Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Sep 14.
Published in final edited form as: J Leukoc Biol. 2025 Mar 14;117(3):qiae233. doi: 10.1093/jleuko/qiae233

Basophils Induce Pro-Tumorigenic Cytokines from A549 Lung Adenocarcinoma via Mechanisms Requiring IgE, Galectin-3, and IL-3 Priming

John T Schroeder 1,†, Laurent Ehrlich 1, Anja P Bieneman 1
PMCID: PMC12212430  NIHMSID: NIHMS2082682  PMID: 39432748

Abstract

Galectin-3 (Gal-3) is implicated in innate immune cell activation in a host of diseases/conditions. We identified a unique response whereby human basophils secrete IL-4/IL-13 when co-cultured with A549 cells –a lung adenocarcinoma. While displaying parameters consistent with standard IgE-dependent activation, these Galectin-3-dependent responses occurred in the absence of specific IgE/allergen and required cell-to-cell contact. We now hypothesize that this mode of activation also impacts A549 function. Our findings show that cytokines are induced in basophil/A549 co-cultures that are not detected when either cell is cultured alone, in particular IL-6. As previously shown for IL-4/IL-13, IL-6 production also required cell-to-cell contact and was dependent on A549-Gal-3, since clones deficient of this lectin induced less cytokine. Using culture-derived basophils (CDBA), we demonstrate that the IL-6 response, and production of another tumorigenic factor, VEGF-A, are induced in CDBA/A549 co-cultures but only after passively sensitizing CDBA with IgE, in a manner similar to IL-4/IL-13. However, IgE-dependent activation of basophils/CDBA cultured alone failed to induce IL-6/VEGF. Importantly, IL-3-primed basophils, even those fixed with paraformaldehyde, readily induced IL-6/ VEGF-A in co-cultures, thus verifying these cytokines are derived from A549. Overall, these results suggest a complex mechanism whereby Gal-3/IgE interactions between IL-3-primed basophils and A549 have the potential to modulate cytokine production by both cells. With Gal-3 implicated in many diseases ranging from asthma to cancer, but also in normal physiological conditions, such as wound healing, these findings are predicted to provide insight into the molecular mechanisms by which this lectin (and IgE) functions in these processes.

Keywords: Human, IgE, allergy, lectin, cytokine, inflammation, tumor immunity

Summary Statement:

Featuring a novel mode of activation, these findings provide mechanistic insight towards the increasing evidence that basophils play a role in wound healing and tumorigenesis.

Introduction

Human basophils have been shown to secrete distinctly high levels of IL-4/IL-13 when co-cultured with A549 epithelial cells (EC) –an adenocarcinoma of lung origin. These cytokines were also secreted by basophils co-cultured with primary lung EC, but to a lesser extent (1). This novel mode of stimulation with A549 proved dependent on cell-to-cell contact and basophil IgE/FcεRI-dependent signaling, was markedly enhanced by IL-3 yet occurred independently of allergen. Involvement of A549-associated galectin-3 (EC-Gal-3) was suspected in this reaction, given its long history as an IgE-binding lectin with multivalent binding capacity and newer evidence of its expression on EC, especially those of cancer origin. Hence, a follow-up study showed that A549 clones (1E2 & 1B10) deficient of this protein, no longer activated basophils in co-culture for production of IL-4/IL-13 (2). In contrast, the capacity to activate basophils was preserved in a control clone (3G8) that retained Gal-3 expression, inducing IL-4/IL-13 levels nearly identical to those in co-cultures with parent A549. Importantly, recombinant human Gal-3 (rhGal-3) failed to stimulate basophils when used in solution, although it suppressed the capacity A549 to activate basophils, presumably through competitive binding. In other experiments, microspheres coated with rhGal-3 (MS-GAL-3) mimicked the capacity of A549 to activate basophils for IL-4/IL-13, although to a lesser extent (2). Overall, the results indicated that immobilized rhGal-3 lectin, and that naturally expressed on the surface of A549, facilitates basophil activation by allowing for more effective crosslinking of IgE/FcεRI.

The uniqueness of the EC-Gal-3-dependent basophil response infers additional functions mediated by these granulocytes, both in normal physiology and in disease. For example, recent evidence, coming mostly from animal studies, indicate that basophils play a role in wound healing and/or tumorigenesis, by directly secreting cytokines (IL-4/IL-13) and/or inducing other cells (e.g. M2-like) to secrete cytokines and growth factors that promote these activities (3). Therefore, given that Gal-3 is implicated in many of these same progressions, we’ve considered the possibility that EC-Gal-3-dependent activation of basophils might be an underlying mechanism in such occurrences, at least in part. In fact, we further hypothesize that this mode of activation impacts A549 function, either directly or via paracrine effects mediated by the cytokines/mediators secreted by basophils.

In addressing this hypothesis, we report herein that other cytokines are, indeed, induced in basophil/A549 co-cultures but are not detected, or only minimally present, when either cell is cultured alone in particular IL-6 and VEGF-A. As previously shown for the IL-4/IL-13 produced in these co-cultures, the induction of IL-6 is also dependent on Gal-3, since co-cultures with A549 clones lacking this lectin have reduced levels of this cytokine. Likewise, the enhanced production of both IL-6 and VEGF is dependent on basophils expressing IgE, yet basophils are not the source of either cytokine. Thus, basophil-bound IgE has the potential to act as a ligand (presumably binding EC-Gal-3) to provoke A549-derived cytokines. Unexpectantly, IL-3-primed basophils fixed with paraformaldehyde (PF) are as capable of inducing IL-6 and VEGF-A from A549 as are live cells primed by IL-3, thus indicating involvement of an additional unknown basophil-associated surface protein(s). Collectively, these results indicate a complex mechanism whereby Gal-3/IgE interactions between IL-3-primed basophils and A549, modulate cytokine production by both cell types.

Materials and Methods

Special Reagents, Buffers, and Media

The following reagents were purchased: crystallized human serum albumin (Calbiochem-Behring Corp, La Jolla, CA); PIPES, FCS and crystallized BSA (Sigma-Aldrich, Allentown, PA); gentamicin, IMDM, and nonessential amino acids (Life Technologies, Inc., Grand Island, NY); Percoll (Pharmacia Biotec, Inc. Piscataway, NJ); rhIL-17 and rhIL-3 (R&D Systems, Minneapolis, MN, cat# 7955-IL010 & 203-IL-010, respectively); MOPS (3-(N-morpholino) propanesulfonic acid) buffer (Honeywell-Fluka, Pittsburg, PA); aldehyde sulfate polystyrene microspheres (MS) were 5μm in size (Invitrogen/ThermoFisher, Grand Island, NY). All PIPES-containing buffers used in this study (e.g. 1x PIPESPIPES/albumin/glucose – PAG, and PAG-EDTA) were made from a 10x solution as previously described (4). C-IMDM consisted of IMDM supplemented with 5% FCS, non-essential amino acids, L-glutamine, and 10 μg/ml gentamicin. The pH was adjusted to 7.2–7.4 before filtering through a 0.22 μm filter. The in-house polyclonal anti-human IgE (anti-IgE) was made in a goat and is well-characterized, having been used in numerous publications. The human IgE was prepared from a subject with hyper-IgE levels-JK). Both reagents were from the DACI lab within the Division here at JHU and provided by Dr. Robert Hamilton, Director. Dupilumab and its control were obtained from Dr. Jody Tversky, JHU.

A549 Cell Culture and Gal-3 Knockdown clones

A549 cells (American Type Culture Collection ATCC, Manassas, VA) were maintained in medium consisting of F-12K nutrient mixture –Kaigh’s Modification, with 10% heat-inactivated FBS and 1% penicillin streptomycin. Cultures were split twice weekly during the duration of these experiments, with no detectable changes in their capacity to activate basophils. Seeding of A549 for co-culture with basophils has been previously described (1) and was similarly done herein. Briefly, A549 cells were plated into wells of 96-well plates in 0.100 mL of medium. These were incubated up to 48h in a humidified incubator (37°C, 5% CO2) to allow cells to adhere and achieve ~90–100% confluency prior to co-culture.

The generation of stable A549 clones deficient of Gal-3 protein has been described (2). In brief, shRNA technology using a Mission® Lentiviral Transduction kit and protocol (Sigma-Aldrich, St. Louis, MO) was used to produce clones (1E2 and 1B10) showing >95% reduction in total Gal-3 protein, including surface Gal-3. This approach also produced a control clone (3G8) whereby ~50% of total Gal-3 protein, relative to A549-wild type (WT) cells, was retained following stable transfection.

Isolation of Blood Basophils

Basophil suspensions were prepared from residual TRIMA cassettes from anonymous subjects undergoing platelet pheresis. In some instances, venipuncture was performed on consenting adults (age range, 21–65) using a protocol approved by the Johns Hopkins University Institutional Review Board. Subjects were selected regardless of allergic status. Buffy-coats from both specimen sources were subjected to density centrifugation using a well-established double-Percoll protocol. This produced both basophil-depleted cell (BDC) and basophil-enriched cell (BEC) suspensions, as described in detail elsewhere (4). Basophils were purified from BEC suspensions by negative selection using an antibody cocktail & microbeads (StemCell Technologies, Vancouver, Canada, cat# 14309-A01P), and collecting the flow-thru from magnetized LS columns (Miltenyi Biotec, Gaithersburg, MD) The negative selection protocol, staining procedure, and enumeration of basophils on a Spiers-Levy counting chamber are described in detail (4). The percentage of cells staining blue, which is a direct assessment of basophil purity, ranged between 98% and >99%. Supplemental Figure 1 shows an example of basophil staining at each step of the purification procedure.

Culture-derived Basophils (CDBA) and passive sensitization with IgE

CDBAs were prepared as previously described (1, 5). In brief, hematopoietic cell precursors were isolated from 1–2×109 BDC using CD34+ positive selection (Miltenyi Biotec, Gaithersburg, MD). The number of CD34+ enriched cells typically ranged between 0.5–2.0×106, which were seeded at 5×104/ml in serum-free StemPro-34 medium containing rhIL-3 (10 ng/ml). After 12–14d at 37°C, 5% CO2, the cells, which had generally expanded in number by ~2 to 3-fold were harvested, washed 2x, counted, resuspended in C-IMDM and split 2-ways for passive sensitization accordingly: 1) medium alone and 2) with IgE (2 μg/ml). These were cultured overnight at 37°C, 5% CO2 before harvesting, washing 2x, and used for functional assays in co-cultures. CDBA were also counted using Alcian Blue staining, but with a slight modification compared to the standard approach done for blood basophils. Specifically, just 0.001 ml of 1N HCL (rather than 0.0025 ml) was added to the staining solution/cell mix. The deficit in volume was corrected by pre-adding 0.0015 ml more of saline/EDTA (0.0465 ml total) to the staining solution. Counts indicated that 94±0.7% of the cells in these suspensions (n=12) were basophil-like as determined by staining blue. A representative staining of CDBA is shown in Supplemental Figure 2.

Co-Culture Conditions

All cultures to induce cytokine production were done as previously described (1). In brief, cells were suspended in C-IMDM such that 1×105 basophils/CDBA were added (in 0.050–0.100 ml volumes) to flat-bottom wells (96-well plates) that had been pre-seeded 2d earlier with 5×103 A549, clones, or medium alone –all at 0.100 ml. Importantly, wells seeded with A549 were 90–100 % confluent at the time basophils were added, with an estimated 2×104 A549 per well. In experiments using antibodies designed to block/neutralize cell surface markers or the effects of cytokines, 0.050 ml of 4–5x antibody, isotype control, or medium alone were added immediately after adding basophils or CDBA. Co-stimulation was achieved by adding 0.050 ml of 5x stimulus (e.g. IL-3, anti-IgE, or medium alone), with co-cultures incubated as indicated at 37°C, 5% CO2. Supernatants were harvested after 20h unless otherwise indicated and tested for cytokine secretion.

Cytokine measurements

Supernatants were analyzed for cytokine protein levels by ELISA. Kits for IL-4, IL-6, and IL-13 (Invitrogen/ThermoFisher, Grand Island, NY), IL-17A (R&D Systems, Minneapolis, MN, and VEGF (BioLegend, San Diego, CA). Prior reports showed that the IL-4 and IL-13 detected in co-cultures is basophil-derived (1). Therefore, the levels of these cytokines were normalized to pg/106 basophils to be consistent with our long practice of reporting in this manner. Since the cell source of IL-6 and VEGF-A in the co-cultures was not known at the onset of the study, levels were reported in pg/ml.

Statistical Analysis

Statistical analyses were performed with Prism 9.0 software (GraphPad, Software, LaJolla, Calif.) Analyses were performed using paired t-test. Differences were considered statistically significant at a P value <0.05.

Results

Induction of IL-6 protein in A549/basophil (BA) co-cultures

As shown in Figure 1A, no detectable levels of IL-6 were seen when basophils alone were stimulated with IL-3 or anti-IgE, even though our previous report had shown that these same supernatants contained relatively high levels of IL-4/IL-13 in response to these stimuli (1). In contrast, IL-6 was readily detected when basophils from the same subjects were co-cultured with A549. In fact, no co-stimulation was necessary to induce detectable IL-6. However, the levels of IL-6 were markedly enhanced ~10-fold with the addition of co-stimuli, in particular IL-3 (10 ng/ml) and to a lesser degree with anti-IgE (20 ng/ml). The lowest concentration of IL-3 tested, 0.1 ng/ml, significantly enhanced IL-6, with 10 ng/ml being optimal (Figure 1B). A time course showed that IL-6 became detectable at ~8h and was increased nearly 8-fold more at the time of harvest at 20h (Figure 1C). In addition, the IL-3 condition required fewer basophils (~6×103) compared to the nearly 10-fold greater number (~5×104) for co-stimulation with anti-IgE to obtain comparable IL-6 levels (Figure 1D).

Figure 1. Induction of IL-6 protein in basophils + A549 co-cultures vs. basophils alone:

Figure 1.

Open in a new tab

A, Protein levels secreted after 20h incubation when co-stimulated with medium, IL-3 (10 ng/ml), and anti-IgE antibody (20 ng/ml) (n=5). B, Effect of IL-3 dosing (n=5). C, Time course for IL-6 secretion in co-cultures co-stimulated as indicated (n=3), and D, numbers of basophils required to induce IL-6 when co-stimulated as indicated (representative exp.). *, p≤0.003; **, p<0.001 vs. corresponding conditions with basophils co-cultured alone. <LOD, denotes below level of detection.

Importantly, we have reported that human basophils co-cultured with BEAS-2B cells (a lung EC line) fail to produce IL-4/IL-13, as seen when co-cultured with A549 (1). However, basophils are reported to induce IL-6 from BEAS-2B when co-cultured in the presence of IL-17A (6). Therefore, we conducted experiments investigating whether IL-17A might act similarly in our A549/basophil co-cultures. This included testing whether basophils produce IL-17A when treated with IL-3, which might then act in an autocrine or paracrine fashion or whether IL-17A is detected in co-culture supernatant. However, we found no evidence to support either outcome (Supplemental Figure 3) and thus continued with testing the effects of IL-3 in the basophil/A549 co-culture model.

Requirement for cell-to-cell contact and the dependency of EC-Gal-3

In observing that the parameters central for IL-6 production in the co-cultures bear remarkable similarity to those identified for the induction of IL-4 and IL-13, we next investigated the importance of cell-to-cell contact and whether EC-Gal-3 is required. As previously reported, both proved essential for the IL-4 and IL-13 produced by basophils when co-cultured with A549 vs. separated using a transwell approach (1, 2). As shown in Figure 2A, basophils, when cultured alone, once again failed to secrete detectable amounts of IL-6 in response to IL-3 or medium alone. In contrast, IL-6 protein was detected after combining basophils (BA) with A549, with levels significantly higher only when the two cell types were allowed to come in direct contact with one another. Importantly, A549, when cultured alone under these conditions, were shown previously to secrete no to little IL-6 (7), and, indeed, this finding was further demonstrated in subsequent figures (see Figures. 3, 4 and 6 below). To substantiate the notion of cell-to-cell contact facilitating IL-6 production in the basophil/A549 co-cultures, additional control-type experiments were conducted yet proved mostly negative. For example, supernatant from A549 failed to support IL-6 production by basophils (Supplemental Figure 4A). Likewise, supernatant from activated basophils, including those primed with IL-3 and/or stimulated with known basophil stimuli (e.g. anti-IgE or FMLP), failed to induce comparable IL-6 secretion when added to A549 (Supplemental Figure 3B). This was true despite detecting histamine, IL-4, and IL-13 in the basophil supernatants (data not shown), and even after blocking the activity of the latter two mediators using dupilumab (Supplemental Figure 4 C,D).

Figure 2. IL-6 production in A549/basophil (BA) co-cultures is augmented with cell-to-cell contact and is augmented by EC-Gal-3.

Figure 2.

Open in a new tab

A, IL-6 protein in culture supernatants (20h) whereby BA and A549 EC were co-cultured together or separated in transwells in medium alone or IL-3 (10 ng/ml). *, p<0.02 when comparing IL-6 levels for cells cultured with IL-3 together vs. separated (mean±SEM, n=3,). B, IL-6 in supernatants (20h) from BA cultured alone, with A549 parent cells (A549-WT), Gal-3-deficient (1B10, 1E2) and Gal-3+ (3G8) clones in medium alone, IL-3 (10 ng/ml), or anti-IgE (20 ng/ml). Results are shown as Box-Whisker plots (n=8–9). ** p<0.001 vs. corresponding conditions with BA+A549-WT. <LOD, denotes below level of detection.

Figure 3. Cytokine production in basophil/A549 co-cultures is dependent on basophils expressing IgE.

Figure 3.

Open in a new tab

Culture-derived basophils (CDBA), obtained by culturing CD34+ precursors in IL-3 (10 ng/ml) for 12–14d, were portioned into those passively sensitized (16–18h) with IgE or those left unsensitized, as described in Materials & Methods. Both were then washed, counted, and then cultured, as indicated. Supernatants were harvested after 20h and assayed for the cytokines listed in panels A-D. Panel E shows the levels of these same cytokines in control cultures of A549 alone (n=4–9). * p<0.0005; ** p<0.0001 vs. corresponding conditions using CDBA not sensitized with IgE.

Figure 4. Matrix-bound IgE is incapable of inducing IL-6 when co-cultured with A549.

Figure 4.

Open in a new tab

Polystyrene MS (5 μM in size) were coated with JK IgE (MS-IgE) or with BSA (MS-BSA) as described in the Materials & Methods. The flow histogram (panel A) shows the staining for IgE on MS-IgE vs, MS-BSA. Three experiments (EXP) were conducted to test the capacity of each MS prep (1×105/culture) to induce IL-6 when added to A549 (panels B,C,D). Control cultures with A549 and basophils (BA) each alone and combined were included as indicated, with all cultures containing IL-3 (10 ng/ml). After 20h incubation, supernatants were assayed for IL-6 by ELISA. Dotted line denotes LOD for each EXP.

Figure 6. IL-3 priming of basophils facilitates their capacity to induce IL-6/VEGF from A549.

Figure 6.

Open in a new tab

Basophils (BA) were culture overnight (16–18h) in medium containing IL-3 (10 ng/ml). After washing, half of the primed (P) BA were portioned and immediately fixed with 4% buffered paraformaldehyde (P-BA-fixed), with the other portion left unfixed (P-BA). Counts were normalized for each suspension before co-culturing 1×105 with A549 or alone. Control cultures with A549 alone were also included, as indicated. Supernatants were collected after 20h incubation and assayed for the cytokines shown. Results are shown as Box-Whisker plots (n=4–5). * p<0.02, ** p<0.007.

We have previously reported the generation of two A549 clones (IE2 and 1B10) that demonstrated >95% suppression of total Gal-3 protein using shRNA, as determined by Western Blot (2). In contrast a 3rd clone, 3G8, showed ~50% reduction, compared to wild-type A549 (A549-WT). Use of these clones have demonstrated the importance of EC-Gal-3 in mediating the activation of basophils, monocytes and DC when co-cultured with A549 (2, 7). Therefore, we next determined the importance of EC-Gal-3 for the induction of IL-6. As shown in Figure 2B, basophils once again failed to secrete detectable levels of IL-6 in medium alone, with IL-3, or in response to anti-IgE. IL-6 protein was detected when basophils were co-cultured with A549-WT, with these levels augmented some 10- to 20-fold in the presence of IL-3. However, these augmented responses with IL-3 co-stimulation were significantly reduced when co-cultured with the Gal-3-deficient clones (1B10 > 1E2) in a nearly identical manner to that previously reported for IL-4 and IL-13 (2). In contrast, IL-6 protein was again detected when basophils were co-cultured with the Gal-3+ clone, 3G8, with levels nearly equal to those induced by A549-WT.

IL-6 was also produced when basophils were co-cultured with A549-WT in the presence of anti-IgE antibody relative to medium control (Figure 2B), although the levels averaged approximately 5-fold less than those in co-cultures receiving IL-3. As previously published, IL-4/IL-13 tracked similarly when using IL-3 co-stimulation (2). However, IL-6 levels in co-cultures receiving anti-IgE were significantly reduced when using clone 1B10, relative to wild-type A549. Predicting that IL-6 was basophil-derived, this latter observation was unexpected given that IL-4/IL-13 production in co-cultures was maintained/rescued in response to standard IgE-dependent activation with anti-IgE and not significantly affected when basophils were co-cultured with either of the Gal-3-deficient clones (2).

Cytokine production in co-cultures is dependent on basophils expressing IgE

Next, we investigated the requirement for basophil-bound IgE, with the hypothesis that it would be required in the same manner as previously shown for the IL-4/IL-13 produced in the co-cultures (1). To facilitate these experiments, we used culture-derived basophils (CDBA). CDBA are differentiated from CD34+ precursors, express FcεRIα, store histamine but are devoid of IgE expression, since propagated in serum-free medium. This latter feature allows passive sensitization with IgE for direct comparison of functional responses (such as cytokine secretion) with a portion of the same CDBA lacking IgE. Flow cytometric analysis shown in Supplemental Figure 5 demonstrates the expression of IgE following passive sensitization. As shown in Figure 3, IL-4 (panel A) and IL-13 (panel B) were monitored as positive controls and were produced only under stimulatory conditions that utilized CDBA passively sensitized with IgE and not by those left unsensitized. As expected, these cytokines were produced when stimulated with an activating anti-IgE antibody, and to an even greater extent after co-culturing with A549, but not with medium alone -all responses consistent with our previous findings (1). As hypothesized, IL-6 was also produced and almost exclusively in co-cultures to which passively sensitized CDBA were added. Some IL-6 was detected in co-cultures using unsensitized CDBA but at levels approximately 100-fold less. Unlike that observed for IL-4/IL-13, sensitized CDBA failed to produce measurable levels of IL-6 when stimulated with anti-IgE in the absence of A549 (Figure 3C). Finally, a nearly identical pattern was observed for VEGF-A (Figure 3D), which was also produced to a greater extent when using sensitized vs. unsensitized CDBA in co-cultures with A549. As observed for IL-6, VEGF-A was not produced by sensitized CDBA when stimulated with anti-IgE alone. Importantly, VEGF-A was clearly produced by A549 cells alone (Figure 3E), albeit at much lower levels than those detected in co-cultures with sensitized CDBA (Figure 3D). Control cultures with A549 alone (Figure 3E) showed no detectable amounts of IL-4 and IL-13, and very low amounts of IL-6 (~2–5 pg/ml) -findings consistent with our previous reports (1, 2, 7).

The findings above did not exclude the possibility that IgE might interact with A549-associated Gal-3 to directly induce IL-6 from these cells, even in the absence of basophils. However, this seems possible only if IgE was to be immobilized onto a solid phase, given that cell-to-cell interaction between basophils and A549 facilitated IL-6 production (Figure 2A). We therefore thought that the best chance of observing an IL-6 response was to couple IgE onto microspheres (MS) and add these to A549. As shown in Figure 4A, IgE was readily coated onto the MS (5 μm in size), as indicated by its detection using flow cytometry. However, these IgE-coated MS (MS-IgE) failed to induce IL-6 when added to A549 in 3 separate experiments (Figure 4B,C,D), further implicating the importance of adding basophils expressing IgE. Control MS coated with BSA (MS-BSA) likewise failed to induce a response. As expected, IL-6 was produced in basophil/A549 co-cultures simultaneously done as positive controls.

Basophils lack the capacity to secrete IL-6 to the extent seen for IL-4/IL-13

The observation that IgE-sensitized CDBA (Figure 3C) and peripheral blood basophils (Figures 1A, & 2B) did not produce IL-6 when stimulated alone with activating anti-IgE antibody was a surprise given the following: 1) IL-6 was induced in co-cultures with A549 -a mode of stimulation that we had previously shown to be dependent on FcεRI-mediated signaling, at least for the induction of IL-4/IL-13 secretion, and 2) there are several publications whereby human (and mouse) basophils and KU812 cells (a basophil-like cell line) produce IL-6, including upon IgE-dependent activation (8–12). We therefore thought it was important to further validate the capacity of human basophils to produce IL-6, contrasting this response with the IL-4 & IL-13 made by the same cells when cultured with well-known stimuli. Figure 5 shows data from experiments using highly enriched (>97%) basophils (n=18), which were stimulated with the indicated stimuli. Surprisingly, only 1 of the 18 preparations produced detectable levels of IL-6, when stimulated with anti-IgE alone (~15 pg/106 basophils), and this response was augmented to nearly 1 ng/106 with the addition of IL-3. However, IL-6 was undetected in the other 17 basophil preparations stimulated in this manner. Importantly, high levels of IL-6 (~1 ng/106) were detected in 3/3 basophil cultures stimulated with the calcium ionophore, ionomycin, thus indicating the potential to make this cytokine. The patterns and levels of IL-4/IL-13 detected in these same supernatants were as expected and quite comparable to what we and others have previously reported (13).

Figure 5. Human basophils do not secrete IL-6 to the same capacity seen for IL-4 & IL-13 when stimulated with IL-3 and/or anti-IgE.

Figure 5.

Open in a new tab

Basophil suspensions were cultured alone in medium containing: no stimulus, IL-3 (10 ng/ml), anti-IgE (20 ng/ml), the combination of IL-3/Anti-IgE, and ionomycin (500 ng/ml). Supernatants were collected after 20h incubation and assayed for the cytokines listed by ELISA. Paired values (n=3 to 18). * p<0.000001 vs. cells cultured in medium alone.

Basophils induce A549 to secrete IL-6 and VEGF-A

In observing that basophils only rarely secrete IL-6 when cultured alone, we conducted a final set of experiments to address whether they might acquire greater capacity do so when co-cultured with A549, or whether the A549 themselves are activated by basophils to secrete this cytokine. To address this possibility, we primed basophils overnight (16h) with IL-3, followed by washing, with a portion also immediately fixed with 4% buffered paraformaldehyde (PF) before adding to A549 co-cultures. As shown in Figure 6A, IL-3-primed basophils immediately fixed with PF (P-BA-fixed) markedly induced IL-6 when co-cultured with A549. In fact, the levels of IL-6 were significantly higher than those induced when using live primed basophils (P-BA) put back into culture with A549 without IL-3. The fixing procedure with PF alone did not account for the IL-6 responses, since other experiments showed that unprimed/fixed basophils did not induce measurable levels of this cytokine compared to those primed/fixed (Supplemental Figure 6). Moreover, adding IL-3 back to unfixed P-BA resulted in IL-6 levels comparable to those induced by P-BA-fixed. Once again, IL-6 was not detected in cultures of basophils alone (unfixed or fixed) and only marginally detected in the A549 cultures. A nearly identical pattern was observed for VEGF-A (Figure 6B), clearly indicating that basophils did not secrete these two cytokines in the co-cultures but instead provoked their production from A549. In contrast, the production of IL-4 (Figure 6C) and IL-13 (Figure 6D) in these same supernatants was clearly derived from basophils, as expected, given that neither cytokine was detected in cultures that contained fixed basophils.

Discussion

The findings presented herein extend on the concept that EC-Gal-3/IgE interactions play a unique role in activating basophils. However, in addition to provoking responses directly from IgE-expressing basophils, we now report reciprocal responses that also occur in the cells expressing Gal-3, in this instance, A549. Again, we have previously shown that basophils produce IL-4/IL-13 when co-cultured with A549, in a reaction requiring: 1) cell-to-cell contact, 2) expression of IgE by basophils, and 3) expression of Gal-3 by A549. Selective inhibitors of IgE/FcεRI signaling inhibited this response, which further supported the role of this pathway. We now show evidence that many of these same parameters are also critical for other cytokines (e.g. IL-6 and VEGF-A) that we have since identified as being produced in our co-culture model. However, we clearly reveal that the production of these latter cytokines is from the A549 cells themselves and that IgE-expressing basophils function as a stimulus for this response, rather than directly secreting these cytokines. We demonstrate this by conducting straightforward experiments using paraformaldehyde (PF)-fixed basophils. Naturally, cells fixed in this manner are dead and incapable of actively secreting cytokines. Indeed, this was very much evident by the fact that we could no longer detect IL-4/IL-13 in the co-cultures when using fixed basophils (Figure 6). However, these fixed cells remained structurally intact and continued to provide the necessary signals to support increased production of IL-6 and VEGF-A from A549 cells. In using live cells, we show that the production of IL-6 and VEGF-A, like that of IL-4/IL-13, require that basophils express IgE, and that A549 bear Gal-3. We initially presumed that basophils were the source of IL-6 and VEGF-A, given evidence in the literature that these cells have the capacity to produce both cytokines, especially via IgE-dependent activation, using human (8,14) and mouse (8–12, 14). However, when unable to confirm these reports, we addressed the possibility that basophils are not the source, but rather they provoke A549 cells to produce IL-6 and VEGF.

In fact, clues at the onset of this study directed our thoughts that basophils may not be the source of IL-6 in the co-cultures. First, basophils cultured alone with IL-3 or anti-IgE did not secrete detectable IL-6 (Figures 1, 2A). Second, and more intriguing, IL-6 levels were markedly reduced when basophils were co-cultured with the Gal-3-deficient clones (1B10 > 1E2), as observed in Figure 2B. Certainly, this was expected in co-cultures also receiving IL-3, which is critical for EC-Gal-3 responses. However, IL-6 levels were reduced in co-cultures receiving anti-IgE especially when using clone 1B10 -a finding unlike what we had previously observed for IL-4/IL-13 where levels of these cytokines were maintained (2). If truly basophil-derived, we reasoned that IL-6 should track like IL-4 and IL-13 and be induced upon co-stimulation with standard IgE-dependent activation (i.e. w/ anti-IgE), even when EC-Gal-3 activation is impaired using Gal-3-deficient A549 clones. Moreover, basophils themselves can produce IL-3 when activated by anti-IgE, with this response mediating autocrine activity (15). This potentially accounted for why basophils co-cultured with A549-WT plus anti-IgE resulted in higher levels of IL-6 compared to co-cultures in medium alone (Figure 2B). If producing IL-3 in response to anti-IgE, basophils can potentially augment EC-Gal-3 responses. In contrast, if A549 are the source of IL-6 resulting from EC-Gal-3 activation, then clones lacking Gal-3 should be less capable of producing IL-6. This is, in fact, what was observed, especially when using clone 1B10. This was less evident using the 1E2 clone, likely due to it spontaneously leaking more IL-6 than 1B10 or A549-WT, which we later discovered (data not shown).

Importantly, Wong et al. reported more than a decade ago that basophils have the capacity to induce IL-6 from bronchial EC/BEAS-2B, including evidence that the two cells directly interact with one another in co-culture to induce this response (6). Thus, while our findings reported herein bear similarity to this publication, there are several significant distinctions to note. First, many of the experiments conducted in the earlier study made use of KU812 cells -a basophil-like cell line, rather than using primary blood basophils. Second, we stress herein the importance of using IL-3-primed basophils to achieve enhanced production of IL-6, whereas the work by Wong et al. reported the role of IL-17A and how it impacted both basophils and BEAS-2B. We investigated whether IL-17A is produced by basophils and/or is present in our co-culture model but saw no evidence that this cytokine accounts for the IL-6 produced (Supplemental Figure 3). Most significantly, our findings clearly point to a molecular mechanism involving IgE and EC-Gal-3, with both being critical for the IL-6/VEGF-A secreted by A549, but also for the IL-4/IL-13 produced by basophils. In the study by Wong et al., adhesion molecules were suggested to play a role in IL-6 produced by BEAS-2B, but direct evidence was not provided (6). It’s reasonable to conclude that the differences between the studies could be due to the different bronchial cell lines employed. For example, we previously tested BEAS-2B for their capacity to activate basophils, finding that they did not induce IL-4/IL-13 (1). We additionally found that BEAS-2B express no to little Gal-3, which presumably accounted for their inability to induce IL-4/IL-13. Whether any of the other parameters identified in our study using A549 (e.g. IgE) might also prove critical to the production of IL-6 by BEAS-2B remain unknown but seem possible.

While IgE (on basophils) and Gal-3 (on A549) have both proved essential for the cytokines produced in our co-culture model (i.e. IL-4/IL-13 from basophils and IL-6/VEGF-A from A549), the overall levels of each are markedly augmented by IL-3. This effect of IL-3 impacts responses by directly affecting basophils (and not A549). Indeed, IL-3 is well known to mediate priming and stimulatory activity in mature basophils in addition to promoting basophil development (16). Most surprisingly, basophils primed overnight (16–18h) with IL-3, washed, and then immediately fixed with PF provoked greater IL-6/VEGF-A responses than those primed, washed, and left unfixed/alive (Figure 6). Certainly, the significance of this outcome requires further elucidation, although the use of fixed cells rules-out any impact IL-3 mediates on signal transduction or in the induction of secreted cytokines/factors contributing to the response. Instead, it strongly signifies that a yet unidentified protein/factor is induced on the surface of basophils, which plays a role in augmenting the IgE/Gal-3-dependent production of IL-6/VEGF-A from A549, and possibly that of IL-4/IL13 from basophils. Moreover, since adding IL-3 back to the primed/washed cells restored the capacity of live basophils to induce these cytokines (Figure 6), then expression of this protein/factor is likely dependent on constant and/or prolonged exposure to IL-3. By fixing basophils immediately after IL-3 priming, levels of this unknown protein/factor are likely captured at peak expression, which allowed these fixed cells to optimally induce IL-6/VEGF-A. Numerous attempts to identify this putative protein/factor using antibodies targeting known proteins induced on basophils with IL-3 priming have thus far proved negative, perhaps requiring additional approaches (JTS, unpublished data).

Again, we could not demonstrate predictable IL-6 and VEGF-A production by human basophils cultured alone, despite several publications in both mouse and human models, indicating that basophils produce these cytokines (6, 8–12, 14, 17). This was particularly true for IL-6 where we investigated up to 18 additional basophil preparations (97 to >99% pure) for production of this cytokine (Figure 5). Results showed that only 1 of 18 basophil preps investigated produced detectable IL-6, and most significantly when using the combination of IL-3 and anti-IgE antibody. It is unknown whether this subject whose basophils produced IL-6 suffers from some clinical disease/condition that might account for this response as all specimens are de-identified and anonymous. However, basophil unresponsiveness cannot be claimed for the other preparations, given that most/all produced IL-4 and IL-13, as expected, to IL-3 and/or anti-IgE dependent activation. Interestingly, IL-6 was secreted in 3/3 basophil preparations when stimulated with calcium ionophore (ionomycin), thus indicating a capacity to produce this cytokine. Nonetheless, the literature is lacking in detail regarding the exact conditions/parameters and donor basophils most capable of producing IL-6, thus requiring further investigation. Likewise, we were unable to demonstrate VEGF-A production by basophils cultured alone, even when stimulating using anti-IgE (Figure 3D). Basophils are reported to rapidly secrete (within 4h) this cytokine (14, 17), which might account for our inability to detect it in 20h cultures. Like IL-4, which generally peaks at 4–6h with standard IgE activation, the levels of VEGF-A may be low, even undetectable, after prolonged (20h) culture. Regardless, the peak VEGF-A levels previously reported by basophils after 4h incubation (up to ~60 pg/106) (14), is far less (10- to 60-fold) than what was produce by A549 in the co-culture experiments conducted herein (Figure 3D). This finding raises the likelihood that greater levels of VEGF-A (and IL-6) seem possible when basophils induce these cytokines from cells more capable of producing them, rather than being a source themselves. Consequently, our VEGF-A findings (like those for IL-6) also prompt the need for more work to address the capacity of human basophils to produce this growth factor.

The findings presented herein raise a unique hypothesis whereby the IgE bound to basophils has the potential to act as a ligand in stimulating Gal-3-expressing cells. In this instance, to secrete IL-6 and VEGF-A, but possibly other pro-angiogenic cytokines not yet assayed. Importantly, IgE bound to IL-3-primed basophils was required for the IL-6/VEGF-A responses, since IgE coated MS (MS-IgE) did not induce these cytokines from A549. (Figure 4). This finding indicates that IgE alone is insufficient to induce these cytokines from A549, further stressing the importance of the unknown putative protein induced on basophils with IL-3 in facilitating the overall response. Our previous work had already shown, and is confirmed again herein, that the IgE/EC-Gal-3 interaction has functional consequences that are more common with IgE/FcεRI crosslinking in inducing basophils to produce IL-4/IL-13. Of course, we provide no direct evidence that IgE and Gal-3 interact with one another to induce these responses yet work done nearly 35 years ago provided solid evidence that Gal-3 does, indeed, bind IgE, especially when this immunoglobulin is desialylated (18, 19). In fact, this early work prompted belief that Gal-3 may very well play a role in the pathogenesis of allergic disease, given its association with IgE. However, solid evidence supporting this concept has not been forthcoming.

Thus, while it remains possible that our findings have relevance to allergic disease, given the involvement of an IgE component, there is plenty accompanying evidence to suggest a greater role in wound healing and/or cancer. Again, a plethora of evidence during the past two decades has since linked Gal-3 to unsuspecting diseases ranging from cardiovascular disease (CVD), cancer, autoimmunity, and fibrosis in general. In fact, Gal-3 serum levels are sometimes used diagnostically as a marker of heart failure progression and for various cancers. Gal-3 involvement in wound healing is also well documented (20–22). Moreover, there is now growing evidence that IL-4/IL-13 play a role, where these cytokines are well known to act on monocyte/macrophages to promote the M2 phenotype -a feature common both to cancer and wound healing. In fact, we showed nearly a decade ago that basophil-derived IL-4/IL-13 promotes M2-like features when basophils are co-cultured with monocytes in vitro (23). Mouse models also report that basophil-derived IL-4/IL-13 is instrumental in promoting M2 development in vivo (24). Similarly, the associations of IL-6 and VEGF-A in cancer are well known, where both are recognized as potent cytokines promoting tumor growth (25).

Therefore, it seems logical to propose that the basophil IgE/EC-Gal-3 axis, as described herein, constitutes a potential mechanism for triggering the cytokines that promote wound healing, but also for the remarkably similar responses linked to tumorigenesis observed in some cancers. In fact, several recent observations are consistent with this hypothesis. For example, mouse basophils were recently shown to be critical in promoting the wound-healing observed following experimental myocardial infarction (26). They did so by acutely producing IL-4/IL-13 needed for the development of M2-like myeloid cells that infiltrated the heart and promoted wound-healing. Although involvement of Gal-3 was not directly investigated in this study, this lectin, as noted above, has been identified as a key component in other models of wound-healing, especially in the heart (21, 22). Likewise, there is emerging evidence that basophils, and their capacity to produce IL-4/IL-13, can also mediate an unfavorable outcome in certain cancers. For example, a mouse model of pancreatic cancer showed evidence that IL-4-producing basophils within the TME help promote Th2-like responses that associated with greater tumor burden (27). The same study showed correlative evidence that human pancreatic cancer patients experienced poorer outcomes the greater the frequency of IL-4-producing basophils detected in their biopsies. Again, this earlier study presented no evidence that tumor-associated Gal-3 stimulated basophils to produce IL-4/IL-13, yet it is well-documented that Gal-3 is a biomarker in pancreatic cancer (28). Finally, the most compelling association comes from a recent study modeling non-small cell lung cancer (NSCLC) in mice, where basophils, and their capacity to produce IL-4/IL-13, were shown essential for promoting tumorigenesis (29). Selective depleting basophils in these mice profoundly reduced the tumor burden and normalized myelopoiesis/M2 development. The clinical significance of these findings was further underscored in the same paper with preliminary data that dupilumab (an IL-4Ra receptor blocker of IL-4/IL-13 binding), when given in conjunction with PD-1/PD-L1 blockade, reduced M2-like development in human NSCLC patients (29). In doing so, more CD8 cytotoxic cells infiltrated the TME, with an improved capacity to target the tumor. Of course, the actual stimulus (or stimuli) responsible for inducing basophils to produce IL-4/IL-13 in this study (like those above) remained mostly unanswered. However, it is our belief that the basophil IgE/EC-Gal-3 interaction, as described herein, constitutes a potential mechanism for triggering the cytokine responses (IL-4/IL-13) that help drive M2 development but also those (e.g. IL-6/VEGF) that further promote tumorigenesis within the TME. Overall, these findings should provide valuable insight into the emerging and compelling interplay between basophils, Gal-3/IgE, IL-4/IL-13, and M2 cell development, and how these collectively function both in health and disease.

Supplementary Material

1

Acknowledgements:

The authors wish to acknowledge colleagues Dr. Robert G. Hamilton in providing the JK IgE used in some of the studies and Dr. Jody Tversky in providing dupilumab.

Funding:

Supported, in part, by Public Health Services Research Grant AI115703 to JTS from the National Institute of Allergy and Infectious Diseases, National Institutes of Health (NIAID, NIH)

Abbreviations:

EC

epithelial cell

PF

paraformaldehyde

Gal-3

Galectin-3

C-IMDM

conditioned Iscove’s Modified Dulbecco’s Medium

PAG, PIPES

albumin, glucose

BDC

basophil-depleted cell

BEC

basophil-enriched cell

CDBA

culture-derived basophils

MS

microspheres

Footnotes

Authorship Contribution:

JS conceived the study, helped conduct experiments and wrote the manuscript. LE and AB provided input regarding experimental design and conducted many of the experiments. All authors contributed to manuscript revision, read and approved the submitted version.

Conflicts of Interest: None

References

  • 1.Schroeder JT, Bieneman AP. Activation of human basophils by A549 lung epithelial cells reveals a novel IgE-dependent response independent of allergen. J Immunol. 2017. −08-01; 199(3): 855–865. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Schroeder JT, Adeosun AA, Do D, Bieneman AP. Galectin-3 is essential for IgE-dependent activation of human basophils by A549 lung epithelial cells. J Allergy Clin Immunol. 2019. −07; 144(1): 312–315.e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Poto R, Gambardella AR, Marone G, Schroeder JT, Mattei F, Schiavoni G, Varricchi G. Basophils from allergy to cancer. Front Immunol. 2022; 13: 1056838. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Schroeder JT, Bieneman AP. Isolation of human basophils. Curr Protoc Immunol. 2016. −02-02; 112: 7.24.1–7.24.8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Langdon JM, Schroeder JT, Vonakis BM, Bieneman AP, Chichester K, Macdonald SM. Histamine-releasing factor/translationally controlled tumor protein (HRF/TCTP)-induced histamine release is enhanced with SHIP-1 knockdown in cultured human mast cell and basophil models. J Leukoc Biol. 2008-10; 84(4): 1151–1158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Wong CK, Cao J, Yin YB, Lam CWK. Interleukin-17A activation on bronchial epithelium and basophils: A novel inflammatory mechanism. Eur Respir J. 2010. −04; 35(4): 883–893. [DOI] [PubMed] [Google Scholar]
  • 7.Schroeder JT, Adeosun AA, Bieneman AP. Epithelial cell-associated galectin-3 activates human dendritic cell subtypes for pro-inflammatory cytokines. Front Immunol. 2020; 11: 524826. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Galeotti C, Stephen-Victor E, Karnam A, Das M, Gilardin L, Maddur MS, Wymann S, Vonarburg C, Chevailler A, Dimitrov JD, Benveniste O, Bruhns P, Kaveri SV, Bayry J. Intravenous immunoglobulin induces IL-4 in human basophils by signaling through surface-bound IgE. J Allergy Clin Immunol. 2019. −08; 144(2): 524–535.e8. [DOI] [PubMed] [Google Scholar]
  • 9.Krüger-Krasagakes S, Möller A, Kolde G, Lippert U, Weber M, Henz BM. Production of interleukin-6 by human mast cells and basophilic cells. J Invest Dermatol. 1996. −01; 106(1): 75–79. [DOI] [PubMed] [Google Scholar]
  • 10.Lin H, Del Rio Castillo AE, González VJ, Jacquemin L, Panda JK, Bonaccorso F, Vázquez E, Bianco A. Effects of industrially produced 2-dimensional molybdenum disulfide materials in primary human basophils. NanoImpact. 2023. −01; 29: 100451. [DOI] [PubMed] [Google Scholar]
  • 11.Yuk CM, Park HJ, Kwon B, Lah SJ, Chang J, Kim J, Lee K, Park S, Hong S, Lee S. Basophil-derived IL-6 regulates TH17 cell differentiation and CD4 T cell immunity. Sci Rep. 2017. −01-30; 7: 41744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Doke T, Abedini A, Aldridge DL, Yang Y, Park J, Hernandez CM, Balzer MS, Shrestra R, Coppock G, Rico JMI, Han SY, Kim J, Xin S, Piliponsky AM, Angelozzi M, Lefebvre V, Siracusa MC, Hunter CA, Susztak K. Single-cell analysis identifies the interaction of altered renal tubules with basophils orchestrating kidney fibrosis. Nat Immunol. 2022. −06; 23(6): 947–959. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Schroeder JT. Basophils: Emerging roles in the pathogenesis of allergic disease. Immunol Rev. 2011. −07; 242(1): 144–160. [DOI] [PubMed] [Google Scholar]
  • 14.de Paulis A, Prevete N, Fiorentino I, Rossi FW, Staibano S, Montuori N, Ragno P, Longobardi A, Liccardo B, Genovese A, Ribatti D, Walls AF, Marone G. Expression and functions of the vascular endothelial growth factors and their receptors in human basophils. J Immunol. 2006. −11-15; 177(10): 7322–7331. [DOI] [PubMed] [Google Scholar]
  • 15.Schroeder JT, Chichester KL, Bieneman AP. Human basophils secrete IL-3: Evidence of autocrine priming for phenotypic and functional responses in allergic disease. J Immunol. 2009. −02-15; 182(4): 2432–2438. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Varricchi G, Poto R, Marone G, Schroeder JT. IL-3 in the development and function of basophils. Semin Immunol. 2021. −04; 54: 101510. [DOI] [PubMed] [Google Scholar]
  • 17.Marone G, Varricchi G, Loffredo S, Granata F. Mast cells and basophils in inflammatory and tumor angiogenesis and lymphangiogenesis. Eur J Pharmacol. 2016. −05-05; 778: 146–151. [DOI] [PubMed] [Google Scholar]
  • 18.Hsu DK, Zuberi RI, Liu FT. Biochemical and biophysical characterization of human recombinant IgE-binding protein, an S-type animal lectin. J Biol Chem. 1992. −07-15; 267(20): 14167–14174. [PubMed] [Google Scholar]
  • 19.Robertson MW, Albrandt K, Keller D, Liu FT. Human IgE-binding protein: A soluble lectin exhibiting a highly conserved interspecies sequence and differential recognition of IgE glycoforms. Biochemistry. 1990. −09-04; 29(35): 8093–8100. [DOI] [PubMed] [Google Scholar]
  • 20.Sciacchitano S, Lavra L, Morgante A, Ulivieri A, Magi F, De Francesco GP, Bellotti C, Salehi LB, Ricci A. Galectin-3: One molecule for an alphabet of diseases, from A to Z. Int J Mol Sci. 2018. −01-26; 19(2): 379. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.McLeod K, Walker JT, Hamilton DW. Galectin-3 regulation of wound healing and fibrotic processes: Insights for chronic skin wound therapeutics. J Cell Commun Signal. 2018. −03; 12(1): 281–287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.González GE, Cassaglia P, Noli Truant S, Fernández MM, Wilensky L, Volberg V, Malchiodi EL, Morales C, Gelpi RJ. Galectin-3 is essential for early wound healing and ventricular remodeling after myocardial infarction in mice. Int J Cardiol. 2014. −10-20; 176(3): 1423–1425. [DOI] [PubMed] [Google Scholar]
  • 23.Borriello F, Longo M, Spinelli R, Pecoraro A, Granata F, Staiano RI, Loffredo S, Spadaro G, Beguinot F, Schroeder J, Marone G. IL-3 synergises with basophil-derived IL-4 and IL-13 to promote the alternative activation of human monocytes. Eur J Immunol. 2015. −07; 45(7): 2042–2051. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Egawa M, Mukai K, Yoshikawa S, Iki M, Mukaida N, Kawano Y, Minegishi Y, Karasuyama H. Inflammatory monocytes recruited to allergic skin acquire an anti-inflammatory M2 phenotype via basophil-derived interleukin-4. Immunity. 2013. −03-21; 38(3): 570–580. [DOI] [PubMed] [Google Scholar]
  • 25.Abolfathi H, Sheikhpour M, Shahraeini SS, Khatami S, Nojoumi SA. Studies in lung cancer cytokine proteomics: A review. Expert Rev Proteomics. 2021. −01; 18(1): 49–64. [DOI] [PubMed] [Google Scholar]
  • 26.Sicklinger F, Meyer IS, Li X, Radtke D, Dicks S, Kornadt MP, Mertens C, Meier JK, Lavine KJ, Zhang Y, Kuhn TC, Terzer T, Patel J, Boerries M, Schramm G, Frey N, Katus HA, Voehringer D, Leuschner F. Basophils balance healing after myocardial infarction via IL-4/IL-13. J Clin Invest. 2021. −07-01; 131(13): e136778. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.De Monte L, Wörmann S, Brunetto E, Heltai S, Magliacane G, Reni M, Paganoni AM, Recalde H, Mondino A, Falconi M, Aleotti F, Balzano G, Algül H, Doglioni C, Protti MP. Basophil recruitment into tumor-draining lymph nodes correlates with Th2 inflammation and reduced survival in pancreatic cancer patients. Cancer Res. 2016. −04-01; 76(7): 1792–1803. [DOI] [PubMed] [Google Scholar]
  • 28.Aderinto N, Abdulbasit MO, Olatunji D, Edun M. Unveiling the potential of galectin-3 as a diagnostic biomarker for pancreatic cancer: A review. Ann Med Surg (Lond). 2023. Oct 2; 85(11): 5557–5567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.LaMarche NM, Hegde S, Park MD, Maier BB, Troncoso L, Le Berichel J, Hamon P, Belabed M, Mattiuz R, Hennequin C, Chin T, Reid AM, Reyes-Torres I, Nemeth E, Zhang R, Olson OC, Doroshow DB, Rohs NC, Gomez JE, Veluswamy R, Hall N, Venturini N, Ginhoux F, Liu Z, Buckup M, Figueiredo I, Roudko V, Miyake K, Karasuyama H, Gonzalez-Kozlova E, Gnjatic S, Passegué E, Kim-Schulze S, Brown BD, Hirsch FR, Kim BS, Marron TU, Merad M. An IL-4 signalling axis in bone marrow drives pro-tumorigenic myelopoiesis. Nature. 2023. Dec 6. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2082682-supplement-1.pdf (876.2KB, pdf)

RESOURCES