 and IL-13 indicating that such an innate, antigen-independent pathway for the production of these crucial Th2 cytokines may also operate in mice. It will be interesting to determine whether complement activation plays a role in the induction and/or amplification of certain Th2-mediated inflammatory conditions. Interestingly, C5a did not induce an LTC4 response despite the presence of IL-3 for unknown reasons requiring further investigation.

It is controversially discussed how basophils develop from hematopoietic progenitors. In humans, bipotent eosinophil/basophil and megakaryocyte/basophil progenitors have been proposed [32]. In the mouse, however, a strong lineage relationship exists between basophils and mast cells where IL-3 is a major differentiation cytokine for both lineages. This is underlined by the outgrowth of both basophils (minor and transient population) and mast cells (dominating after 3 weeks in culture) upon incubation of bone marrow with IL-3 and the description of a common basophil/mast cell progenitor in the mouse [20, 33, 34]. The IL-3−condHoxb8 basophil model may represent a useful tool to study individual stages of basophil development as well as the developmental relationship between mouse basophils and mast cells.

As reported for human basophils, we observed spontaneous apoptosis of end-differentiated basophils to be significantly slower than of neutrophils (Fig. 2A), a finding supported by the very high levels of BCL-2 in mature basophils compared with hardly detectable BCL-2 in mature neutrophils (Fig. 2B,C and [16]). IL-3−condHoxb8 progenitors and basophils both critically depended on IL-3 for their survival as IL-3 deprivation led to rapid apoptosis. Somewhat surprisingly, caspase-3 was already partially activated in healthy cells. Interestingly, the basophil-specific mast cell protease mMCP-8 is highly homologous with granzyme B [35], and it will be interesting to test the potential of mMCP-8 to directly activate procaspases. We found that IL-3 deprivation leads to a rapid upregulation of the pro-apoptotic BH3-only proteins BIM and PUMA, which is in accordance with reports in other IL-3-dependent systems [17-19]. Hoxb8 seems to negatively regulate Bim expression, as indicated by the lack of BIM in IL-3−condHoxb8 progenitors (Fig. 6A and Salmanidis et al., in revision). Our data further support recent reports on an essential role for PIM-1 kinase in prosurvival IL-3 signaling in human basophils [16] and Pim-1 downregulation in a murine system [36].

No suitable mouse or human basophil cell lines exist to date, and, with the exception of the slow growing LAD2 mast cell line, there are no human cell lines expressing the high-affinity IgE receptor. Comparable, IgE receptor positive cell lines of human origin would therefore be of great interest, also with respect to potential applications in diagnostics and for drug testing purposes.

In summary, we describe a novel and potentially useful cellular model to produce large amounts of mature mouse basophils in vitro. As with any model system, in vitro differentiated basophils cannot be seen as a general substitute for primary cell analysis or in vivo studies. However, given the rare frequency of basophils in vivo, we are convinced that the here described model constitute a powerful tool to dissect cellular and molecular aspects of basophil biology and may prove a valid alternative method to isolating primary basophils in many respects.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Author contributions
  8. Conflict of interest
  9. References
  10. Supporting Information

We thank Barbara Geering for discussions and critical comments on the manuscript; Andrea Kaufmann and the FACS sorting facility of the Department of Clinical Research, University of Bern, for expert technical assistance; John Silke (Melbourne, AU) for the lentiviral inducible system; and Andreas Strasser, David Huang, Lorraine O'Reilly (all Melbourne, AU), Gunnar Nilsson (Stockholm, SE), Georg Häcker (Munich, DE), Christoph Borner (Freiburg, DE), Shida Yousefi, and Nadia Corazza (all Bern, CH), for mice, antibodies, and reagents. This study was supported by grants to TK form the Swiss National Science Foundation (PP0033_119203) and the 3R Research Foundation Switzerland (3R-127-11).

Author contributions

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Author contributions
  8. Conflict of interest
  9. References
  10. Supporting Information

UG and TK designed and performed research and wrote the manuscript.TR, NE, and DB performed research. CAD provided critical inputs, designed, and performed research. MS, GB, PGE, and HUS provided critical tools and intellectual inputs.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Author contributions
  8. Conflict of interest
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Author contributions
  8. Conflict of interest
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
all12140-sup-0001-FigS1.pdfapplication/PDF360KFigure S1. Differentiation IL-3−condHoxb8 progenitors into mature basophils within six days upon removal of 4-OHT.
all12140-sup-0002-FigS2.pdfapplication/PDF406KFigure S2. Basophil-committed IL-3−condHoxb8 progenitors display bipolarity in FcεRI surface expression.
all12140-sup-0003-supplemental-textS1.docWord document55KData S1. Materials and methods.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.