Ultrasound reduces inflammation by modulating M1/M2 polarization of microglia through STAT1/STAT6/PPARγ signaling pathways

Abstract Introduction Activated microglia can be polarized to the pro‐inflammatory M1 phenotype and the anti‐inflammatory M2 phenotype. Low‐intensity pulsed ultrasound (LIPUS) can attenuate pro‐inflammatory responses in activated microglia. Objective This study aimed to investigate the effects of LIPUS on M1/M2 polarization of microglial cells and the regulatory mechanisms associated with signaling pathways. Methods BV‐2 microglial cells were stimulated by lipopolysaccharide (LPS) to an M1 phenotype or by interleukin‐4 (IL‐4) to an M2 phenotype. Some microglial cells were exposed to LIPUS, while others were not. M1/M2 marker mRNA and protein expression were measured using real‐time polymerase chain reaction and western blot, respectively. Immunofluorescence staining was performed to determine inducible nitric oxide synthase (iNOS)‐/arginase‐1 (Arg‐1)‐ and CD68‐/CD206‐positive cells. Results LIPUS treatment significantly attenuated LPS‐induced increases in inflammatory markers (iNOS, tumor necrosis factor‐α, interleukin‐1β, and interleukin‐6) as well as the expression of cell surface markers (CD86 and CD68) of M1‐polarized microglia. In contrast, LIPUS treatment significantly enhanced the expression of M2‐related markers (Arg‐1, IL‐10, and Ym1) and membrane protein (CD206). LIPUS treatment prevented M1 polarization of microglia and enhanced or sustained M2 polarization by regulating M1/M2 polarization through the signal transducer and activator of transcription 1/STAT6/peroxisome proliferator‐activated receptor gamma pathways. Conclusions Our findings suggest that LIPUS inhibits microglial polarization and switches microglia from the M1 to the M2 phenotype.


| INTRODUC TI ON
Neuroinflammation is one of the major neuropathological characteristics of neurodegenerative diseases, such as Alzheimer's disease and Parkinson's disease. 1,2Clinical failures in treating these diseases underline the obstacles to drug development, including delivery of a high enough drug concentration to the brain, because the bloodbrain barrier (BBB) often prevents therapeutic levels from being achieved.Indeed, only a fraction of the drug that enters the brain reaches microglia.Therefore, the continuing lack of effective drugs to treat these diseases highlights the unmet need for developing innovative treatments. 3,4method of manipulating the microglial phenotype could potentially be an effective treatment for multiple human neurodegenerative diseases.[7] Moreover, PPARγ activation has a protective effect by attenuating inflammation in the central nervous system (CNS).In previous studies, low-intensity pulsed ultrasound (LIPUS) attenuated proinflammatory responses as well as the decline in brain-derived neurotrophic factor (BDNF) in LPS-treated microglia. 8Moreover, LIPUS stimulation induced neuroprotection by inhibiting LPS-induced activation of TLR4/NF-κB inflammatory signaling and by enhancing the associated CREB/BDNF expression in LPS-treated mice. 9Zhang et al. showed LIPUS induced the polarization change of inflammatory macrophages and resident macrophages in a rat model of spinal fusion. 10Nevertheless, whether LIPUS is involved in regulating microglial polarization and its molecular mechanisms is still unknown.
Microglia, the resident macrophages of the CNS, maintain the neural environment and respond to neural damage and repair by switching between different activation states. 11Inflammatory responses in the brain, which are indicated by changes in the properties of microglia, play a crucial role in multiple CNS diseases. 12croglia are the first line of defense in the brain and have protective functions.They initiate an inflammatory response when a danger signal is detected.This may be triggered by microglial activation and consequent inducible nitric oxide synthase (iNOS) upregulation and secretion of several inflammatory cytokines. 13On the other hand, microglia can produce anti-inflammatory cytokines, inhibit local inflammation, and promote trophic support by secreting neurotrophins. 14Therefore, microglia can be either cytotoxic or neuroprotective depending on their current phenotype.The shifting microglial phenotype has recently become a therapeutic strategy to treat multiple neurodegenerative diseases. 15croglia can be polarized to the classically activated proinflammatory M1 phenotype or the alternatively activated antiinflammatory M2 phenotype. 16M1 and M2 phenotypes are stimulated by LPS and interleukin-4 (IL-4), respectively. 17M1 microglia release high levels of pro-inflammatory mediators, such as tumor necrosis factorα (TNFα), interleukin-1β (IL-1β), and interleukin-6 (IL-6), in addition to increased levels of oxidative metabolites, such as iNOS, and cell surface markers, CD86 and CD68. 18,19These inflammatory factors are mediated by the signal transducer and activator of transcription 1 (STAT1) signaling pathway.In contrast, M2 microglia release beneficial mediators, including IL-10, arginase-1 (Arg-1), Ym1, mannose receptor (CD206), transforming growth factor beta (TGFβ), and several neurotrophic factors, such as BDNF, insulin-like growth factor-1 (IGF-1), and nerve growth factor (NGF), which are mediated by the STAT6 signaling pathway. 16,20,21Furthermore, the M2 microglial phenotype can be categorized into M2a, M2b, and M2c subtypes, which can be distinguished by changes in expression of the relevant factors. 22,23M1/M2 polarization of microglia can be regulated through STAT1 or STAT6 phosphorylation. 24nce LIPUS is a local treatment and is noninvasive, its targeted stimulation to activate microglia at the site of neuroinflammation could be highly beneficial.Therefore, this study aimed to investigate whether LIPUS has the ability to inhibit M1 polarization and promote M2 polarization and subsequently induce an M1-to-M2 phenotypic shift in microglia.The results of our study may have implications for treating inflammation-related neurological diseases.

| Pulsed ultrasound system
LIPUS was generated by a therapeutic ultrasound generator (ME740, Mettler Electronics) and a 1-MHz plane transducer (ME7410, Mettler Electronics; 9.3-cm 2 effective radiating area) with 2-ms burst lengths at a 20% duty cycle and repetition frequency of 100 Hz.The spatial average intensity (I SAI ) over the plane transducer head was 30 mW/cm 2 ; it was measured with a radiation force balance (Precision Acoustics) in degassed water.The ultrasound parameters were selected based on the results of our previous study. 8The LIPUS was transmitted from the plane transducer to the bottom of the cell culture plate.Ultrasound transmission gel (Pharmaceutical Innovations) was used to cover the area between the transducer and the plate to maximize ultrasound transmission.Each microglial cell sample was treated with three episodes of LIPUS stimulation using triple sonication.To reduce the thermal effect of ultrasound, the duration of each sonication was 5 min, with a 5-min interval between two sonications.Thus, the total sonication time of LIPUS stimulation was 15 min.

| Cell cultures
BV-2 cells, derived from primary mouse microglial cells, are the most frequently used substitute for primary microglia.There were no animal subjects in this study.The mouse BV-2 microglial cell line was purchased from American Type Culture Collection (ATCC).The cells were grown on a six-well plate (each well diameter = 34.8mm) at an appropriate density in Dulbecco's Modified Eagle's Medium (Gibco) supplemented with 10% (v/v) heat-inactivated fetal bovine serum, 100 μg/mL of streptomycin, and 100 U/mL of penicillin (Hyclone).
They were maintained in a humidified 5% CO 2 /95% air environment at 37°C.The cells were incubated for a further 24 h in a humidified incubator.A cell density of 1 × 10 5 cells/well was prepared for subsequent experiments.To promote polarization to the M1 or M2 phenotype, microglia were treated with LPS (10 ng/mL) or IL-4 (20 ng/ mL), respectively, for 8 h.BV2 cells were divided into six groups: control, LIPUS, LPS, LPS + LIPUS, IL-4, and IL-4 + LIPUS.The control group received no treatment except for a change of medium.The LIPUS group received the medium and LIPUS.The LPS group and LPS + LIPUS group consisted of microglia stimulated with LPS without and with LIPUS, respectively.The IL-4 group and IL4 + LIPUS group consisted of microglia stimulated with IL-4 without and with LIPUS, respectively.The LPS + LIPUS and IL-4 + LIPUS groups were treated with LIPUS for a sonication time of 15 min 4 h after LPS and IL-4 administration, respectively.

| Real-time polymerase chain reaction
Total RNA was isolated from cells using an RNeasy Mini Kit (QIAGEN) according to the manufacturer's instructions.RNA concentration and purity were determined using NanoDrop ND-1000 (Thermo Fisher Scientific) according to the manufacturer's instructions.
Reverse transcription was performed using a ToolsQuant II Fast RT Kit (BioTools) according to the manufacturer's instructions.RT-PCR assay was performed using SYBR (KAPA SYBR FAST qPCR Master Mix (2×), ABI Prism™) in StepOnePlus™ (Applied Biosystems), according to the manufacturer's instructions.Table S1 summarizes the primers used in this study.β-actin was used to normalize mRNA in the cells.Each sample was analyzed in triplicate for yield validation.
The 2 −ΔΔCt method was used to determine the relative quantification of gene expression levels.

| Immunofluorescence
BV2 cells were incubated with 10 ng/mL of LPS in humidified 5% CO 2 /95% air for 24 h at 37°C.The cells were fixed in 4% paraformaldehyde for 10 min at room temperature and washed three times with PBS.The cells were treated with 0.1% Triton X-100 for 5 min at room temperature.The washed cells were incubated with a blocking serum for 1 h and then incubated overnight with primary antibody

| Statistical analysis
All data are shown as mean ± standard error of the mean.The Shapiro-Wilk test was used to test for normality.One-way ANOVA, followed by Tukey's post hoc test, was used to determine differences between groups.The level of statistical significance was set at p < 0.05.

| Effects of LIPUS, LPS, and IL-4 treatment on microglia
First, we examined the effect of LIPUS on microglia (Figure 1A) as it can increase the expression of neurotrophic factors in microglia. 8When exposed to LIPUS, microglia exhibited significant time-dependent changes in IGF-1 mRNA expression (p < 0.01; Figure 1B).The expression of IGF-1 reached a maximum at 4 h after LIPUS stimulation (p < 0.001; Figure 1B).On the other hand, LIPUS significantly decreased TNFα mRNA expression 4 h after stimulation (p < 0.05; Figure 1C).LIPUS, LPS, IL-4, or a combination of the three did not affect cell viability compared with the control cells (Figure 1D).Second, we examined the effect of LPS on microglia (Figure 1E).iNOS and TNFα mRNA expression showed timedependent changes in LPS-treated microglia (Figure 1F,G).iNOS and TNFα expression reached a maximum at 8 and 2 h after LPS treatment, respectively (p < 0.001; Figure 1F,G).However, no significant difference in Arg-1 expression was observed at any time points in LPS-treated microglia (Figure 1H).Third, we examined the effect of IL-4 on microglia (Figure 1I).Arg-1 and IGF-1 mRNA expression showed time-dependent changes in IL-4-treated microglia (Figure 1J,K).The expression of both Arg-1 and IGF-1 reached a maximum at 8 h after IL-4 treatment (p < 0.001; Figure 1J,K).In contrast, iNOS mRNA expression showed a time-dependent decrease in IL-4-treated microglia (Figure 1L).iNOS expression reached a minimum at 8 h after IL-4 treatment (p < 0.01; Figure 1L).

| LIPUS inhibits LPS-induced M1 markers and selectively promotes IL-4-induced M2 markers
Based on the temporal expression profile shown in Figure 1

| LIPUS inhibits LPS-induced iNOS expression, but not IL-4-induced Arg-1 expression
The differential expression of iNOS and Arg-1 in the M1 profile compared with the M2 profile provides a model for microglial phenotypes.LPS treatment increased the mRNA and protein expression of pro-inflammatory cytokine iNOS (both p < 0.001; Figure 3A,C), indicating the shifting of microglia to the M1 phenotype.LPS treatment also decreased the expression of M2-related Arg-1 in microglia.LIPUS treatment not only significantly inhibited iNOS mRNA and protein expression in LPS-treated microglia (both p < 0.001; Figure 3A,C), but also significantly promoted Arg-1 mRNA and protein expression (both p < 0.01; Figure 3B,D).However, LIPUS treatment did not induce higher Arg-1 mRNA and protein expression in IL-4-treated microglia (Figure 3B,D).Therefore, these data suggest that LIPUS prevents LPS activation of microglial cells and facilitates their polarization to the M2 phenotype.

| LIPUS regulates M1/M2 polarization of microglial cells
To validate the effect of LIPUS treatment on polarization, microglial cells were stimulated by LPS to an M1 phenotype or by IL-4 to an M2 phenotype, and then treated with LIPUS.The levels of expression of M1 and M2 intercellular proteins, iNOS and Arg-1 were detected using immunofluorescence (Figure 4A).Consistent with the mRNA and protein expression shown in Figure 3, iNOS expression was significantly elevated in LPS-treated microglia (p < 0.01; Figure 4B), while it was significantly reduced following LIPUS treatment (p < 0.05; Figure 4B, third and fourth columns).In addition, no change in Arg-1 expression was observed in the microglia treated with LPS, but LIPUS treatment significantly increased Arg-1 in LPStreated microglia (p < 0.05; Figure 4C, third and fourth columns).
However, no significant difference in Arg-1 was found in microglia treated with IL-4 or the combination of IL-4 and LIPUS (Figure 4C, fifth and sixth columns).The levels of expression of M1 and M2 cell surface markers, CD68 and CD206, were also detected using immunofluorescence (Figure 5A).CD68 expression was significantly elevated in LPS-treated microglia (p < 0.01; Figure 5B), while it was reduced following LIPUS treatment.In addition, no change in CD206 expression was observed in microglia treated with LPS, but LIPUS treatment significantly increased CD206 in LPS-treated microglia (p < 0.01; Figure 5C, third and fourth columns).However, no significant difference in CD206 was found in microglia treated with IL-4 or the combination of IL-4 and LIPUS (Figure 5C, fifth and sixth columns).These results suggest that LIPUS reduced M1 polarization of microglia and switched them from M1 to M2 polarization.

| LIPUS induces M2 subtype expression in microglia and regulates M1/M2 polarization via the STAT1/STAT6/PPARγ signaling pathways
CD206, CCL1, and TLR8 are markers for microglial subtypes M2a, M2b, and M2c, respectively. 22All three M2 phenotypes are antiinflammatory repair microglial cells.LIPUS significantly promoted CD206 and TLR8 expression in control, LPS-treated, and IL-4-treated microglia (all p < 0.05; Figure 6A, C), while it did not affect CCL1 expression in the three treatments (all p > 0.05; Figure 6B).STAT1 and STAT6 have a reciprocal inhibitory relationship and regulate the transcription of M1 or M2 genes involved in the polarization of microglia. 16,25To understand the underlying mechanism of the M1/ M2 polarization switch after LIPUS treatment, the effects of LIPUS on STAT1/STAT6 were observed.Microglia were treated with LPS to activate the STAT1 pathway, which is necessary for M1 polarization.The level of phosphorylated (p)-STAT1 protein was significantly increased by LPS and attenuated by LIPUS treatment (Figure 6D,E

| DISCUSS ION
Many drugs inducing the M1-to-M2 phenotype switch in microglia have therapeutic potential for treating neurodegenerative diseases but have failed due to the limited permeability of the BBB. 26,27erefore, there is clinical interest in changing microglial polarization through innovative approaches without the impediment of the BBB.In vivo and in vitro experiments provide increasing evidence of the anti-inflammatory effects of LIPUS on LPS-induced inflammation. 8,9  An initial investigation of time points was performed with LPS, IL-4, or LIPUS to determine the optimal expressive time points of these three stimulators (Figure 1).To mimic a pro-inflammatory (M1like) or an anti-inflammatory (M2-like) classification of microglial activation, we pretreated microglia with LPS or IL-4, respectively, for 4 h.We then stimulated the microglial cells with LIPUS for 15 min, and then collected the samples.The MTT assay showed that LPS, IL-4, and LIPUS at these doses did not induce cytotoxicity (Figure 1D).Subsequently, depending on whether the microglial phenotype was M1 or M2 following treatment with LPS, IL-4, or LIPUS (Figure 3), microglia promoted either iNOS or Arg-1 to metabolize ʟ-arginine for the M1 pathogen-killing response or the M2 wound healing response, respectively. 28These results indicate that LIPUS stimulation switched the production of cytotoxic nitric oxide to metabolization of ʟ-arginine for wound healing in microglia, which (α-tubulin).The cells were then washed and incubated with secondary antibodies for 1 h in a dark room.For nuclear staining, the cells were incubated with 10 μL of DAPI for 10 min.The slide was finally washed and mounted for microscopic examination.The stained cells were detected by fluorescence microscopy.The number of cells positive for α-tubulin was counted in a 0.302-mm 2 area in three non-overlapping fields under 200× magnification.The mean signal intensities for iNOS, Arg-1, CD68, and CD206 were quantified in a 0.302-mm 2 area in three non-overlapping fields with Image-Pro Plus software (Media Cybernetics) in a blinded manner.
Figure 6I, third and fourth columns).Moreover, IL-4 induced a significant increase in PPARγ mRNA expression compared with control cells; this increase was further enhanced in the presence of LIPUS (both p < 0.001; Figure 6I, fifth and sixth columns).Taken together, our results suggest that LIPUS regulates M1/M2 polarization of microglia via the STAT1/STAT6/PPARγ pathways.

F I G U R E 3
Effects of LIPUS treatment on iNOS and Arg-1 expression.Microglia were treated with LPS, IL-4, LIPUS, or a combination of LIPUS and LPS or IL-4.Representative changes in (A) iNOS and (B) Arg-1 mRNA were detected by qRT-PCR at 4 h after treatment with LIPUS.The relative levels of (C) iNOS and (D) Arg-1 protein were detected by western blotting at 4 h after treatment with LIPUS.*p < 0.05; **p < 0.01; ***p < 0.001; N.S. = not significant; n = 8. | 4121 HSU et al. neuroinflammation by inhibiting TLR4/NF-κB signaling and enhancing TrkB/CREB expression in LPS-treated microglia. 9In this study, we demonstrated that noninvasive LIPUS stimulation inhibited M1 polarization and promoted the switch from M1 to M2 polarization in LPS-treated microglia.Our results suggest further development of LIPUS stimulation as a potential treatment for multiple neurodegenerative diseases.