Differential regulation of TNFalpha and GM-CSF induced activation of P38 MAPK in neutrophils and eosinophils

P38 MAPK is a central mediator in cytokine signalling in human leukocytes. P38 MAPK is activated by phosphorylation of a conserved Thr180-X-Tyr182 motif by dual phosphorylation via the upstream kinases MKK3 and MKK6. Alternatively, P38 MAPK can be activated via autophosphorylation when associated with TAB1. In this study P38 MAPK phosphorylation and activation (measured via phosphorylation of P38 MAPK downstream target MK2) were investigated upon engagement of the GM-CSF- and TNFalpha-receptors expressed on both eosinophils and neutrophils. The MKK3/MKK6 pathway mediated neutrophil P38 MAPK activation after stimulation with TNFalpha (100U/ml) or GM-CSF (10(-10)M). Under these conditions the activation but not phosphorylation of P38 MAPK could be inhibited by SB203580 (10(-5)M or 10(-6)M). In eosinophils SB203580 (10(-6)M) inhibited both the phosphorylation and activation of P38 MAPK after stimulation with several doses of TNFalpha (10-10000 U/ml) or GM-CSF (10(-11) to 10(-9)M), indicating that P38 MAPK activation is mediated via autophosphorylation in eosinophils. This hypothesis was supported by the finding that in marked contrast to neutrophils, MKK3/MKK6 did not show enhanced phosphorylation in eosinophils after cytokine stimulation, but were constitutively phosphorylated. Therefore, the involvement of TAB1 was investigated with regard to this cytokine-induced autophosphorylation. Co-immunoprecipitation experiments showed that TAB1 was constitutively associated with P38 MAPK in eosinophils and neutrophils and that cytokine-induced autophosphorylated P38 MAPK was co-precipitated with TAB1. These findings are consistent with the hypothesis that cytokine-induced autophosphorylation of P38 MAPK in primary granulocytes depends on the interaction with TAB1.     

MicroRNA-21* regulates the prosurvival effect of GM-CSF on human eosinophils

Eosinophils are the principal effector cells of allergic inflammation, and hematopoietic cytokine granulocyte macrophage colony-stimulating factor (GM-CSF) is the primary cytokine that activates and prolongs the survival of eosinophils in local inflammatory sites by mediating anti-apoptotic activity in allergic inflammation. To investigate the immunopathological role of microRNA (miRNA) in allergic inflammation, we elucidated the regulatory mechanisms of miRNA on the GM-CSF-mediated in vitro survival in eosinophils. Eosinophils were purified from fresh human peripheral blood buffy coat fraction obtained from adult volunteer using microbead magnetic cell sorting. The apoptosis, viability and phosphorylation of extracellular signal-regulated kinase (ERK) were assessed by flow cytometry, and the expression of miRNA was analyzed using Agilent Human miRNA Microarray with Human miRNA Microarray Version 3 and real time RT-PCR. We have confirmed the increased in vitro viability of GM-CSF-treated eosinophils and upregulated expression of miRNA-21* (miR-21*), a complementary miRNA of miR-21, in GM-CSF-treated eosinophils. The transfection of pre-miR miR-21* precursor molecule could up-regulate the miR-21* expression, subsequently enhance the GM-CSF-activated ERK pathway and reverse the apoptosis of eosinophils, while anti-miR-21* inhibitor could down-regulate the miR-21* expression, suppress the GM-CSF-activated ERK pathway and enhance the apoptosis. Our results should shed light on the potential immunopathological role of miRNA-21* regulating the in vitro apoptosis of eosinophils and development of novel molecular treatment of allergic inflammation.     

Ligation of intercellular adhesion molecule 3 inhibits GM-CSF production by human eosinophils

BACKGROUND: Intercellular adhesion molecule 3 (ICAM-3) has recently been identified on the surface of eosinophils. OBJECTIVE: The purpose of this study was to characterize ICAM-3 expression on eosinophils in response to cytokines and to determine whether ligand binding of ICAM-3 modulates inflammatory responses of eosinophils, as it does in other leukocytes. METHODS: To determine effects of ICAM-3 on eosinophil function, we isolated human eosinophils and used a monoclonal antibody directed against the epitope of ICAM-3 that binds to leukocyte-function antigen-1 to mimic binding of ICAM-3 and this natural ligand. We measured granulocyte-macrophage colony stimulating factor (GM-CSF) production by unstimulated eosinophils and eosinophils stimulated with ionomycin (1 micromol/L), both in the presence and absence of this anti-ICAM-3 antibody. RESULTS: We found that 99% of eosinophils expressed ICAM-3, regardless of whether allergic symptoms were present or absent. Expression of ICAM-3 was not enhanced by proinflammatory cytokines. Expression of ICAM-3 was reduced in apoptotic cells and in cells incubated with the combination of GM-CSF and tumor necrosis factor-alpha (n = 3). Antibody binding of ICAM-3, which mimics leukocyte-function antigen-1 binding, had no effect on baseline GM-CSF production but reduced by 80% the production of GM-CSF stimulated by ionomycin (control 1969 pg/mL +/- 1259 SD versus anti-ICAM-3 396 pg/mL +/- 207 SD, n = 8) and reduced GM-CSF mRNA content. CONCLUSIONS: ICAM-3 is highly expressed on the surface of human eosinophils, and downregulation of GM-CSF production by anti-ICAM-3 mAb suggests that ICAM-3 ligation may inhibit eosinophil inflammatory responses and survival.     

Expression and regulation of monocyte chemoattractant protein-1 by human eosinophils

Several recent studies have identified eosinophils as a cellular source of various cytokines, indicating that eosinophils play not only an effector role, but also a regulatory role within the allergic inflammatory cell network. In this study, we demonstrate that eosinophils can generate and secrete monocyte chemoattractant protein-1 (MCP-1), a prototype of C-C chemokines. Eosinophils generated immunoreactive MCP-1 in response to such diverse stimuli as C5a, formylmethionyl-leucyl-phenylalanine (FMLP) and ionomycin, but MCP-1 production was not induced by interleukin (IL)-1 or tumor necrosis factor-α. C5a- and FMLP-induced eosinophil MCP-1 production was absolutely dependent on pretreatment with cytochalasin B. Eosinophils elaborated significantly more MCP-1 than neutrophils. Immunoreactive MCP-1 was detected at 6 h of incubation with C5a or FMLP. Expression of MCP-1 mRNA reached a maximum within the first 3 h after stimulation and then declined rapidly to a very low and stable level by 18 h. Pretreatment with IL-5 markedly amplified C5a-induced MCP-1 production, and the enhancement occurred at the pretranslational level. Eosinophilactive chemokines such as eotaxin failed to induce MCP-1 generation, even when eosinophils were primed by IL-5. Since MCP-1 exerts a potent histamine-releasing effect on human basophils, our results indicate that eosinophils may regulate basophil mediator release with possible consequent contribution to the pathogenesis of allergic inflammation via a paracrine mechanism.     

Excretory-secretory products secreted by Paragonimus westermani delay the spontaneous cell death of human eosinophils through autocrine production of GM-CSF

BACKGROUND: Eosinophils play important roles in tissue inflammatory responses associated with helminth infections. Excretory-secretory products (ESP) produced by tissue-invasive helminths contain a large quantity of proteolytic enzymes that can modulate the host's immune responses. However, little is known regarding the roles of worm-derived products that are responsible for eosinophilic inflammatory responses in helminth infections. OBJECTIVE: In the present study, we investigated whether ESP produced by Paragonimus westermani, which cause pulmonary or extrapulmonary paragonimiasis in human beings, regulates both cell survival and death of human eosinophils. METHODS: The ESP was obtained from P. westermani newly excysted metacercariae (PwNEM). Eosinophils were purified from peripheral blood of healthy donors, and the purified eosinophils were incubated with or without the ESP secreted by PwNEM. The viability of eosinophils was assessed by staining with propidium iodide using the flow cytometer. RESULTS: When eosinophils were incubated with a low concentration of the ESP produced by PwNEM, which totally consists of proteolytic enzymes, eosinophil cell death was delayed compared with results for cells incubated with medium alone. In fact, the ESP at a low concentration stimulated eosinophils to produce detectable levels of GM-CSF that can delay eosinophil cell death. In contrast, eosinophil cell death was dose-dependently accelerated when cells were incubated with high concentrations of the ESP. To see whether the dose-dependent biphasic survival effect of the ESP on eosinophils is primarily due to the protease activity contained in the ESP, a high dose of the ESP was treated with heat at 56 degrees C for 30 min before being added to eosinophils. Attenuating protease activity in a high dose of the ESP by heat treatment reversed the ESP-afforded eosinophil cell death. This prolonged survival of eosinophils induced by the heated ESP was remarkably inhibited by anti-GM-CSF-neutralizing mAb and Jak2 kinase inhibitor AG-490. CONCLUSION: These results suggest that the proteases in the ESP secreted by PwNEM are able to regulate eosinophil survival through the autocrine production of GM-CSF. Thus, the enhanced eosinophil survival induced by Paragonimus-secreted products may contribute to the elicitation of eosinophilic inflammatory responses at the worm-infected lesion in human paragonimiasis.     

CR3-dependent negative regulation of human eosinophils by Mycobacterium bovis BCG lipoarabinomannan

Eosinophils have recently been shown to participate in innate immune responses against mycobacteria. We have investigated whether Mycobacterium bovis BCG regulate the human eosinophil immune response. A negative correlation between mycobacteria internalization and eosinophil activation was observed. In addition, mannose-capped lipoarabinomannan from M. bovis BCG (ManLAM) failed to induce a significant release of eosinophil peroxidase and TNF-α. Noteworthy, ManLAM exhibited a potent inhibitory effect on eosinophil peroxidase release by TLR2-activated eosinophils involving the complement receptor-3 molecule and the phosphatidylinositol-3 kinase pathway. ManLAM, generally present in pathogenic mycobacteria, plays an important role in modulating eosinophil-dependent immune response.     

IL-9 expression by human eosinophils: regulation by IL-1beta and TNF-alpha

BACKGROUND: IL-9 is a pleiotropic cytokine that exhibits biologic activity on cells of diverse hemopoietic lineage. IL-9 stimulates the proliferation of activated T cells, enhances the production of IgE from B cells, and promotes the proliferation and differentiation of mast cells and hematopoietic progenitors. OBJECTIVE: In this study we evaluated the expression of IL-9 messenger (m)RNA and protein by human peripheral blood eosinophils. We also investigated the role of IL-1beta and TNF-alpha in the release of IL-9 from human peripheral blood eosinophils. METHODS: RT-PCR, in situ hybridization, and immunocytochemistry were used to investigate the presence of IL-9 mRNA and protein in human peripheral blood eosinophils from asthmatic patients and normal control subjects. Furthermore, biologic assay was used to investigate the release of IL-9 protein from IL-1beta- or TNF-alpha-stimulated eosinophils in vitro. RESULTS: RT-PCR analysis showed the presence of IL-9 mRNA in human peripheral blood eosinophil RNA preparations from subjects with atopic asthma, as well as in the eosinophil-differentiated HL-60 cell line. By using in situ hybridization, a significant difference (P <.01) in IL-9 mRNA expression was detected in human peripheral blood eosinophils freshly isolated from asthmatic subjects compared with those isolated from normal control subjects. Furthermore, the percentage of IL-9 immunoreactive eosinophils from asthmatic patients was increased compared with that found in normal control subjects (P <.01). We also demonstrate that cultured human peripheral blood eosinophils from asthmatic subjects synthesize and release IL-9 protein, which is upregulated on stimulation with TNF-alpha and IL-1beta. CONCLUSION: Human eosinophils express biologically active IL-9, which suggests that these cells may influence the recruitment and activation of effector cells linked to the pathogenesis of allergic disease. These observations provide further evidence for the role of eosinophils in regulating airway immune responses.     

Incidence of GM-CSF antibodies in cancer patients receiving GM-CSF for immunostimulation

We have assessed the immunogenicity profile of GM-CSF in patients with either colorectal carcinoma (CRC) at different stages of disease or with multiple myeloma who were given recombinant human GM-CSF (Escherichia coli-derived) combination therapy. Metastatic CRC patients received a colon carcinoma-reactive antibody and high doses of GM-CSF (425--500 microg/day for 10 days), while other CRC patients and those with myeloma received low doses of GM-CSF (75--80 microg/day for 4 days) as an adjuvant along with appropriate tumor antigens. We found that 55% of the patients (11/20) given high doses of GM-CSF developed GM-CSF-reactive antibodies in comparison with an incidence of only 16% (4/25) in patients given low doses of GM-CSF. None of the patients developed neutralizing antibodies and so the biological effects of GM-CSF were not compromised. A majority of patients (80%) (36/45) also developed antibodies to E. coli proteins that were present as trace contaminants in the GM-CSF product. Treatment with recombinant GM-CSF products, therefore, may induce antibodies against this cytokine depending on the regimen and the amounts used. In this study, multiple immunizations with low doses of GM-CSF was associated with a low incidence of GM-CSF antibodies, which did not neutralize the effect of the cytokine. This therapeutic strategy was effective in inducing adjuvant-type effects and needs to be explored in further clinical trials with this cytokine.     

GM-CSF Biology

Granulocyte macrophage-colony stimulating factor (GM-CSF) was originally defined by its ability to generate in vitro granulocyte and macrophage colonies from bone marrow precursor cells. Apart from its physiological role in the control of alveolar macrophage development, it now appears more likely that its major role lies in its ability to govern the properties of the more mature myeloid cells of the granulocyte and macrophage lineages, particularly during host defence and inflammatory reactions. This review summarizes the in vivo evidence to support this proposition. This evidence includes both the findings obtained by administration of GM-CSF, e.g. as an adjuvant, and also includes those observed in depletion studies, e.g. during inflammatory reactions where GM-CSF can be shown to have a proinflammatory action.     

The effects of recombinant granulocyte-macrophage colony-stimulating factor (GM-CSF) and interleukin-3 on the secretory capacity of human blood eosinophils.

The effects of recombinant human GM-CSF and interleukin-3 (IL-3) on human blood eosinophil survival, activation, and secretion were studied. Purified normal density eosinophils from patients with the idiopathic hypereosinophilic syndrome (HES) survived in culture for 7 days (50% viable) in the presence of 50 nM GM-CSF or 50 nM IL-3. Neutrophils did not survive after 4 days. No eosinophils survived in the absence of GM-CSF or IL-3. In two out of five patients studied, the cultured eosinophils became elongated with numerous processes. In all five patients the cells became adherent, but there were no morphological signs of degranulation. Both GM-CSF and IL-3 activated eosinophils, transforming the storage form of eosinophil cationic protein (ECP) into the secreted form. The proportion of activated cells increased from less than 20% to over 50% after 4 days in culture. However, GM-CSF and IL-3 did not induce secretion on their own. On the other hand, when GM-CSF/IL-3-activated eosinophils were exposed to known secretory stimuli, there was a six-fold increase in the amount of ECP released when the cells were stimulated with sepharose coated with C3b, and a two-fold increase when they were stimulated with sepharose-activated whole autologous serum. Eosinophils from patients taking steroids were unable to secrete their granule contents, even though they became activated by GM-CSF and IL-3. A novel finding was that sepharose-activated whole serum was an extremely potent secretory signal for ECP, releasing up to 50% of the total ECP content. These studies showed that GM-CSF and IL-3 prime eosinophil effector function by initiating granule solubilization which is the first step in the secretory event, without affecting the subsequent extracellular release of granule proteins.     

Intraepithelial eosinophils in esophagitis

Red blood cell alloantibody production was studied in 90 neonates who received a mean of 14.1 transfusions (range 2-35) from an average of 8.9 donors during the first three months after birth. Standard antibody detection procedures were done with the use of a selected red blood cell panel. No unexpected alloantibodies were detected. These findings suggest, at a 99% confidence level, that neonates do not make red blood cell alloantibodies in response to transfusion, indicating that repeated compatibility testing is probably unnecessary. Thus, following initial antibody screening and compatibility tests, further compatibility testing can be eliminated.     

Gastrointestinal eosinophils

Summary: The gut‐associated lymphoid tissue (GALT) is composed of lymphocytes residing in Peyer's patches, lamina propria, and intraepithelial compartments. In addition to these features which distinguish GALT from other peripheral sites of the immune system, the gastrointestinal immune system is also composed of resident eosinophils. Eosinophils are generally considered to be peripheral blood leukocytes that have an important pro‐inflammatory role in various immune disorders. Although most research concerning this cell has focused on understanding its trafficking and function in the blood and lung, recent studies have also started to elucidate its regulation and function in the gastrointestinal tract. Interestingly, eosinophil numbers in the gastrointestinal tract are substantially higher than in other tissues. At baseline (healthy conditions), most eosinophils reside in the lamina propria in the stomach and intestine. Eosinophil homing to these sites occurs during embryonic development and their levels in perinatal mice are comparable to those in adults, indicating that their homing is not dependent upon the presence of intestinal flora. Furthermore, eosinophil localization to the lamina propria at baseline is critically regulated by eotaxin, a chemokine constitutively expressed throughout the gastrointestinal tract. Although eotaxin is required for eosinophil homing, its expression in the esophagus is not sufficient for eosinophil accumulation, since this organ is devoid of eosinophils at baseline. During Th2‐associated inflammatory conditions (e.g. interleukin (IL)‐5 overexpression or oral allergen challenge), marked increases of eosinophils occur not only in the lamina propria but also in Peyer's patches. The accumulation of Peyer's patch eosinophils, which mainly occurs in the outer cortex and interfollicular regions, is critically regulated by IL‐5 and less significantly by eotaxin, suggesting the involvement of other eosinophil chemokines in this lymphoid compartment. Preliminary investigations have shown that gastrointestinal eosinophils express the α4β7 integrin and that this molecule is responsible, in part, for eosinophil homing. In summary, eosinophils are resident cells of the gastrointestinal immune system whose levels can be induced by antigen exposure under Th2 conditions, in a manner that is critically regulated by eotaxin and IL‐5. We propose that eosinophils are integral members of the gastrointestinal immune system and are likely to be important in innate, regulatory and inflammatory immune responses. This work was supported in part by the National Health Medical Research Council (Australia) C.J. Martin Post‐doctoral Fellowship (S.P.H.), the Jaffe Family Fund of the American Academy of Allergy, Asthma, and Immunology (S.P.H.), NIH grant R01 AI45898 (M.E.R.) and the Human Frontier Science Program (M.E.R.). The authors wish to thank Drs. K. Frank Austen, Mitchell Cohen, Paul Foster, Glenn Furuta, and Nives Zimmermann for helpful discussions, as well as numerous other colleagues.