C3a signaling is not involved in eosinophil migration during experimental allergic lung inflammation in mice

To the Editor,

Eosinophilia of the lungs and airways is a hallmark feature in allergic asthma. Eosinophils can exert proinflammatory and epithelial‐damaging effects.1 New therapies in the field of asthma targeting eosinophils have shown benefit in patients with high blood and sputum eosinophilia, including fewer exacerbation and improved lung function,2 thereby pointing to the potential importance of therapeutics that interfere with eosinophil recruitment in the treatment of asthma. In recent years, data from preclinical studies unveiled an important role for anaphylatoxins, components of complement system activation, in the pathogenesis of allergic airway disease.3 Signaling of the anaphylatoxin C3a through its receptor (C3aR) was documented to promote the onset of Th2 responses in different allergen‐induced asthma models. Deficiency in or pharmacological blocking of either C3a or C3aR attenuated the allergen‐induced Th2 response, which included a marked reduction of the eosinophilia in lung and airways.3 While many studies focused on the interaction between C3a signaling and cells of the adaptive immune system during allergic inflammation,4 only few investigated a potential direct effect of C3a signaling on eosinophil function. Infusion of C3a induced eosinophil adherence to postcapillary venules of IL‐1β‐stimulated mesenteric blood vessels of rabbits, but did not influence subsequent transmigration of eosinophils, suggesting a selective effect of C3a on eosinophil adhesion.5 It remains elusive if C3a signaling in eosinophils affects their migration to lungs during an allergic response. This study aimed to investigate the role of C3a signaling in eosinophils in their recruitment to the airways during allergic lung inflammation.

We used an established mouse model of airway sensitization and challenge with the clinically relevant allergen house dust mite (HDM) (for details see Figure S1 and Appendix S1). HDM challenge elicited similar increases (P = .94) in C3a concentrations in bronchoalveolar lavage (BAL) fluid from HDM‐sensitized WT and eosinophil lineage‐deficient ΔdblGATA KO mice (Figure 1A). As expected, HDM challenge induced eosinophilia in the lungs and airways of WT mice whereas ΔdblGATA KO mice showed a complete deficit in eosinophils in both lung tissue and BALF (Figure 1B). Next, we harvested bone marrow cells from wild‐type (WT) and C3aR knockout (KO) mice and differentiated these into mature bone marrow‐derived eosinophils (bmEos) ex vivo (Figure S2A) as described in detail in the Appendix S1. We confirmed that C3aR mRNA was present in WT bmEos but completely abrogated in C3aR KO bmEos after 14 days in culture (Figure S2B). Total cell numbers grew similarly from WT and C3aR KO bone marrow to approximately 30 × 10⁶ cells on day 14 (P = .91)( Figure S2C). Over a time span of 14 days, a high purity of eosinophils, defined as CCR3 and Siglec‐F double‐positive cells, was achieved from WT and C3aR KO mice (>95% at day 14, P = .92; Figure 1C). Giemsa staining showed the stereotypical bilobed nucleus and eosinophilic granules in eosinophils from both WT and C3aR KO (Figure S2D) cultures.6 Moreover, circular nuclear morphologies, previously described in mouse eosinophils from peripheral blood were also observed.6 Together, these data indicate that signaling through C3aR is not essential for the proliferation and differentiation of eosinophils.

Figure 1.

Generation of bone marrow‐derived eosinophils ex vivo and adoptive transfer in HDM‐challenged eosinophil‐deficient mice. A, BALF C3a in WT and ΔdblGATA KO mice 24 h after the last saline or HDM challenge. B, Identification of bone marrow‐derived eosinophils in the lung by flow cytometry in WT and ΔdblGATA KO mice 24 h following last HDM challenge. C, Identification of eosinophils by double‐positive staining for CCR3 and Siglec‐F in response to IL‐5 on each indicated time point. Percentage of eosinophils following incubation with IL‐5 from day 4 till 14. Number of eosinophils in the (D) lung or (E) BALF of recipient (ΔdblGATA KO) mice following adaptive transfer of either WT or C3aR KO bmEos. A parametric t test was used for the comparison between groups. Data are representatives of at least two independent experiments and expressed as means ± SEM from 6 to 8 separate cultures or mice per group. ***P < .001

Eosinophils are not required for the generation of memory T cells during the sensitization phase.7 Thus, we adoptively transferred 5 × 10⁶ WT or C3aR KO bmEos (intravenously) 24 hours after the first challenge (on day 15) into ΔdblGATA KO mice. Previously, bmEos were shown to have a half‐life of eight days following chemotaxis into the lung,8 enabling the investigation of migrated eosinophils after completing the challenge phase in our HDM model. Twenty‐four hours after the last challenge (on day 20), saline‐challenged mice demonstrated a lack of bmEos accumulation in both lung tissue and BAL fluid (data not shown). HDM‐challenged mice showed similar numbers of WT and C3aR KO bmEos in their lungs (P = .53) (Figure 1D) and BAL fluid (P = .94) (Figure 1E). Together, these data suggest that C3aR‐deficient eosinophils are not impaired in their migration toward the lung and airways following HDM challenge in spite of the presence of elevated C3a levels in the airways.

We next examined the potential of C3a as a chemoattractant for bmEos in vitro using a transwell system (for protocol see Appendix S1). The potency to attract bmEos was expressed as chemotactic index (CI) defined as: number of cells migrated in response to chemoattractant/ number of cells migrated in response to vehicle control. As expected, WT bmEos migrated toward hCCL24, a chemoattractant for both human and murine eosinophils,9 in a dose‐dependent fashion with a maximum CI of 11 at 1 µmol/L hCCL24 (Figure 2). C3a‐mediated chemotaxis of WT eosinophils exhibited a maximum response of CI = 1.3, indicating that C3a at best is a very weak chemoattractant for eosinophils, thereby corroborating the in vivo findings. In addition, these in vitro results are consistent with the findings from a previous study.5

Figure 2.

C3A is not an important chemoattractant for bone marrow‐derived eosinophils ex vivo. Chemotaxis of WT bmEos in response to increasing dose of hCCL24 or C3a. Chemotactic index (CI) is used as measure for the extent of ex vivo chemotaxis. A parametric t test was used for the comparison between groups. Data are representatives of at least two independent experiments (n = 6‐8 per group) and expressed as means ± SEM. *P < .05, **P < .01 for comparison between hCCL24 and C3a at the indicated dose

In conclusion, this study shows that C3a does not exert important chemoattractant activity on bmEos during HDM‐induced allergic lung inflammation in mice. As such, this report supports the notion 3 that C3a signaling promotes eosinophilia during allergic inflammation via an altered Th2 response rather than through a direct effect on eosinophils.

CONFLICTS OF INTEREST

The authors declare that they have no conflicts of interest.

Supporting information