Eosinophil-expressed galectin-3 regulates cell trafficking and migration

Abstract

Galectin-3 (Gal-3), a β galactoside-binding lectin, is implicated in the pathogenesis of allergic airway inflammation and allergen-challenged mice deficient in Gal-3 (Gal-3(-/-)) exhibit decreased airway recruitment of eosinophils (Eos). Gal-3 is expressed and secreted by several cell types and can thus function extracellularly and intracellularly to regulate a variety of cellular responses. We sought to determine the role of Eos-expressed Gal-3 in promoting Eos trafficking and migration in the context of allergic airway inflammation using bone marrow (BM)-derived Eos from wild-type (WT) and Gal-3(-/-) mice. Airway recruitment of Eos in acute (4 weeks) and chronic (8-12 weeks) allergen-challenged WT mice correlated with Gal-3 expression in the lungs. BM-derived Eos were found to express Gal-3 on the cell surface and secrete soluble Gal-3 when exposed to eotaxin-1. Compared to WT Eos, Gal-3(-/-) Eos exhibited significantly reduced rolling on vascular cell adhesion molecule 1 (VCAM-1) and decreased stable adhesion on intercellular adhesion molecule 1 (ICAM-1) under conditions of flow in vitro. Evaluation of cytoskeletal rearrangement demonstrated that relatively fewer adherent Gal-3(-/-) Eos undergo cell spreading and formation of membrane protrusions. In addition, cell surface expression of integrin receptor αM (CD11b) was lower in Gal-3(-/-) Eos, which is likely to account for their altered adhesive interactions with VCAM-1 and ICAM-1. Gal-3(-/-) Eos also exhibited significantly decreased migration toward eotaxin-1 compared to WT Eos irrespective of similar levels of CCR3 expression. Further, eotaxin-induced migration of WT Eos remained unaffected in the presence of lactose, suggesting a role for intracellular Gal-3 in regulating Eos migration. Overall, our findings indicate that Gal-3 expression in the lungs correlates with Eos mobilization during allergic airway inflammation and signaling involving intracellular Gal-3 and/or secreted Gal-3 bound to the cell surface of Eos appears to be essential for Eos trafficking under flow as well as for migration.

Keywords: allergic airway inflammation; cell trafficking; eosinophils; galectin-3; migration.

FIGURE 1

Airway eosinophilia in allergen-challenged mice is associated with elevated Gal-3 expression. (A) Eos in BALF of WT C57BL/6 mice after allergen challenge for 4, 8, or 12 weeks (n = 5 mice/group). (B) Gal-3 levels in BALF of allergen-challenged and control mice by ELISA (top, n = 6 mice/group) as well as Western blot analysis (bottom). Representative results for two mice from 4 and 12 week allergen-challenged groups along with corresponding controls (n = 3–4 mice/group) are shown. (C) Quantitation of Gal-3 expression in lung tissue of allergen-challenged and control mice by densitometric analysis of Western blots. Representative results for each group are shown below (n = 3–4 mice/group). Data represent mean ± SEM. *p <0.01 in (A) and (C) and <0.03 in (B) versus control mice; ^#p <0.01 versus 4 week allergen-challenged mice in (A).

FIGURE 2

Expression and release of Gal-3 by BM-derived mouse Eos. (A) Cell surface expression of Gal-3 in BM-derived Eos from WT and Gal-3^-/- mice by flow cytometry with rabbit polyclonal antibody against Gal-3 and rabbit IgG as negative control. (B) Cell surface expression of Gal-3 in non-permeabilized BM-derived Eos by confocal microscopy using monoclonal antibodies against Gal-3. Representative images for isotype control (mouse IgG, top), WT Eos (middle), and Gal-3^-/- Eos (bottom) are shown at a magnification of 600×. (C) Gal-3 expression in WT and Gal-3^-/- Eos lysates by Western blot analysis using rabbit polyclonal antibodies against Gal-3. (D) Gal-3 in culture supernatant of Eos incubated with eotaxin-1 (100 nM) or media alone for 30 min or 6 h by Western blot analysis followed by densitometry (Mean ± SEM). *p <0.05 versus Eos cultured without eotaxin-1 in (D). Representative data of two to four independent experiments in (A–C) and of three independent experiments in bottom panel of (D) performed with BM-derived Eos from different mice is shown.

FIGURE 3

Gal-3 deficient Eos exhibit decreased trafficking and altered cell morphology after adhesion. (A) Rolling and (B) adhesion of BM-derived Eos from WT and Gal-3^-/- mice on rmVCAM-1- and rmICAM-1-coated coverslips under conditions of flow (~2.0 dynes/cm²) in an in vitro flow chamber. (C) BM-derived Eos from WT and Gal-3^-/- mice adherent on rmVCAM-1-coated coverslips were stained with Alexa Fluor 488 phalloidin and DAPI and examined by confocal microscopy. Adherent cells in five randomly selected fields of each coverslip were counted and the number of cells that exhibited spreading and membrane protrusions from a round cell body was identified and expressed as a percentage of the total number of adhered cells in that field. Results presented are relative to WT Eos. Combined data (mean ± SEM) of two to three independent experiments is shown in (A–C). *p <0.05 versus PBS-coated coverslips; ^#p <0.05 versus WT Eos.

FIGURE 4

Gal-3 deficient Eos express decreased levels of αM. Expression of CD11b (αM), CD11a (αL), L-selectin, and CD49d (α4) on the cell surface of WT and Gal-3^-/- Eos was examined by flow cytometry using rat mAbs against CD11b, CD11a, CD49d, and L-selectin, respectively, followed by FITC-conjugated goat anti-rat IgG. Rat IgG2b (for CD11b and CD49d) and rat IgG2a (for L-selectin and CD11a) were used as isotype-matched control antibodies. Representative data of two to five independent experiments with Eos from different mice is shown.

FIGURE 5

Gal-3 deficient Eos exhibit decreased migration toward eotaxin-1. (A) Migration of WT and Gal-3^-/- BM-derived Eos toward murine eotaxin-1 (100 nM) in vitro using 96-well Transwell ^® Chambers. Results are represented as percent cell migration relative to WT Eos. (B) Expression of CCR3 by WT and Gal-3^-/- Eos by flow cytometry using FITC-conjugated rat anti-mouse CCR3 with FITC-conjugated rat IgG2a as isotype control. Representative data of two independent experiments with Eos from different mice is shown. (C) Migration of WT Eos suspended in medium alone or medium containing lactose or maltose toward murine eotaxin-1. Number of cells that migrated in each case was determined and expressed as the average number of cells/field. Combined data (mean ± SEM) from three independent experiments in duplicate is shown in (A) and (C). ^#p <0.05 versus WT Eos in (A).