Background
Granulocyte–macrophage colony-stimulating factor (GM-CSF) is a hematopoietic growth factor that stimulates the proliferation of granulocytes and macrophages from bone marrow precursor cells [
1]. Recent studies suggest that GM-CSF has many pro-inflammatory functions and plays an important role in the development of autoimmune and inflammatory diseases [
2]. For example, GM-CSF plays a central role in the pathogenesis of rheumatoid arthritis (RA) by activating the differentiation and survival of macrophages and neutrophils in the rheumatoid synovium [
3]. Moreover, a case report showed that the administration of GM-CSF exacerbated RA [
4]. Conversely, therapies targeting GM-CSF have been demonstrated to be effective against patients with active RA [
5]. These findings suggest that GM-CSF can prime monocyte/macrophage activation and inflammatory cytokine production, leading to a pro-inflammatory network loop that is maintained in rheumatoid synovitis [
6].
GM-CSF has also been implicated in the progression of arthritis in mice expressing a transgene encoding human interleukin-1 alpha (IL-1α) [
7], indicating a possible link between IL-1 and GM-CSF in inflammatory arthritis. IL-1β is a key cytokine involved in the regulation of immune responses as well as several inflammatory disorders [
8]. The secretion of IL-1β, in contrast to that of other inflammatory cytokines, is a tightly controlled two-step process involving the induction of pro-IL-1β and its processing into mature IL-1β by caspase-1, in which NLR family pyrin domain-containing 3 (NLRP3) inflammasome activation plays a critical role [
9]. Recent investigations suggest that GM-CSF can act as an enhancer of inflammasome-dependent IL-1β secretion in response to stimuli such as monosodium urate [
10]. The GM-CSF receptor is expressed on myeloid cells, including neutrophils. It is a heterodimer of an α-subunit that binds a common β-subunit (c β) [
11]. This c β subunit constitutively associates with Janus kinase 2 (JAK2) and undergoes tyrosine phosphorylation prior to the initiation of signaling [
12]. Signal transducer and activation of transcription (STAT) is recruited into the cytoplasmic domain of cytokine receptors and is phosphorylated by receptor-associated JAK family kinases [
13,
14].
The function of the JAK/STAT signaling pathway in neutrophils and its relationship with IL-1β production are poorly understood. Therefore, this study examined the role of GM-CSF in the cytokine network by determining its effect against neutrophils. We also determined whether an alternation of signal transduction by a JAK inhibitor could regulate the secretion of IL-1β. We report that GM-CSF directly induces IL-1β secretion from neutrophils by activating the NLRP3 inflammasome.
Methods
Reagents
Recombinant human GM-CSF and tumor necrosis factor-alpha (TNF-α) were purchased from PeproTech (Rocky Hill, NJ, USA). Anti-β-actin antibodies were purchased from Santa Cruz Biotechnology Inc. (Dallas, TX, USA). Anti-NLRP3 antibody was purchased from Merck Millipore (Billerica, MA, USA). Human IL-1β and caspase-1 (p20) enzyme-linked immunosorbent assay (ELISA) kits were purchased from R&D Systems (Minneapolis, MN, USA). Human IL-18 ELISA kits were purchased from MBL (Nagoya, Japan). Phospho-specific antibodies against JAK-1 (Tyr1022/1023), JAK-2 (Tyr1007/1008), STAT-5 (Tyr701), and STAT-3 (Tyr705) were purchased from Cell Signaling Technology (Beverly, MA, USA). Phospho-specific antibody against JAK3 (Tyr980) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Anti-phospho-1κB-α (Ser32), anti-phospho-NF-κB p65 (Ser536), anti-IκBα, and anti-NF-κB antibodies were purchased from Cell Signaling Technology (Danvers, MA, USA). Anti-phospho-specific Asc (Tyr-144) antibody was purchased from ECM Biosciences (Versailles, KY, USA). Anti-caspase-1 antibody (14F468) was purchased from Novus Biologics (Littleton, CO, USA). Tofacitinib was purchased from Sigma-Aldrich (Tokyo, Japan).
Neutrophil isolation
Venous peripheral blood was collected from healthy volunteers. Written informed consent for blood donation was obtained from each individual. The blood was layered on a Polymorphprep TM (Axis-Shield, Oslo, Norway) cushion and cells were isolated in accordance with the protocol of the manufacturer. Briefly, neutrophils were isolated on the basis of density, washed once in 0.5 N RPMI-1640 to restore osmolality, and then washed once more in RPMI-1640. Using this procedure, we obtained higher purity of CD13+ neutrophils. The cells were subsequently diluted in complete medium consisting of RPMI-1640.
ELISA and Western blot analysis
Neutrophils (2 × 106/mL) were seeded in 24-well plates containing RPMI1640 supplemented with 10% heat-inactivated fetal bovine serum and stimulated with TNF-α or GM-CSF. Cell-free supernatants were collected by centrifugation at 400g for 5 min and assayed for IL-1β or caspase-1 (p20) using ELISA kits. Caspase-1 p20 detection was carried out by using a commercially available ELISA kit (Quntikine human caspase-1 p20, R&D Systems) in which monoclonal antibody specific to the p20 subunit of caspase-1 was pre-coated onto a microplate as captured antibody and bounded caspase-1 are detected by another p20-specific polyclonal antibody.
Reverse transcription–polymerase chain reaction
Total RNA was extracted from neutrophils by using the RNeasy total RNA isolation protocol (Qiagen, Crawley, UK) in accordance with the protocol of the manufacturer. First-strand cDNA was synthesized from 1 μg of total cellular RNA by using an RNA polymerase chain reaction kit (Takara Bio Inc., Otsu, Japan) with random primers. Thereafter, cDNA was amplified by using specific primers respectively. The amplification of the IL-1β transcripts was also accomplished on a Light Cycler (Roche Diagnostics, Mannheim, Germany) by using specific primers. The housekeeping gene fragment of glyceraldehydes-3-phosphates dehydrogenase (GAPDH) was used for verification of equal loading.
Cell lysis and Western blotting
Freshly isolated neutrophils were stimulated with GM-CSF (50 ng/mL) for the times indicated in the figure legends, and the cells were washed by ice-cold phosphate-buffered saline and lysed with RIPA Buffer (Sigma-Aldrich) supplemented with 1.0 mM sodium orthovanadate, 10 μg/mL aprotinin, and 10 μg/mL leupeptin for 20 min at 4 °C. After 5 min on ice, the cell lysates were centrifuged at 10,000g for 10 min at 4 °C. After centrifugation, cellular lysates (30 μg) were also subjected to 12% SDS-PAGE followed by Western blot with antibodies against human NLRP3 or β-actin with an ECL Western blotting kit (Amersham, Little Chalfont, UK). Only in the signal transduction analysis, cells were pretreated with tofacitinib for 30 min and then stimulated with GM-CSF. Western blot analysis using phospho-specific anti-JAK and STAT antibodies was performed with an ECL Western blotting kit (Amersham).
Statistical analysis
Differences between groups were examined for statistical significance by using the Student t test. P values of less than 0.05 were considered statistically significant.
Discussion
GM-CSF is a well-known hematopoietic factor but its function exceeds that of a simple growth factor [
17]. GM-CSF modulates key aspects of both innate and adaptive immunity and plays an important role in the communication between pathogenic auto-reactive T helper (Th) cells and members of the myeloid lineages [
18]. In experimental autoimmune encephalitis, the production of GM-CSF by CD4
+ T cells is required to induce encephalitis [
19], while GM-CSF secreted by RA synovial CD4
+ T cells promotes the differentiation of inflammatory dendritic cells [
20]. The success of blocking the GM-CSF receptor in RA therapy suggests that neutralizing the GM-CSF axis could be a useful therapeutic strategy in RA [
21].
Activated neutrophils possess many of the molecular properties of macrophages as drivers of inflammatory processes, and several drugs used to treat RA can target neutrophil functions [
22]. In this study, we investigated the biological effects of GM-CSF against neutrophils as a major cell of the myeloid lineage. We describe the novel finding that GM-CSF induces inflammasome-dependent IL-1β secretion in human neutrophils. In fact, unlike TNF-α stimulation, GM-CSF stimulation resulted in marked IL-1β secretion and enhanced IL-1β gene expression in neutrophils without the need for a priming signal. Our data strongly suggest that GM-CSF upregulates IL-1β gene expression at the transcriptional as well as the post-translational level in human neutrophils. Furthermore, GM-CSF stimulation was shown to induce the secretion of caspase-1 (p20) in parallel with the upregulation of NLRP3 protein expression in neutrophils. This increased NLRP3 protein expression likely contributes to the heightened IL-1β production following GM-CSF stimulation.
Type 1 and type 2 cytokine receptors lack intrinsic enzymatic activity and associate with a family of cytoplasmic protein tyrosine kinases known as JAKs [
23]. Upon cytokine-induced activation, JAKs phosphorylate the cytoplasmic tail of the receptors, leading to the recruitment of STATs, which are also phosphorylated by JAKs [
23]. Activated STATs dimerize, translocate to the nucleus, and regulate the expression of target genes [
14]. In this study, we demonstrated that tofacitinib, an inhibitor of the JAK family, interferes with JAK2-dependent GM-CSF–driven signaling. It has previously been demonstrated that JAK2 is important in the signal transduction cascade of GM-CSF signaling [
13]. Tofacitinib demonstrated selectivity for JAK1 and JAK3 over JAK2 in a whole blood assay in which JAK2 was in its native conformation [
24,
25]. Although it is unclear how tofacitinib might spare JAK2 in experiments with isolated JAK kinases, our data clearly showed that tofacitinib efficiently blocked GM-CSF–induced JAK2 phosphorylation in human neutrophils. Tofacitinib was first designed as a JAK3-specific inhibitor, but recent studies suggest that it could be a pan-JAK inhibitor [
26].
Our data confirm that neutrophils respond to GM-CSF, which activates JAK2 and the downstream STAT3/STAT5 pathway, resulting in NLRP3 protein expression and IL-1β secretion. Neutrophils are thought to be essential in the pathogenesis of rheumatoid synovitis [
27]. Our data suggest that GM-CSF–driven neutrophils play a non-redundant role during the T cell–mediated inflammatory process by polarizing the innate immune activation, including inflammasome activation and IL-1β induction as demonstrated in neutrophils previously [
28]. Our data showed that GM-CSF sustains the expression of NLRP3 in neutrophils. Given that NLRP3 expression is known to activate innate immune cells [
29], a functional NLRP3 inflammasome resulting in IL-1β induction within the synovium would be expected to contribute to rheumatoid synovitis. Our observations highlight the role of IL-1β by arthrogenic Th cells in the inflamed synovium. Therapeutic interventions targeting GM-CSF during rheumatoid inflammation are likely to restrict not only activated T cells producing GM-CSF but also GM-CSF–responding neutrophils and subsequent IL-1β secretion. This study provides evidence for how GM-CSF impacts on rheumatoid synovitis and which cell type requires a GM-CSF signaling event to mediate inflammation. Pathogenic T cells are the most abundant cellular infiltrates in the rheumatoid synovium and thus cause rheumatoid inflammation [
30]. Our data suggest that JAK inhibitors have the potential to block multiple cytokine pathways, including the GM-CSF–mediated autoinflammatory cascade.
There was a limitation in our study. We measured the p20 subunit of caspase-1 in culture supernatants by using caspase-1 p20-specific ELISA. However, Western blot or enzymatic assay should be required to demonstrate the bioactive p20 subunit of caspase-1.
Conclusion
We have shown that GM-CSF is a strong inducer of IL-1β by activating the inflammasome in neutrophils. Our results indicate that GM-CSF signaling controls the pathogenic expression of IL-1β in neutrophils, which may cause innate cell activation, inflammation, and cartilage damage in RA. Therefore, GM-CSF emerges as a communicator between pathogenic lymphocytes and neutrophils through activating the NLRP3 inflammasome.