Oral Syk inhibitor attenuated inflammation in FcγRIIb−/− lupus mice.
To determine the efficacy of Syk inhibitor against SLE, R788 (fostamatinib) was orally administrated for four weeks to 40-wk-old female FcγRIIb−/− mice (symptomatic lupus model) and age-gender matched wide type (WT) mice, to imitate the clinical situation (Fig. 1A). Indeed, 40-wk-old FcγRIIb−/− mice showed lupus characteristics, including increasing levels of serum anti-dsDNA and proteinuria with proliferative glomerulonephritis in renal histology (Supplement Fig. 1A-C), spontaneous elevation of serum cytokines, as evaluated by tumor necrosis factor alpha (TNFa) and interleukin-6 (IL-6), and an extracellular traps biomarker (serum citrullinated histone H3; CitH3) (Supplement Fig. 1D-G). Additionally, lipopolysaccharide (LPS) and beta-glucan (BG), the major components of bacteria and fungi in gut microbiota, were observed in serum of FcγRIIb−/− mice together with gut permeability defect, as tested by a fluorescein isothiocyanate-dextran (FITC-dextran) assay (Supplement Fig. 1H-J). These data support active lupus with impaired gut permeability (leaky gut) and elevated microbial molecules in serum of 40-wk-old female FcγRIIb−/− mice, supporting previous studies in patients and in mice (10, 11).
With Syk inhibitor, lupus characteristics and systemic inflammation in FcγRIIb−/− mice were less prominent than the mice without inhibitor, as indicated by serum anti-dsDNA, proteinuria, and serum cytokines (TNFa but not IL-6) (Fig. 1C-F). The heatmap shows a summary of the alteration between FcγRIIb−/− mice with versus without the inhibitor (Fig. 1G). Similarly, the mice with Syk inhibitor had less severe renal histology (Fig. 1H), as indicated by glomerular expansion (Fig. 1I) with a trend to reduced renal tubular injury (Fig. 1J) compared with vehicle-treated FcγRIIb−/− mice. In accordance with these findings, renal immune-complex accumulation (Fig. 1K-L) and level of immune-complex deposition in glomeruli (Fig. 1M-N) were attenuated in Syk inhibitor-treated group. However, gut leakage parameters (serum FITC-dextran, serum endotoxin, and serum BG) did not show a significant difference between Syk inhibitor treatment and vehicle groups (Supplement Fig. 1K).
Syk inhibitor attenuated responses of FcγRIIb −/− macrophages after activation by lipopolysaccharide (LPS) plus whole glucan particle (WGP)
Macrophages play a critical role in inflammation by producing various inflammatory mediators after recognizing several stimuli, including LPS and BG, by several receptors on the cell surface that activates several downstream signals, including several kinase enzymes (14, 26). The extraction of bone marrow-derived macrophages (BMDMs) from femurs yielded more than 90% purity, as indicated by F4/80 macrophage indicator in flow-cytometry analysis (Fig. 2A). Because FcγRIIb (CD32b) was not detectable in FcγRIIb−/− BMDMs, the effect of microbial molecules (LPS and BG) on CD32b alteration was observed in WT BMDMs (Fig. 2B) using flow cytometry. In WT BMDMs, only LPS alone and LPS with whole glucan particle (WGP; the representative BG), elevated abundance of CD32b, whereas WGP alone did not elevate FcγRIIb. The reduced inhibitory FcγRIIb after LPS + WGP compared with LPS alone (bacterial molecule alone) might be responsible for the more prominent inflammation in LPS + WGP (combined the molecules from bacteria and fungi) (Fig. 2C). Because of the co-elevation of LPS and BG in the serum of mice with active lupus (Supplement Fig. 1H, I), LPS + WGP with and without a Syk inhibitor (R406; the active form of Syk inh), but not LPS alone, were further tested.
In comparison with WT macrophages, LPS + BG-activated FcγRIIb−/− BMDMs demonstrated prominent protein expressions of both Syk and phosphorylated Syk (p-Syk), as evaluated by Western blotting, with the more prominent M1 pro-inflammatory macrophage polarization (CD206low, CD86high) as analyzed by flow cytometry (Fig. 2D-H). In LPS + BG-activated FcγRIIb−/− BMDMs with Syk inh, the attenuated of p-syk with a shift toward M2 macrophage polarization (CD206high, CD86low; anti-inflammatory marker) were demonstrated in a dose-dependent manner (Fig. 2D-H). In parallel, supernatant cytokines (TNFa and IL-6, but not IL-10) in FcγRIIb−/− BMDMs were higher than WT BMDMs and Syk inh also attenuated these cytokines in a dose-dependent manner (Fig. 2I-K). Due to the well-known importance of DNA extracellular traps (ETs) in macrophages (27), macrophages with ETs were measured by co-staining immunofluorescent between DAPI (4′,6-diamidino-2-phenylindole; blue color) nucleus staining and citrullinated histone 3 (citH3; green color) together with supernatant citH3 (28) (Fig. 2L-N). Indeed, macrophage extracellular traps (METs) of LPS + WGP-activated FcγRIIb−/−macrophages were prominent than LPS + WGP-stimulated WT cells. Notably, Syk inhibitor inhibited METosis (cell death after METs) in both WT and FcγRIIb−/− BMDMs (Fig. 2L-N) with a reduction in apoptosis (annexin V and propidium iodide measured by flow cytometry) in a dose-dependent manner (Fig. 2O-P).
In accordance with these results, the hyper-activated Syk in LPS + WGP-stimulated FcγRIIb−/− macrophages compared with the stimulated WT cells might be associated with hyperinflammatory responses, as indicated by cytokine release, M1 polarization, cell apoptosis, especially METs. As expected, lists of the genes to explain METosis following the KEGG pathway of neutrophil extracellular traps was applied on a heat map (Supplement Fig. 2). The results of LPS + WGP stimulated macrophages from WT and FcγRIIb−/− mice demonstrated several genes that might be associated with NETosis, including NET stimulators (FcγR genes), NET downstream signals (PIK3K-ATK, phosphokinase C, and protein kinase C), chromatin de-condensation genes, and oxidative stress genes, as demonstrated which were reduced in Syk inh treatment condition consistency with the METosis result (Fig. 2L-N).
Syk inhibitor attenuated inflammation through Syk-p38MAPK-dependent pathway.
To demonstrate the molecular mechanism of the Syk inhibitor in macrophages, transcriptomic analysis was performed. In comparison between LPS plus WGP-stimulated FcγRIIb−/− BMDMs without Syk inhibitor (KO_SYN) and with Syk inhibitor (KO_SYKI), there were 1,346 up- and 2,199 down-regulated genes, as indicated by the heat map and the Volcano plot analysis (Fig. 3A-B) with the highest alteration in gene-related signal transduction (560 genes) (Fig. 3C). Meanwhile, the mitogen-activated protein kinase (MAPK) pathway and TNFa signaling were the top 2 pathway with the highest significant degrees of enrichment in the network analysis of pathway terms (Fig. 3D). According to the enrichment pathway in the group with Syk inhibitor, the major direction of the expressed genes was the downregulation of MAPK signaling pathway, as there were 91 up-regulated MAPK-related genes with only 23 down-regulated genes (Supplement Fig. 3). Consistent with previous enrichment, the comparative between unstimulated and stimulated (LPS + WGP) conditions in FcγRIIb−/− BMDMs showed the MAPK signaling pathway as the secondary highest degree of enrichment (Supplement Fig. 4A). The subgroup analysis in stimulated condition between WT and FcγRIIb−/− BMDMs (heat map) on the 2 top highest significant degrees of enrichment-related pathways (MAPK and TNF signaling) also showed the upregulation of MAPK and TNF signaling-related genes in FcγRIIb−/− BMDMs (Supplement Fig. 4B, 4C). These data implied the possible importance of MAPK in LPS + WGP activation in FcγRIIb−/− macrophages.
Based on the correlation of MAPK to other well-known molecules (29, 30), Syk might regulate MAPK signaling through extracellular signal-regulated kinases (ERK), c-Jun N-terminal kinases (JNK). Then, these molecules were explored in FcγRIIb−/− macrophages using western blot (WB) analysis (Fig. 3E). In FcγRIIb−/− macrophages without Syk inh, LPS + WGP more prominently activated phosphorylated Syk (p-Syk) and phosphorylated p38MAPK (p-p38MAPK), but not phosphorylated ERK and JNK (p-ERK and p-JNK) (Fig. 3F-3I). The WT macrophages and Syk inh also demonstrated significantly reduced the abundance p-Syk and p-p38MAPK (Fig. 3F-3I). In parallel, p38MAPK inhibitor (Adezmapimod) was tested for the correlation between P38MAPK and LPS + WPG activation in macrophages. As such, the western blot analysis demonstrated that the p38MAPK inhibitor reduced p-p38MAPK abundance without an impact on p-Syk (Fig. 3J-3L). Additionally, p38MAPK inhibitor also attenuated LPS + WGP-induced METs formation as determined by nuclear morphology (DAPI) colocalized with citH3 (FITC conjugated) (Fig. 3M-O) and the supernatant cytokines (Fig. 3P-R) in both FcγRIIb−/− and WT BMDMs in a dose-dependent manner. Although the anti-inflammatory effect of p38MAPK inhibitor was similar to Syk inhibitor, the apoptosis inhibition by p38MAPK inhibitor was not observed (Supplement Fig. 5). Despite a similar impact of LPS + WGP in both WT and FcγRIIb−/− macrophages on inflammatory induction, Syk and inflammatory markers of FcγRIIb−/− cells were more prominent and anti-inflammatory impact of Syk inh was more obvious in FcγRIIb−/− macrophages.
Syk inhibitor also reduced inflammatory responses of FcγRIIb −/− neutrophils through Syk-p38MAPK dependent pathway.
A potential role of neutrophils in SLE pathogenesis and organ damage is well described, including neutrophil extracellular traps (NETs). Here, neutrophils were isolated from BM by a magnetic-based assay with an approximate purity of 82% (Fig. 4A). After LPS + WGP activation, FcγRIIb−/− neutrophils demonstrated prominent increased p-Syk with similarly elevated p-p38MAPK when compared with LPS + WGP-activated WT neutrophils with Syk inh, as indicated by WB analysis (Fig. 4B-D). Likewise, LPS + WGP also elevated neutrophil supernatant cytokines, including TNFa, and IL-6, but not IL-10 (ELISA assay), NETosis (using colocalized FITC-citH3 with DAPI-nuclear morphology and supernatant citH3 levels), and apoptosis (flow cytometry by annexin V and propidium iodide) that were more prominent in FcγRIIb−/− than WT neutrophils and were attenuated by Syk inh in both cell types (Fig. 4E-L). With p38MAPK inhibitor in LPS + WGP activation, NETs and supernatant cytokines (TNFa and IL-6) were attenuated without an effect on supernatant IL-10 and apoptosis in both FcγRIIb−/− and WT neutrophils (Supplement Fig. 6). These findings implied a crucial role of Syk in stimulating FcγRIIb−/− neutrophils through the Syk-p38MAPK axis, similar to macrophages.
Syk inhibitor attenuates inflammation and extracellular traps (ETs) formation in FcγRIIb −/− lupus mice.
Due to the Syk inh impact against pro-inflammatory responses and ETs formation of LPS + WGP-activated macrophages and neutrophils (Fig. 3, 4), Syk inh was further tested in mice using 4-wk-oral administration of Syk inh in 40-wk-old FcγRIIb−/− mice, a symptomatic lupus model, as indicated by positive anti-dsDNA with proteinuria, leaky gut (FITC-dextran assay), endotoxemia, and glucanemia (Supplement Fig. 1). As such, the reduced Syk activation in several organs (kidneys, spleens, and large intestines) in Syk inh-administered FcγRIIb−/− mice with a prominent decrease in abundance in the spleen of Syk, p38MAPK, and apoptosis (cleavage activated caspase 3), as assessed by immunohistochemistry, was demonstrated (Fig. 5A-C). The co-staining of anti-F4/80 with anti-p-Syk immunofluorescent staining revealed that 62.5 ± 12.5% of the total Syk-positive cells at the white pulp of the spleen were macrophages, and Syk inh reduced Syk abundance in both macrophages (red-colored bar) and non-macrophages (gray-colored bar) (Fig. 5D-F).
Not only Syk abundance, Syk inh also decreased ETs formation, as indicated by reduced serum citH3 and serum ds-DNA after 2 and 4 wks of administration (Fig. 5G-H) and decreased METs formation in spleen, as determined by colocalized F4/80 with citH3 immunofluorescence (Fig. 5I-K). Interestingly, the total positive citH3 cells were mainly F4/80-positive cells (macrophages) at approximately 80.5 ± 10.2%, supporting the role of METosis in SLE pathogenesis (Fig. 5J). The F4/80hiCD11bhi macrophages (mature resident macrophages) in the spleen (flow cytometry) were not decreased by Syk inh; however, CD86-positive macrophages (active pro-inflammatory M1 macrophage polarized cells) (the possible drivers of ET-related pathogenesis) (27), were decreased in FcγRIIb−/− mice (Fig. 5L-N). In the kidney, METs (co-localization of F4/80 with citH3 immunofluorescence) mostly presented in the tubulointerstitial area but not in the glomeruli; however, renal METs were also reduced by Syk inh (Fig. 5O-Q). Parallelly, the prominent citH3 positive cells in kidneys were also macrophages (F4/80 positive) at approximately 63.4 ± 19.5% of all renal citH3-positive cells (Fig. 5P). Taken together, these results support the anti-inflammatory impact of Syk inh in FcγRIIb−/− mice and might be useful for patients.