Abstract
Abstract 2134
Siglec-8 is a membrane protein predominantly expressed on eosinophils, where its ligation induces cell death. Paradoxically, Siglec-8-induced cell death is markedly enhanced by the eosinophil activation and survival factor IL-5. Thus, Siglec-8 induces cell death preferentially in activated eosinophils, making it an attractive therapeutic target for eosinophil-mediated diseases. However, the mechanism of this survival factor-enhanced cell death is not known. While Siglec-8 ligation (by anti-Siglec-8 antibody) induces caspase-dependent apoptosis in resting eosinophils, it induces caspase-independent cell death in activated eosinophils. We hypothesize that co-stimulating the Siglec-8 and IL-5 pathways induces a necrotic cell death pathway. By morphologically characterizing human peripheral blood eosinophils as “apoptotic” (i.e., shrunk cells with condensed chromatin) or “necrotic” (i.e., swollen cells, disrupted membrane integrity), we found that anti-Siglec-8 + IL-5 co-stimulation yielded more necrotic eosinophils (P = 0.055, 6 donors) than stimulation with anti-Siglec-8 alone. Additionally, we stained with Annexin V and 7AAD and assessed the percent of Annexin V+ cells that are 7AAD+ as an indicator of increased transition of apoptotic cells to secondary necrosis and/or cells dying primarily by necrosis. We found that anti-Siglec-8 + IL-5 co-stimulated cells had a higher ratio of 7AAD+ cells compared with cells treated with anti-Siglec-8 alone (P < 0.001, 25 experiments with 11 independent donors). This higher 7AAD+ ratio, the morphological characteristics and the caspase-independent cell death of co-stimulated cells suggest that Siglec-8 ligation induces a necrotic form of cell death in IL-5-stimulated eosinophils by activating a specific and distinct biochemical pathway. Our previous studies have shown that reactive oxygen species (ROS) production is involved in Siglec-8-induced cell death in both resting and activated eosinophils. However, we have observed that phosphorylation of ERK1/2 and MEK1 was significantly increased in cells co-stimulated with anti-Siglec-8 + IL-5 compared to cells stimulated with IL-5 alone; anti-Siglec-8 alone did not cause ERK1/2 phosphorylation. MEK1 inhibitors U0126 and PD184352 completely blocked anti-Siglec-8 + IL-5-induced cell death; however, intracellular ROS production induced by Siglec-8 ligation was MEK1-independent. In contrast, the ROS inhibitor DPI prevented the anti-Siglec-8 + IL-5-induced enhancement of ERK1/2 phosphorylation and subsequent cell death. Enhanced ROS accumulation in IL-5 treated cells was sufficient to induce enhanced cell death, similar to anti-Siglec-8 treatment. These findings suggest that Siglec-8 ligation leads to ROS-dependent enhancement of IL-5-induced ERK1/2 phosphorylation, which results in enhanced Siglec-8-induced eosinophil cell death. How ERK phosphorylation induces cell death in co-stimulated eosinophils is not known, and ERK's involvement is surprising considering its role in activation of IL-5-stimulated eosinophils. However, recent studies have shown that ERK can be involved in specific types of cell death, namely necroptosis or autophagy, and that spatiotemporal parameters determine whether ERK will induce cell death or activation. Thus, we hypothesized that ERK localization will be altered in eosinophils co-stimulated with anti-Siglec-8 + IL-5 compared with cells treated with IL-5 alone. Western blotting of nuclear and cytoplasmic fractions and immunofluorescence suggest that enhanced ERK1/2 localization and phosphorylation are sustained for at least 2 hours in the nucleus of anti-Siglec-8 + IL-5 co-stimulated cells; cells treated with IL-5 alone have only brief ERK1/2 nuclear localization. The sustained nuclear activation of ERK may explain the change in IL-5 function from eosinophil activation/survival to necrotic death upon Siglec-8 ligation. In summary, ERK is involved in regulating the decision point for eosinophil activation, apoptosis or regulated necrosis.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.