The Mincle ligand trehalose dibehenate differentially modulates M1‐like and M2‐like macrophage phenotype and function via Syk signaling

Abstract Introduction Macrophages play a significant role in the progression of diseases, such as cancer, making them a target for immune‐modulating agents. Trehalose dibehenate (TDB) is known to activate M1‐like macrophages via Mincle, however, the effect of TDB on M2‐like macrophages, which are found in the tumor microenvironment, has not been studied. Methods qRT‐PCR, flow cytometry, cytokine ELISA, and Western Blotting were used to study the effect of TDB on GM‐CSF and M‐CSF/IL‐4 derived bone marrow macrophages (BMMs) from C57BL/6 and Mincle−/− mice. Results TDB treatment up‐regulated M1 markers over M2 markers by GM‐CSF BMMs, whereas M‐CSF/IL‐4 BMMs down‐regulated marker gene expression overall. TDB treatment resulted in Mincle‐independent down‐regulation of CD11b, CD115, and CD206 expression by GM‐CSF macrophages and CD115 in M‐CSF/IL‐4 macrophages. GM‐CSF BMMs produced of significant levels of proinflammatory cytokines (IL‐1β, IL‐6, TNF‐α), which was Mincle‐dependent and further enhanced by LPS priming. M‐CSF BMMs produced little or no cytokines in response to TDB regardless of LPS priming. Western blot analysis confirmed that the absence of cytokine production was associated with a lack of activation of the Syk kinase pathway. Conclusion This study illustrates that TDB has the potential to differentially regulate M1‐ and M2‐like macrophages in the tumor environment.

The majority of studies on the activation of macrophages by TDM and TDB have focussed solely on inflammatory macrophages and dendritic cells (DCs) [13,18,20,24,25]. Macrophages, however, exist as a continuum of phenotypes ranging from the classically activated ''M1-like'' (inflammatory) to the alternatively activated ''M2-like'' (anti-inflammatory/wound healing) macrophage [26]. Moreover, they display tremendous plasticity and readily alter their phenotype following exposure to changes in the local microenvironment [27]. Understanding how these different phenotypes respond to agents such as TDB is key to understanding how best to exploit these cells in therapy.
In diseases such as cancer, different macrophage populations display distinct characteristics with often opposing immunological functions [28,29]. Macrophage progenitors exposed to a variety of immune regulatory cytokines, including IL-4, can differentiate into alternatively activated (M2-like) macrophages with tumor-promoting properties [30][31][32]. Thus, the ability to deplete tumor-associated macrophages (TAMs) or convert them from an M2-like phenotype to a more tumor-suppressive phenotype represents a promising new approach for anti-cancer therapies [32]. As TDB and related trehalose glycolipids exhibit anti-cancer activities and have been found to be effective adjuvants [9][10][11][12][13][14][15], we compared the effect of TDB on M1and M2-like macrophages.

Materials and Methods
Animals C57BL/6 wild-type mice and Mincle À/À mice were bred and housed in a conventional animal facility at the Malaghan Institute of Medical Research, Wellington, New Zealand. All animals used for the experiments were aged between 8 and 12 weeks. All experimental procedures were approved by the Victoria University Animal Ethics Committee in accordance with their guidelines for the care of animals (protocol nr 22371).

Generation and stimulation of bone marrow-derived macrophages
Bone marrow cells were collected from the tibia and femur of C57BL/6 or Mincle À/À mice and cultured (250,000 cells/mL) in complete RPMI media (RPMI-1640 [Gibco, UK] with 10% heat inactivated fetal bovine serum [Gibco], 100 U/mL penicillin-streptomycin [Gibco], and 2 mM Glutamax [Gibco]). Macrophage differentiation was induced by either 50 ng/mL GM-CSF (PeproTech, Israel) or 10 ng/mL M-CSF (PeproTech) with 10 ng/mL IL-4 (PeproTech) added to the cRPMI [33,34]. Cells were incubated at 378C (5% CO 2 ) for 8 days (cells fed on days 3 and 6). Where indicated, BMMs were primed with 0.5 ng/mL LPS on day 7 for 24 h. On day 8, the media was removed and the cells were washed with DPBS to remove all non-adherent and loosely adherent cells, and fresh complete RPMI was added to the cells followed by stimulation with 40 or 100 mg/mL TDB (2.5 mg/mL stock in DPBS with 2% DMSO), or 100 ng/mL LPS as positive control. TDB was synthesized according to previously published procedures [35] and was confirmed to be free of endotoxin at a sensitivity of 0.125 EU/mL by the Limulus amebocyte lysate (LAL) assay using an endotoxin kit (Pyrotell, MA, USA).

Cytokine analysis
Levels of IL-1b, IL-6, and TNF-a cytokines in the supernatants were determined by sandwich ELISA (BD Biosciences, CA, USA) according to the manufacturer's instructions.

Quantitative RT-PCR
GM-CSF and M-CSF/IL-4 BMMs were differentiated over 8 days and stimulated with 100 mg/mL TDB for 48 h. Total RNA was extracted using Quick-RNA TM MiniPrep kit (Zymo Research, CA, USA) followed by cDNA synthesis using iScript (BioRad, CA, USA) according to the manufacturer's instructions. Quantitative RT-PCR of 18s, Cd74, Nos2, CD86, Chil3, Retnla, and Mrc1 (QuantiTect primer assay [Qiagen, Germany] and KAPA SYBR FAST qPCR Master Mix [x2] [Kapa Biosystems, MA, USA]) was performed using ABI 7500 platform. Cycle threshold (CT) was determined in the exponential phase of the amplification curve and CT of Cd74, Nos2, CD86, Chil3, Retnla, and Mrc1 were normalized to the CT of 18s ribosomal RNA (DCT). Amplification efficiency of QuantiTect primers are equivalent, so the DDCT method was used to determine fold change (2 ÀDDCT ). Results are expressed as log2 (fold change). All experiments were performed in triplicate.

Statistics
Statistical significance of differences was assessed using two tailed Student's t-tests, 2-way ANOVA with Bonferroni posthoc test, where appropriate, using Prism v7 software (GraphPad, CA, USA). A P value less than 0.05 was considered statistically significant.

TDB activates GM-CSF macrophages and depolarizes M-CSF/IL-4 macrophages
Macrophages often require a priming signal to boost their response to inflammatory stimuli [41]. As TDB can deliver both priming and stimulatory signals [13,20,42], in this study TDB was added to the GM-CSF and M-CSF/IL-4 BMMs without separate priming. Following TDB treatment, GM-CSF BMMs exhibited a shift toward a more proinflammatory phenotype, as indicated by the increased expression of iNOS and CD86 and the decreased expression of the M2 markers Fizz1 and CD206 ( Fig. 2A). In contrast, M-CSF/IL-4 BMMs showed decreased expression of both M1-and M2-like mRNA markers after TDB treatment (Fig. 2B).
Cell surface marker analysis showed that TDB treatment resulted in a significant decrease in the expression of CD11b for GM-CSF BMMs (Fig. 2C) and lowered the expression of CD115 (Fig. 2D) for both macrophage phenotypes. Further analysis of CD115 expression on both GM-CSF and M-CSF/ IL-4 BMMs revealed a discrete CD115 high subpopulation (Fig. 2E) that decreased after exposure to TDB (Fig. 2F). Consistent with the mRNA results, TDB treatment also down regulated the relative expression of the M2-like marker CD206 on GM-CSF BMMs, but had no effect on the expression of CD206 on M-CSF/IL-4 BMMs (Fig. 2G). Although the activation of antigen presenting cells due to the uptake of mycobacteria is commonly associated with an increased expression of MHC class II and co-stimulatory molecules such as CD86 [43], TDB treatment did not alter the cell surface expression of MHC II or CD86 on either macrophage phenotype (Fig. 2H). TDB also had no effect on the expression of another commonly used murine macrophage marker F4/80 (Fig. 2I).
Next, we looked at the effect of TDB treatment on changes to the cell surface expression of Mincle. Consistent with earlier findings [13,18], TDB stimulated a rapid increase in Mincle expression on GM-CSF BMMs (Fig. 3A). In contrast, the stimulation of M-CSF/IL-4 BMMs with a high concentration of TDB only led to a slight increase in Mincle expression over time. To determine whether the decreased expression of CD11b, CD115, and CD206 following TDB treatment required Mincle, BMMs were generated from Mincle À/À bone marrow cells. The down regulation of CD115 expression for GM-CSF and MCSF/IL-4 BMMs (Fig. 3B) and CD11b for GM-CSF BMMs (Fig. 3C) was still observed, indicating that Mincle was not necessary for inducing these effects. In the absence of Mincle, however, GM-CSF BMMs did not down-regulate CD206 following stimulation with TDB (Fig. 3D).

TDB triggers pro-inflammatory activation of GM-CSF BMMs, but not M-CSF/IL-4 BMMs
We next explored the effect of TDB on the function of GM-CSF and M-CSF/IL-4 BMMs. In accordance with previous studies using M1-like macrophages [13,20,35], the exposure of GM-CSF BMMs to TDB triggered the production of IL-1b, IL-6, and TNF-a (Fig. 4A). As expected [13], this activity was abolished when using GM-CSF BMMs from Mincle À/À mice (Fig. 4B). Consistent with previous studies [23,43], M-CSF-differentiated macrophages responded to TDB (Fig. SI1), however, TDB treatment induced little or no cytokine production by M-CSF/IL-4 polarized macrophages despite these cells still being able to respond to LPS (Fig. 4A). This observation is supported by earlier findings that show that IL-4 downregulates the expression of Mincle in human and mouse macrophages and DCs [44]. Priming of the macrophage populations with LPS enhanced TDB-induced increases in both Mincle expression (Fig. 5A) and cytokine production by GM-CSF macrophages, but did not affect the response of M-CSF/IL-4 macrophages (Fig. 5B-D). Priming of the macrophages with LPS also did not enhance the response of either macrophage phenotype to LPS (Fig. 5B-D).
Phosphorylation of Syk [45] and the downstream formation of the Card9-Bcl10-Malt1 complex are essential for the activation of NFkB-mediated gene expression via Mincle [20,23], which in turn leads to pro-inflammatory cytokine production and Mincle up-regulation. To gain insight into the differences in cytokine production between the GM-CSF and M-CSF/IL-4 macrophage populations, we analyzed signaling through the Syk-Card9 pathway by western blot. Analysis of whole cell lysate showed comparable Syk expression in both GM-CSF and M-CSF/IL-4 BMMs, however, only GM-CSF BMMs showed Syk phosphorylation (Y317 and Y519/520) after TDB treatment ( Fig. 6A and B). The tyrosine phosphatase SHP-2 has been shown to mediate C-type lectin receptorinduced Syk activation [46], and western blot analysis showed that TDB treatment induced SHP-2 phosphorylation (Y542 and Y580) in GM-CSF BMMs alone

K. Kodar et al.
TDB differentially alters macrophage phenotype ( Fig. 6A and C). These data indicate that M-CSF/IL-4 BMM do not switch on the Syk-Card9 pathway required to induce pro-inflammatory responses via Mincle.

Discussion
M1-and M2-like macrophages display distinct characteristics and opposing immunological functions that can be targeted to treat disease [26]. In this study, we demonstrate that TDB differentially modulates GM-CSF (M1-like) and M-CSF/IL-4 (M2-like) macrophages. The effect of TDB on the two macrophage phenotypes is substantial, with TDB enhancing the M1-like phenotype of GM-CSF macrophages and down-regulating M2-like markers commonly associated with TAMs and cancer progression [31,47,48] for both GM-CSF and M-CSF/IL-4 BMMs. TDB differentially alters macrophage phenotype K. Kodar et al.

508
The ability of TDB to decrease the expression of CD115 may play a role in the reported anti-cancer activity of the glycolipid. The accumulation of M2-like TAMs is associated with a negative outcome in cancer [32], and accordingly, there has been much interest in targeting macrophages to block the development of this M2-like phenotype [49,50]. Indeed, the inhibition of CD115 by monoclonal antibodies has been shown to inhibit TAMs and has proven to be a promising new anti-cancer therapy [51][52][53][54]. Similarly, the observed decrease in CD11b following TDM treatment could also contribute to the anti-cancer effects of TDB. CD11b has been correlated to a poor prognosis for certain cancers [55], and blocking CD11b has led to enhanced antitumor therapies and outcomes [56,57]. The observed changes in the expression of CD115 and CD11b were independent of Mincle. We have previously reported that the cellular uptake of TDB is Mincle independent [58], and therefore it is possible that TDB is acting intracellularly to down regulate surface marker expression. Alternatively, other, yet unidentified, TDB receptors might be involved in these processes.
Consistent with previous work on M1-like macrophages and DCs [13,20], TDB-treated GM-CSF BMMs produced IL-1b, IL-6, and TNF-a in a Mincle-dependent manner. Similarly, M-CSF BMMs also produced cytokines in response to TDB [23,43]. This proinflammatory response by GM-CSF (M1-like) macrophages, and M-CSF BMMs, likely contributes to TDB's anti-cancer properties as the immunological activity of both TDM and TDB has been linked with inflammation and lymphocyte sensitization [10]. Notwithstanding, Mincle signaling has recently been associated with oncogenesis in pancreatic ductal adenocarcinoma (PDA) [59]. However, PDA is characterized by inflammatory cells in the tumor microenvironment [60], while many other cancers contain high numbers of anti-inflammatory macrophages [32]. Accordingly, it is necessary to understand the tumor microenvironment and how different tumor macrophage populations are likely to respond to immunomodulating agents such as TDB in order to ensure that the appropriate local response is raised for the best therapeutic effect.
In contrast to GM-CSF and M-CSF BMMs, M-CSF/IL-4 BMMs did not produce inflammatory mediators in response to TDB despite showing the capacity to respond to LPS stimulation. Subsequent analysis of the cell surface expression levels of Mincle indicated that TDB induced rapid and significant increases in Mincle expression in GM-CSF BMMs but not for M-CSF/IL-4 BMMs, which is consistent with the earlier findings by Hupfer et al. [44]. Moreover, while LPS priming increased Mincle expression and cytokine production by GM-CSF BMMs treated with TDB, LPS priming had no appreciable effect on Mincle expression or cytokine  production by strongly M2-polarized macrophages treated with TDB.
Our data on the activation of the Mincle-Syk pathway also demonstrates that while both Syk and SHP-2 proteins were phosphorylated in TDB-treated GM-CSF BMMs, this did not occur in M-CSF/IL-4 BMMs, with the likely explanation being the observed low expression levels of Mincle. However, based on work by Strasser et al. [61] showing that protein kinase C d (PKCd) links Syk activation by C-type lectin receptors (including Mincle) to the Card9-Bcl10-Malt1 complex, a phosphoproteomic study could provide more insight into the Mincle-driven signaling pathways for both macrophage phenotypes.
Taken together, our data demonstrates that TDB can modulate the immune response of M1-like and M2-like macrophages, with both macrophage subsets losing characteristics associated with an M2-like phenotype. These functional changes are more pronounced for macrophages already skewed toward the pro-inflammatory phenotype, with GM-CSF BMMs becoming more pro-inflammatory upon stimulation with TDB. The treatment of M-CSF/IL4 BMMs with TDB, generates a more neutral macrophage phenotype, as evidenced by the decrease in the relative expression of CD115, deficit cytokine production, and an overall decrease in the expression of marker mRNA. Our data also suggests that TDB has the potential to alter macrophage phenotypes to one that disfavors tumorgrowth. Given that TDB is a well-tolerated vaccine adjuvant, this work further supports the development of modified trehalose glycolipids with even more promising as anticancer activity.

SUPPORTING INFORMATION
Additional supporting information may be found in the online version of this article at the publisher's web-site. Figure SI1. TDB induces the production of cytokines in M-CSF differentiated bone marrow macrophages. BMMs from WT bone marrow were differentiated over 8 days using M-CSF followed by stimulation with 40 mg/mL TDB, or 100 ng/ mL LPS as positive control. Levels of IL-1b, IL-6, and TNFa were measured by the ELISA from supernatant at 48 h. Mean AE SEM of triplicate samples from one experiment are shown. Ã P 0.05; ÃÃÃÃ P 0.001 (1-way ANOVA).