Inhibition of Bruton's TK regulates macrophage NF‐κB and NLRP3 inflammasome activation in metabolic inflammation

Background and Purpose There are no medications currently available to treat metabolic inflammation. Bruton's tyrosine kinase (BTK) is highly expressed in monocytes and macrophages and regulates NF‐κB and NLRP3 inflammasome activity; both propagate metabolic inflammation in diet‐induced obesity. Experimental Approach Using an in vivo model of chronic inflammation, high‐fat diet (HFD) feeding, in male C57BL/6J mice and in vitro assays in primary murine and human macrophages, we investigated if ibrutinib, an FDA approved BTK inhibitor, may represent a novel anti‐inflammatory medication to treat metabolic inflammation. Key Results HFD‐feeding was associated with increased BTK expression and activation, which was significantly correlated with monocyte/macrophage accumulation in the liver, adipose tissue, and kidney. Ibrutinib treatment to HFD‐fed mice inhibited the activation of BTK and reduced monocyte/macrophage recruitment to the liver, adipose tissue, and kidney. Ibrutinib treatment to HFD‐fed mice decreased the activation of NF‐κB and the NLRP3 inflammasome. As a result, ibrutinib treated mice fed HFD had improved glycaemic control through restored signalling by the IRS‐1/Akt/GSK‐3β pathway, protecting mice against the development of hepatosteatosis and proteinuria. We show that BTK regulates NF‐κB and the NLRP3 inflammasome specifically in primary murine and human macrophages, the in vivo cellular target of ibrutinib. Conclusion and Implications We provide “proof of concept” evidence that BTK is a novel therapeutic target for the treatment of diet‐induced metabolic inflammation and ibrutinib may be a candidate for drug repurposing as an anti‐inflammatory agent for the treatment of metabolic inflammation in T2D and microvascular disease.


| INTRODUCTION
The worldwide prevalence of obesity has doubled since 1980 with over 1.9 billion people being considered overweight or obese (Swinburn et al., 2011). Diet-induced obesity induces a state of chronic metabolic inflammation (Saltiel & Olefsky, 2017). The control of energy and metabolism in tissues is directed largely by innate immune cells, such as macrophages leading to the production of soluble effector molecules including cytokines and chemokines (Tanti, Ceppo, Jager, & Berthou, 2013). This state of "low-grade" chronic inflammation predisposes individuals to metabolic syndrome and the associated co-morbidities that develop over time including insulin resistance, Type 2 diabetes (T2D), cardiovascular disease and nonalcoholic fatty liver disease (Lusis, Attie, & Reue, 2008) (Esposito & Giugliano, 2004).
Immune cell accumulation specifically from the myeloid linage coupled with the activation of the pathways including those involving NF-κB and the NLRP3 inflammasome have been heavily implicated in orchestrating the inflammatory response in T2D (Arkan et al., 2005) (Vandanmagsar et al., 2011). Pharmacological inhibition or genetic deletion of components of the NF-κB pathway (i) prevents the development of HFD-induced insulin resistance (Chiazza et al., 2015) (Benzler et al., 2015) and (ii) slows the progression of microvascular disease. Equally, inhibition or genetic deletion of key components the NLRP3 inflammasome reduced systemic inflammation in models of diet-induced obesity and protected again the development of peripheral insulin resistance in mice (Chiazza et al., 2016).
Currently, there are no medicines for T2D that target metabolic inflammation and/or prevent the development of microvascular complications. One option for generating new treatments for rapid patient benefit is to repurpose existing U.S. Food and Drug Administration (FDA) or European Medicines Agency (EMA) approved drugs.
Exploring existing medications that have under-appreciated antiinflammatory effects for new indications could be a time-and costeffective approach to prevent and/or treat the development of microvascular complications of T2D. Anti-inflammatory agents have been shown to be powerful tools to study disease pathophysiology in preclinical models of T2D. However, to date, little translational research as followed up on these successes.
Germline mutations in Bruton's tyrosine kinase (BTK) have been implicated in the primary immunodeficiency disease X-linked agammaglobulinaemia, through its essential role in B lymphocyte development (Nyhoff et al., 2018). The first BTK inhibitor to be FDA approved for clinical use was ibrutinib for the treatment of chronic lymphatic leukaemia (CLL) (Davids & Brown, 2012). However, BTK has an additional role in regulating inflammation, being highly expressed in monocytes/macrophages (Ito et al., 2015). BTK has a proposed role in signal transduction downstream of several toll-like receptors (TLR) which results in reduced activation of NF-κB (Ní Gabhann et al., 2014) and also regulates NLRP3 inflammasome assembly in macrophages (Ito et al., 2015). Importantly, BTK is not expressed in hepatocytes or adipocytes, the other cell types responsible for peripheral insulin resistance (Uhlén et al., 2015) and so has a specific role in macrophages, the key cell type that drives the production of pro-inflammatory mediators that leads to the development of metabolic syndrome and insulin resistance.
Ibrutinib has therapeutic utility in a number of preclinical models of human disease; however, most of these models have a strong B-cell component (Honigberg et al., 2010;Chalmers et al., 2018). In the present study, we investigated whether treatment with ibrutinib reduces metabolic inflammation in a murine model of HFD-induced obesity. HFD feeding results in an increase in macrophage number in peripheral tissues including the liver, adipose tissue, and kidney. We hypothesized that ibrutinib treatment would target the cells, which express high levels of BTK, namely, monocytes and macrophages, reducing their pro-inflammatory activity in vivo. We predicted that inhibition of BTK signalling would result in decreased activation of both NF-κB and the NLRP3 What is already known • BTK is a druggable target, with FDA approved medications available.

What does this study add
• BTK expression and activation are increased in male C57BL/6J mice fed a high-fat diet.

What is the clinical significance
• We have identified BTK as a new pathway that is activated in metabolic inflammation.
• Ibrutinib may represent a novel candidate for drug repurposing for the treatment of metabolic inflammation. inflammasome in mice fed a HFD. We report for the first time that treatment with a BTK inhibitor results in reduced metabolic inflammation and improved function in the liver, adipose tissue, and the kidney. Animal studies are reported in compliance with the ARRIVE guidelines (Kilkenny et al., 2010) and with the recommendations made by the British Journal of Pharmacology.

| High-fat diet induced insulin resistance
Ten weeks old male C57BL/6J mice purchased from Charles River Ltd, housed in the same unit under conventional housing conditions at 25 ± 2 C, were randomly assigned either normal diet (chow) (7% simple sugars, 3% fat, 50% polysaccharide, 15% protein) or high-fat diet (HFD) (D12331 diet, Research Diet Inc., USA). All mice had access to food and water ad libitum. After 6 weeks of dietary manipulation, mice were randomly assigned to a treatment group receiving either ibrutinib (3 or 30 mgÁkg −1 , in vehicle) or vehicle alone (5% DMSO, 30% cyclodextrin) for 5 days per week for 6 weeks by oral gavage.
The doses used here are within the range that have previously been reported to produce selective inhibition of BTK, in a short-term model of cerebral ischaemia (Ito et al., 2015) and in a chronic model of arthritis (Honigberg et al., 2010). One week prior to the end of the experiment, an oral glucose tolerance test was performed. 24 h before the end of the experiment, mice were placed in metabolic cages and urine collected. Blood was collected by cardiac puncture under general anaesthesia (Ketamine/Xylazine [100 mgÁkg −1 and 10 mgÁkg −1 ; i.p.] [Centaur Services, UK]).

| XID mice
XID mice (B6.CBA-Btk xid /AllmJ) (Lindsley, Thomas, Srivastava, & Allman, 2007) are an inbred strain on the CBA background purchased from The Jackson Laboratory (#009361). They have a point mutation rendering the kinase domain of BTK inactive. Specifically, there is a C to T substitution at coding nucleotide 82, which alters the amino acid sequence; substituting an arginine for cysteine.
The substitution is in a conserved PH domain and blocks the activation of the kinase (Rawlings et al., 1993) preventing BTK phosphorylation at Tyr 223 , which is a key activating site.

| Murine bone marrow-derived macrophages (BMDMs)
Bone marrow-derived macrophages were generated as previously described (Recio et al., 2018). Briefly, fresh bone marrow cells from tibiae and femurs of male C57BL/6 or XID mice aged 8-10 weeks were cultured in DMEM containing 4.5 gÁL −1 glucose, 2 mM L-glutamine, 50 unitsÁml −1 penicillin and 50 μgÁml −1 streptomycin, 10% heatinactivated FBS, 10% L929 cell-conditioned media (as a source of macrophage colony-stimulating factor) and for 7 days. Bone marrow cells were seeded into 8 ml of medium in 90 mm non-tissue culture treated Petri dishes (ThermoFisher Scientific, Sterilin, UK). On Day 5, an additional 5 ml of medium was added. Gentle scraping was used to lift cells. BMDMs were then counted and suspended in FBS free media at the desired cell concentration.
The luminescence was read with a microplate spectrophotometer (PherastarFSX, BMG Lab).
Cycle threshold values were determined by the StepOne software, and target gene expression was normalized to housekeeping gene (β-actin or 18S).

| Quantification of cytokine level
BMDM or hMoDM (1.5 × 10 6 ) were seeded in 6-well plates. BMDM or hMoDM were pretreated with ibrutinib for 1 h prior to stimulation with LPS (100 ngÁml −1 ) for 8 h. Culture medium was adjusted to 5 mM ATP in the last 60 min to allow secretion of IL-1β.

| Bead assay
Measurement of protein levels of secreted IL-1β, TNFα, and IL-6 in cell supernatants from 1.5 × 10 6 human MoDM's was performed by Legend Plex multi-analyte assay kit (BioLegend) according to the manufacturer's instruction. Data were acquired using a BD Fortessa X20 cytometer (BD Biosciences) and analysed using LegendPlex Software (BioLegend).

| Western blot
The immuno-related procedures used in this study comply with the recommendations made by the British Journal of Pharmacology . Tissues (liver and kidney) were lysed, as previously described (Purvis et al., 2017). Briefly, samples (100 mg) of of snap frozen tissue was suspended in 0.3 ml of RIPA buffer containing phosphatase and protease inhibitors (Sigma, UK) for 30 mins and cells lysed by mechanical disruption, then centrifuged at 10,000g for 15 mins at 4 C and the supernatant collected, Protein concentration was determined by using a BCA protein assay kit (ThermoFisher Scientific). Total cell protein (30 mg) was added to 4× Laemmli buffer (250 mM Tris-HCl, pH 6.8, 8% SDS, 40% glycerol, 0.004% bromophenol blue, 20% β-mercaptoethanol) and AB_2576217) (diluted 1:500 in 4% BSA/PBS for 1 hour at room temperature. DAPI was used for nuclear counterstaining. The slide was mounted with Fluormount-G ® , and cells were visualized using a confocal microscope.

| Oil Red O staining
Frozen liver samples were embedded in OCT and cut in 10 μm sections. Sections were brought to room temperature, fixed with 10% buffer formalin for 5 min, washed with 60% isopropanol, then saturated with Oil Red O (1% w/v, 60% isopropanol) for 15 min, washed in 60% isopropanol and rinsed in distilled water. Then sections were then mounted in aqueous mounting medium with coverslips. Images were acquired using a NanoZoomer Digital Pathology Scanner (Hamamatsu Photonics K.K., Japan) and analysed using the NDP Viewer software.

| Periodic acid Schiff's staining
Kidney samples were obtained at the end of the experiment and fixed in 10% neutral-buffered formalin for 48 h, and histology staining was performed. Briefly, kidney tissue was embedded in paraffin and processed to obtain 4 μm sections. After deparaffinization, sections were rehydrated through graded alcohol to distilled water. The sections were then incubated to saturation in periodic acid Schiff's solution (Sigma, UK) for 30 min and washed in distilled water. Then sections were then dehydrated through graded alcohols and cleared before mounting with coverslips. Images were acquired using a NanoZoomer Digital Pathology Scanner and analysed using the NDP Viewer software .

| Immunohistochemistry
Kidney sections cut at 4 μm were deparaffinized to PBS. Antigen retrieval was performed by in citrate buffer (pH 6.0) for 15 min. Once Statistical analysis was only undertaken for studies where each group size was at least n = 5. For western blot analysis, some representative data are shown where group size is less then n = 5. When the mean of two experimental groups were compared, a two-tailed Students t-test was performed. Normally distributed data without repeated measurements were assessed by a one-way ANOVA followed by Bonferroni correction if the F value reached significance. In all cases a P < 0.05 was deemed significant.

| Data and resource availability
The datasets generated during and/or analysed during the current study are available from the corresponding author upon reasonable request. No novel resources were generated during the current study.

| Nomenclature of targets and ligands
Key protein targets and ligands in this article are hyperlinked to corresponding entries in the IUPHAR/BPS Guide to PHARMACOLOGY (http://www.guidetopharmacology.org) and are permanently archived in the Concise Guide to PHARMACOLOGY 2019/20 (Alexander, Fabbro, et al., 2019a, b;Alexander, Kelly, et al., 2019a, b).

| RESULTS
3.1 | Treatment with ibrutinib reduces the diabetic phenotype in a murine model of HFD feeding C57BL/6J mice were fed either chow or HFD for 6 weeks and then treated with ibrutinib (3 or 30 mgÁkg −1 ; p.o.; five times per week) or vehicle for a further 6 weeks. At the time of treatment, mice fed a HFD had developed a small, but significant, augmentation in oral glucose tolerance test (OGTT) ( Figure S1A, B). Importantly, by week 12 of HFD feeding, control mice fed a HFD gained more weight than mice fed a chow diet, specifically they gained more fat mass (Table 1).
Chronic treatment with ibrutinib (3 or 30 mgÁkg −1 ) to chow or HFDfed mice did not alter calorific intake, weight gain or cause hepatocellular damage (Figure S1C-F; Table 1). When compared to mice fed a chow diet, mice fed a HFD and treated with vehicle exhibited a significant elevation in fasting blood glucose (values at time=0; Figure 1a)).
Mice fed a HFD and treated with either 3 or 30 mgÁkg −1 ibrutinib had significantly lower fasting blood glucose, compared with that in mice fed a HFD and treated with vehicle (values at time=0; Figure 1a).
When compared to mice fed a chow diet that underwent an oral glucose tolerance test (OGTT), mice fed a HFD exhibited a significant and prolonged elevation in blood glucose levels (Figures 1a, b). HFD mice also had elevated terminal plasma insulin levels ( Figure 1c

| Inhibition of BTK with ibrutinib reduces inflammation in the liver of HFD fed mice
We next wanted to investigate if the improvements in the diabetic phenotype seen in mice treated with ibrutinib could be attributed to reduced activation of pro-inflammatory pathways in the liver. BTK has been implicated in pro-inflammatory signalling; therefore, inhibiting BTK in infiltrating monocytes/macrophages could protect against the development of insulin resistance, through reduced production of soluble inflammatory mediators. When compared to mice fed chow diet, liver tissue from mice fed a HFD exhibited a significant increase in the phosphorylation of Tyr 223 and hence activation of BTK (Figure 3a,b), resulting in increased downstream signalling through PLCγ (Figure 3a,b). Mice fed a HFD treated with ibrutinib have significantly reduced activation of BTK and downstream effector PLCγ (Figure 3a,b). 3.3 | Treatment with ibrutinib reduces inflammation in adipose tissue of HFD fed mice Following HFD feeding, there is a significant expansion of peripheral adipose tissue deposits that is typically associated with an increase in the adipose monocyte/macrophage population. Mice fed a HFD had significantly larger fat pads compared to chow fed mice as expected (Table 1). Treatment with ibrutinib attenuated the increases in weight of inguinal and epididymal fat pads compared to vehicle treated control (Table 1). We also show that mice fed a HFD have high expression of monocyte/macrophage chemoattractants (CXCL1, CCL2, and CCL5) in the epididymal fat and in the inguinal fat pads (Figure 4a,c), which was attenuated by ibrutinib treatment (Figure 4a

| Ibrutinib treatment reduces NF-κB and NLRP3 inflammasome activation in murine and human macrophages
Macrophages are hypothesized to be a key cell type involved in the local amplification of metabolic inflammation in peripheral tissues following HFD feeding in this study. Therefore, we wanted to directly confirm that ibrutinib inhibits the activation of both NF-

| DISCUSSION
In this study, we showed for the first time that BTK was activated as a result of HFD feeding in a model of diet-induced metabolic inflammation. BTK has previously been shown to regulate pro-inflammatory pathways including NF-κB and the NLRP3 inflammasome. We, therefore, set out to investigate if the anti-cancer agent ibrutinib could represent a novel drug repositioning opportunity for the treatment of HFD induced metabolic inflammation. The main findings of this preclinical study are that chronic treatment of mice fed a HFD with ibrutinib inhibits the activation and downstream signalling of BTK and inhibits the activation of NF-κB and the NLRP3 inflammasome in the liver and kidney. Reduced metabolic inflammation in the liver and adipose tissue resulted in improved glycaemic control via restoration of IRS-1/Akt/GSK-3β signalling pathway in the liver. These improvements in glycaemic control were independent of changes in body weight or calorific intake. Importantly, treatment with ibrutinib also protected mice from the development of hepatosteatosis and proteinuria. We demonstrated that the expression of BTK significantly correlated with expression of CD68 a marker of infiltrating monocytes/macrophages, while ibrutinib treatment significantly reduced the expression of monocyte/macrophage chemoattractants in the liver, adipose, and kidney. We also showed, in primary murine and human macrophages, that ibrutinib treatment reduced LPSstimulated NF-κB activation and attenuated caspase-1 activity, resulting in a reduction in pro-inflammatory gene expression and a reduction in secreted cytokines (IL-1β and TNFα). Critically, our work has identified a novel therapeutic arget in diet induced inflammation, namely, monocyte/macrophage expressed BTK. Importantly, we provide "proof-of-concept" data that ibrutinib reduces HFD-induced inflammation and could be a candidate for repurposing to treat inflammation in metabolic diseases. Mice with myeloid specific deletion of IKK-β fed a HFD do not develop insulin resistance, while mice fed a HFD with hepatocyte specific deletion of IKK develop less severe insulin resistance (Arkan et al., 2005;Ke et al., 2015). However, deletion of NF-κB activity in skeletal muscle does not protect mice from the development of insulin resistance (Röhl et al., 2004). Taken together these published studies demonstrate that while systemically inhibiting the NF-κB pathways does prevent the development of insulin resistance, it is the myeloid specific inhibition that is the critical cellular target in vivo. We would postulate that the BTK expressed in monocytes/macrophages within metabolic tissues (liver, adipose tissue and kidney) is the pharmacological target of ibrutinib, because BTK expression in monocytes/macrophages is significantly higher than in hepatocytes or adipocytes. Myeloid activation of NF-κB is correlated with functional decline in models of HFD induced obesity and insulin resistance.
Indeed, here we show that ibrutinib treatment of mice fed a HFD results in reduced activation of the NF-κB pathway in both the kidney and liver, which as a consequence of HFD feeding have increased accumulation of macrophages. Activation of NF-κB results in the production of key components of the NLPR3 inflammasome pathway including pro-caspase 1 and pro-IL-1β. Ablation of the NLRP3 inflammasome has been shown to improve glycaemic control in obese mice by reducing caspase-1 cleavage and Il-1β/IL-18 activation (Vandanmagsar et al., 2011). Here, we have demonstrated that treatment with ibrutinib of mice fed a HFD attenuated the formation of the NLRP3 complex and blocked the proteolytic cleavage of pro-caspase 1 to active caspase 1 resulting in attenuated production of IL-1β in the liver and kidney.
One important consequence of having reduced metabolic inflammation is improved glycaemic control. We demonstrate that ibrutinib treatment of mice fed a HFD improved glycaemic control (measured as OGTT) and lowered both terminal blood glucose and insulin levels. Mechanistically, we also show that treatment with ibrutinib of mice fed a HFD decreased the phosphorylation of Ser 307 on IRS-1 and increased activation of Akt and GSK-3β in the liver. It should be noted that although phosphorylation of Ser 307 is a widely used marker of insulin resistance, it is not causative in vivo (Copps et al., 2010). Some studies have reported that IKK-β can also regulate insulin sensitivity through direct phosphorylation of IRS-1 suggesting that there may also be a transcriptionindependent mechanism of the NF-κB pathways in our model (Arkan et al., 2005). Indeed, we show that mice fed a HFD and treated with ibrutinib have less activation of IKK-β, which correlates with a reduction in IRS-1 phosphorylation. Thus, we have demonstrated for the first time a mechanistic link between BTK inhibition and IKK-β/IRS-1 signalling in vivo.
Even with careful management of blood glucose and strategies to lower circulating lipids, microvascular complications develop over time in patients with T2D. These complications occur predominantly in tissues where glucose uptake is insulin-independent, such as the kidney, retina and vascular endothelium, as these tissues are exposed to glucose levels close to blood glucose levels. The extent of inflammatory cell accumulation in the diabetic kidney has been associated with the decline of renal function, suggesting a causative link (Macisaac, Ekinci, & Jerums, 2014). We speculated that BTK is a pivotal kinase in tissue resident macrophages needed for the production of chemokines that recruit immune cells to sites of inflammation. Tissue levels of key chemokines were reduced in the liver, adipose tissue and kidney in mice fed a HFD and treated with ibrutinib, including the chemokine CCL2 which is pivotal for monocyte/macrophage recruitment. In many preclinical models of diet induced obesity and microvascular disease genetic deletion or pharmacological inhibition of CCR2 dramatically reduces macrophage recruitment and protects the kidney from functional decline (Awad et al., 2011;Sayyed et al., 2011). In the present study, ibrutinib treatment significantly reduced macrophage accumulation due to lower expression of CCL2 mRNA in the kidney of mice fed a HFD.
These results are consistent with previous data generated in experimental models of diabetic nephropathy, where macrophage accumulation is reduced in Ccl2 −/− mice. Here, inhibition of BTK with ibrutinib reduced both NF-κB and NLRP3 inflammasome activation in the kidney, reduced classical histological markers of early diabetic nephropathy, and protected mice from the development of proteinuria. These data are consistent with other studies reporting that targeting the inflammatory component of early stage diabetic kidney disease improves functional outcome (Anders, 2016). Indeed, individually targeting NF-κB and the NLRP3 inflammasome have been shown to be efficacious in protecting against microvascular disease in diabetes. While this accumulation of evidence from genetic modifications in preclinical models validated this pathway as a therapeutic target, our study is unique in identifying an upstream molecular target and demonstrating the efficacy of our approach using an orally available, drug-repurposing candidate. Using an FDA approved medicine, ibrutinib, we decreased NF-κB and NLRP3 inflammasome driven inflammation, improved glycaemic control and elicited protection from microvascular damage in this model of dietinduced metabolic inflammation.
In conclusion, the concept that the development of T2D is intrinsically linked to the extent of myeloid cell activation is emerging in the field of diabetic medicine. Here, we demonstrate that BTK expression and activation are increased in mice fed a HFD, and this significantly correlated with monocyte/macrophage infiltration into the liver and kidney. Therapeutic treatment with the FDA approved BTK inhibitor ibrutinib not only reduced the extent of monocyte/macrophage infiltration into the liver and kidney of mice fed a HFD but also attenuated the activation of NF-κB and the NLRP3 inflammasome.
Critically, we also demonstrated that treatment with ibrutinib reduced systemic inflammation and resulted in improved glycaemic regulation, lowering blood glucose and insulin levels, by restoring signalling through IRS-1/Akt/GSK-3β. Improved glycaemic control also protected mice fed a HFD from the development of hepatosteatosis and proteinuria. Taken together, we have identified an FDA approved medication that has many of the ideal properties of a candidate medication for repurposing into the treatment of metabolic inflammation, in diseases such as Type 2 diabetes.