T0901317

Anti‑inflammatory effects of naturally occurring retinoid X receptor agonists isolated from Sophora tonkinensis Gagnep. via retinoid X receptor/liver X receptor heterodimers

Wei Wang1 · Ken‑ichi Nakashima1 · Takao Hirai1 · Makoto Inoue1

Abstract

Retinoid X receptor (RXR) ligands have a wide range of beneficial effects in mouse models of Alzheimer’s disease (AD). Recently accumulated evidence suggests that early neuroinflammation may be a therapeutic target for AD treatment. We therefore investigated the anti-inflammatory effects of the prenylated flavanoids SPF1 and SPF2, which were previously isolated from root of Sophora tonkinensis and identified as potent ligands for RXR, and potential mechanisms involved. SPF1 and SPF2 efficiently reduced interleukin (IL)-1β messenger RNA (mRNA) and IL-6 mRNA levels in lipopolysaccharide- stimulated and tumor necrosis factor-α-stimulated RAW264.7 cells, whereas SPF3—which has a structure similar to SPF1 and SPF2 but no RXR ligand activity—did not exhibit such effects. Intriguingly, the liver X receptor (LXR) ligand T0901317 reduced proinflammatory cytokine mRNA levels, and these effects were potentiated by SPF1. With regard to the mechanism underlying the anti-inflammatory effects, SPF1 induced significant amounts of activating transcription factor 3 (ATF3) mRNA and protein, and this effect was potentiated by T0901317. SPF1 also reduced translocation of nuclear factor κB (NF-κB) into nuclei. The production of proinflammatory cytokines was significantly inhibited by SPF1, and this effect was primarily exerted via RXR/LXR heterodimers. The effects of SPF1 may partly depend on the induction of ATF3, which may bind to the p65 subunit of NF-κB, resulting in reduced translocation of NF-κB into nuclei and reduced NF-κB transcription. Although inflammatory effects mediated by RXR/LXR heterodimers have not been thoroughly investigated, the above-described results shed light on the mechanism of the anti-inflammatory effect via RXR/LXR heterodimer.

Keywords Activating transcription factor 3 · Antiinflammation · Liver X receptor · Naturally occurring agonist · Retinoid X receptor · RXR/LXR heterodimer

Introduction

Inflammation plays a critical role in the onset and progres- sion of numerous conditions, including Alzheimer’s disease (AD), diabetes, and atherosclerosis [1]. Neuroinflammation is most commonly initiated in response to stimuli such as infection [2], brain injury [3], stress [4], or aging [5], and elicits activation of microglia, macrophages, and astrocytes. Although neuroinflammation was initially considered to be a consequence of AD pathology, accumulating evidence now suggests that early disease-exacerbating neuroinflamma- tion may start decades before the manifestation of cognitive decline, and plays a primary causative role in AD pathogen- esis [6]. Moreover, neuroinflammation has different effects on AD pathology in the early and late stages of the condi- tion, and systemic inflammation causes the inflammatory response of the central nervous system (CNS), contributing to chronic neuroinflammation and neurodegeneration [7, 8]. In the context of an urgent need to develop therapeutic agents, early neuroinflammation is considered a potential therapeutic target to treat the onset and progression of AD. Nuclear receptors (NRs), which were initially consid- ered ligand-activated transcription factors comprising a superfamily, regulate biological interactions involved in a wide range of physiological phenomena by controlling gene expression or the functions of transcription products [9, 10]. A growing body of recent evidence suggests that pharma- cological activation of particular NRs may be a promising therapeutic approach in AD [11], and that NR ligands for peroxisome proliferator-activated receptors (PPAR) γ and δ, liver X receptor (LXR), and retinoic acid receptor (RAR) in addition to retinoid X receptor (RXR) ameliorate amy- loid pathology by suppressing expression of proinflamma- tory genes, promoting phagocytosis of plaques, reducing soluble amyloid β (Aβ) levels, and enhancing cognition in various mouse models of AD [12–16]. RXR ligands poten- tially represent therapeutic agents, because RXR serves as the common obligate heterodimer of LXR and PPARs, and ligation of RXR promotes transcription of genes associated with Aβ clearance and abrogation of inflammation in AD mouse models [17–19]. A particularly important regulatory characteristic of RXR heterodimers classified as permis- sive heterodimers such as PPARs/RXR and RXR/LXR is that combined activation with both RXR and the partner NR ligand induces a cooperative or synergistic response greater than that of a single NR ligand, suggesting that RXR permissive heterodimers are capable of sensing even relatively small changes in intracellular molecules derived from diet, associated metabolites, or therapeutic agents, and can cause profound biological responses resulting in main- tenance of homeostasis.
The RXR-selective ligand bexarotene (BEX) can report- edly enhance soluble Aβ clearance, and reverse cognitive decline in mouse AD models [20]. However, the beneficial effects of BEX on Aβ metabolism and plaque burden are controversial [21], because in some in vivo studies there has been no significant relationship between Aβ deposition and cognitive decline [21]. Additionally, in one recent study, BEX suppressed inflammation and astrogliosis, activated microglial phagocytosis [14], and affected signaling path- ways related to neuronal differentiation and neuron projec- tions [22]. Thus, BEX seems likely to have ameliorating effects on cognitive decline or dementia via multifaceted mechanisms including anti-inflammatory effects. However, serious side-effects inherent to BEX treatment including hyperlipidemia [23], hypothyroidism [24], hepatomegaly [25], and anemia [26] have been reported that limit its use for treatment of cutaneous T cell lymphoma.
In previous study, two prenylated flavanoids referred to as SPF1 and SPF2, which were isolated from root of Sophora tonkinensis, exhibited selective RXR agonist activity, with respective half-maximal effective concentration (EC50) val- ues of 0.77 μM and 0.78 μM, and activated PPARs/RXR and RXR/LXR heterodimers alone and in combination with the corresponding PPARs and LXR ligands [27]. Additionally, SPF1 and SPF2 protected sympathetic neuron-like nerve growth factor-differentiated PC12 cells from Aβ-induced apoptosis via activation of RXR/LXR heterodimers, accom- panied by ABCA1 upregulation [28]. Thus, investigation of the anti-inflammatory effects of naturally occurring RXR ligands with properties distinct from BEX is warranted, as are attempts to identify novel alternative RXR ligands with- out severe side-effects.
In the current study, we investigated the anti-inflamma- tory effects of SPF1 and SPF2, and found that they inhibited proinflammatory cytokine production in RAW264.7 cells and peritoneal macrophages by modulating nuclear factor κB (NF-κB) activation via induction of activating tran- scription factor 3 (ATF3) following RXR/LXR heterodimer activation.

Materials and methods

Reagents

2-[{3-Hydr oxy -2 ′ , 2-dime t h y l-8-(3-me t h y l- 2-buten y l)}c hr oman-6-y l]-7-h ydr o xy -8-(3- methyl-2-butenyl)-chroman-4-one (SPF1) and 2-[{2-(1-hydroxy-1-methylethyl)-7-(3-methyl-2-butenyl)-2′, 3-dihydrobenzofuran}-5-yl]-7-hydroxy-8-(3-methyl-2- butenyl)chroman-4-one (SPF2) were isolated from root of Sophora tonkinensis Gagnep. as described in our previous paper [27]. The RXR agonist bexarotene (BEX) and LXR agonist T0901317 (T090) were purchased from Cayman Chemicals (Ann Arbor, MI, USA). All chemicals were dissolved in dimethylsulfoxide (DMSO; Fujifilm Wako Pure Chemical, Osaka, Japan) and stored at −20 °C until use. Dulbecco’s modified Eagle’s medium (DMEM) and lipopolysaccharide (LPS, type B5) were obtained from Sigma-Aldrich (St. Louis, MO, USA). Fetal bovine serum (FBS) was purchased from Nichirei Biosciences Inc. (Tokyo, Japan). Thioglycollate (TGC) medium was obtained from Nihon Pharmaceutical Co., Ltd. (Tokyo, Japan). Tumor necrosis factor-α was purchased from PeproTech Inc. (Rocky Hill, NJ, USA).

Cell culture

Mouse macrophage-like cell line RAW264.7 was provided by the RIKEN Bio Resource Center (Tsukuba, Japan) and maintained in DMEM supplemented with 10 % FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin (DMEM complete medium) at 37 °C in a humidified atmosphere of 5 % CO2 in air.

Animals

Male C57BL/6J mice (6–8 weeks of age) obtained from Japan SLC, Inc. (Hamamatsu, Japan) were housed in cages under controlled environment (12 h light/dark cycle, 23 ± 1 °C) with food and water ad libitum at the Labora- tory Animal Center of School of Pharmacy, Aichi Gakuin University. All animal procedures were approved by the Animal Care and Use Committee for the School of Phar- macy, Aichi Gakuin University.

Isolation of mouse peritoneal macrophages

Mice were injected with 1/10 volume per body weight of 3 % (W/V) TGC medium that had been autoclave-sterilized and stored in the dark for > 1 month. The mice were killed by decapitation 6 days later. Primary peritoneal mac- rophages were obtained immediately from the abdominal cavity, and seeded into plates at density of 1 × 106 cells/ ml. The cells were incubated in DMEM complete medium at 37 °C and 5 % CO2 for 2 h, then nonadherent cells were washed out and adherent macrophages were used in sub- sequent experiments.

Total RNA extraction and real‑time qPCR

RAW264.7 cells were seeded into plates at density of 5 × 105 cells/ml, and treated with compounds for 8 h, fol- lowed by stimulation with LPS (100 ng/ml) for 12 h. After treatment, total RNA was extracted from cells using RNAiso Plus (Takara Bio Inc., Kusatsu, Japan) according to the manufacturer’s instructions. A total of 250 ng RNA was reverse-transcribed using ReverTra Ace quantitative poly- merase chain reaction (qPCR) RT master mix with gDNA remover kit (Toyobo Co. Ltd., Osaka, Japan). Quantitative real-time qPCR with SYBR Green (Thermo Fisher Scien- tific, K.K., Tokyo, Japan) was performed using the TP800 Thermal Cycler Dice® real-time system (Takara Bio Inc., Kusatsu, Japan). qPCR was carried out under the following conditions: 50 cycles of 5 s at 95 °C and 30 s at 60 °C. The relative expression of target gene mRNA was calculated by the ΔΔCt method and normalized to the amount of β-actin expression. The primer sequences are listed in Table 1.

Enzyme‑linked immunosorbent assay

RAW264.7 cells were seeded into plates at density of 5 × 105 cells/ml, then treated with compounds for 8 h, fol- lowed by stimulation with LPS (100 ng/ml) for 12 or 21 h. After treatment, proinflammatory cytokines in the culture medium were assayed using mouse IL-6 enzyme-linked immunosorbent assay (ELISA) kits (Diaclone, France) and mouse IL-1β ELISA kits (R&D Systems Inc., Minneapolis, MN, USA) in accordance with manufacturers’ instructions.

Analysis of nuclear translocation of NF‑κB

RAW264.7 cells were seeded onto cover glasses in the wells of culture dishes at density of 5 × 105 cells/ml, then treated with SPF1 (10 μM) for 8 h, followed by stimulation with LPS (100 ng/ml) for 30 min. The cells were then fixed with 4 % paraformaldehyde at 20 °C for 10 min. After wash- ing with phosphate-buffered saline (PBS), the cells were incubated in blocking buffer (5 % normal goat serum/0.3 % Triton X-100/PBS) for 60 min. They were then incubated with anti-NF-κB p65 subunit antibody (#8242, Cell Signal- ing Technology, 1:400 dilution) in antibody dilution buffer (1 % bovine serum albumin/0.3 % Triton X-100/PBS) at 4 °C overnight, followed by Alexa Flour 488-conjugated secondary anti-rabbit immunoglobulin G (IgG) (#4412, Cell Signaling Technology, 1:1000) at room temperature for 2 h. After thorough washing, the cells were incubated in 4′,6-diamidino-2-phenylindole (DAPI) solution (H-1200, Vector Laboratories, Burlingame, CA, USA) for 10 min in the dark. Fluorescence was visualized using a BZ-9000 fluo- rescence microscope (Keyence Corp., Itasca, IL, USA).

Western blot analysis

RAW264.7 cells were seeded into six-well plates at density of 5 × 105 cells/ml. After treatment with the compounds, total protein was extracted from the cells using radioimmu- noprecipitation assay (RIPA) lysis buffer, then the protein concentration was determined by Bradford assay (Bio-Rad, USA). Protein samples were separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) and transferred to polyvinylidene fluoride (PVDF) mem- branes (Immobilon®, USA). The membranes were blocked with 5 % (w/v) bovine serum albumin (BSA) in TBS-T (Tris-buffer saline containing 0.1% Tween 20) buffer at room temperature for 1 h and incubated with anti-ATF3 antibody (1:1000, Abcam) and anti-β-actin antibody (1:1000, BioLe- gend) at 4 °C overnight. The next day, the membranes were washed with TBS-T three times and incubated with horse- radish peroxidase-linked anti-mouse IgG and anti-rabbit IgG antibodies (1:5000, Cell Signaling Technology) at room temperature for 1 h. Then, the membranes were washed with TBS-T four times, and bands were detected using the enhanced chemiluminescence reagent (ImmunoStar®, Fuji- film Wako Pure Chemical) and imaged by ImageQuant LAS- 4000mini (GE Healthcare, USA).

Statistical analysis

Data are presented as mean ± standard deviation (SD). Sta- tistical significance was analyzed by one-way analysis of variance (ANOVA) followed by Bonferroni’s t test, Dun- nett’s test, or Student’s t test. p-Values less than 0.05 were considered statistically significant.

Results

Inhibitory effects of RXR agonists on proinflammatory cytokine production

To investigate whether SPF1 and SPF2 had anti-inflam- matory effects (Fig. 1), we examined the effects of SPF1 and SPF2 on LPS-stimulated interleukin (IL)-1β mRNA induction in RAW264.7 cells. Pretreatment of RAW264.7 cells with 10 µM SPF1 reduced the level of IL-1β mRNA induced by LPS (Fig. 2a). When RAW264.7 cells were treated with increasing concentrations of SPF1 for 8 h, SPF1 reduced the level of IL-1β mRNA in a concentra- tion-dependent manner (Fig. 2b). Similarly, 10 µM SPF2 and 10 µM BEX reduced IL-1β mRNA levels significantly, whereas SPF3—which does not exert agonistic effects on RXR despite having a similar structure to SPF1 and SPF2 [27]—did not. IL-6 mRNA levels were also reduced in a concentration-dependent and preincubation-time-depend- ent manner by SPF1 (Fig. 2c, d). The mRNA levels of tumor necrosis factor-α (TNF-α), inducible nitric oxide synthase (iNOS), and cyclooxygenase 2 were also reduced by SPF1 (data not shown). SPF1, SPF2, and BEX sig- nificantly reduced production of IL-1β and IL-6 (Fig. 3). In addition, SPF1 and SPF2 markedly suppressed IL-1β and IL-6 levels in peritoneal macrophages derived from C57BL/6 mice (Fig. 4). When TNF-α was used as a stimu- lus instead of LPS, both SPF1 and SPF2 reduced the levels of IL-1β mRNA and TNF-α mRNA (Fig. 5). These results suggest that SPF1 and SPF2 isolated from Sophora tonki- nensis have relatively potent anti-inflammatory activity by way of inhibiting proinflammatory cytokine production in macrophage-like cell lines and peritoneal macrophages.

Cooperative effects of SPF1 and LXR agonists on proinflammatory cytokine production

RXR is known to form permissive heterodimers with LXR. The RXR/LXR heterodimer is subject to agonistic regula- tion, and the combination of RXR and LXR agonism mark- edly enhances transcriptional activity. In a previous study, we found that SPF1 exerted cooperative effects on the expression of RXR/LXR target genes in combination with the LXR agonist T0901317 [27]. Therefore, we sought to investigate whether the anti-inflammatory activity of SPF1 was affected by T0901317. T0901317 reduced IL-1β mRNA and IL-6 mRNA levels significantly, and these effects were markedly enhanced in the presence of SPF1, in a cooperative manner (Fig. 6a, b). The combined effect was also observed in the mouse primary peritoneal macrophages (data not shown). These results suggest that the combination of SPF1 and LXR agonism potentiated anti-inflammatory activity via RXR/LXR heterodimer activation.

Induction of ATF3 by SPF1 alone and in combination with T0901317

Microarray analysis of mRNA expression in RAW264.7 cells treated with SPF1 revealed that SPF1 increased the group. b, d RAW264.7 cells were pretreated with SPF1, SPF2, SPF3, and BEX at the indicated concentrations for 8 h, followed by stimulation with LPS (100 ng/ml) for 12 h. Levels of IL-1β mRNA and IL-6 mRNA were assessed using RT-qPCR and normalized to β-actin mRNA levels, and are represented as fold changes relative to the control group (Ctrl). Data represent the mean ± SD (n = 3) of a representative experiment from two independent experiments with similar results, and statistical significance was evaluated via one-way ANOVA followed by Dunnett’s test. *p < 0.05, **p < 0.01 versus Ctrl mRNA levels of genes associated with anti-inflammatory action. Among them, we focused on ATF3, which is known to interact with the p65 subunit of NF-κB directly and negatively regulate NF-κB activity in toll-like recep- tor 4 (TLR4) signaling [29, 30]. Following treatment of RAW264.7 cells with SPF1, ATF3 mRNA levels were significantly increased, and this effect was enhanced by T0901317. T0901317 alone increased ATF3 mRNA levels slightly, but not significantly (Fig. 7a). Both SPF1 and T0901317 induced ATF3 protein, which was significantly increased in the presence of SPF1 and T0901317 (Fig. 7b). Effect of SPF1 on NF‑κB translocation into nuclei Recent evidence suggests that ATF3 binds to the p65 subu- nit of NF-κB, resulting in suppression of translocation of NF-κB into nuclei and transcription of NF-κB in RAW264.7 cells stimulated by LPS [30]. Thus, to further elucidate the inhibitory effect of SPF1 on proinflammatory cytokine production, we examined the effect of SPF on transloca- tion of the p65 subunit of NF-κB into nuclei following LPS stimulation using an anti-NF-κB p65 subunit antibody and DAPI staining. p65 was mainly expressed in the cytoplasm of unstimulated RAW264.7 cells, but it translocated into the nucleus following LPS stimulation (Fig. 8a). When RAW264.7 cells were pretreated with SPF1 at concentra- tion of 10 µM before LPS stimulation for 30 min, the trans- location of p65 into the nuclei was significantly reduced (Fig. 8a, b). ATF3 is also induced by LPS treatment, which reportedly inhibits TLR-stimulated inflammatory responses as part of a negative feedback loop. When RAW264.7 cells were stimulated with LPS, ATF3 mRNA was significantly induced, but SPF1 did not affect ATF3 mRNA induction (data not shown). Discussion AD-associated inflammation in the CNS is primarily elic- ited by microglia, and increases with AD progression. As well as resident microglia, infiltrating monocytes or mac- rophages from peripheral circulation also contribute to the pathogenesis of AD [31]. Furthermore, systemic inflamma- tion exacerbates the inflammatory environment of the CNS and contributes to chronic neuroinflammation and neuro- degeneration [7, 8]. Thus, as inflammation is intricately involved in the onset and progression of AD, we investigated the anti-inflammatory effects of the naturally occurring pre- nylated flavanoids SPF1 and SPF2, which reportedly have protective effects against Aβ25–35-induced neurotoxicity [28]. In the present study, SPF1 and SPF2 suppressed induction of proinflammatory cytokines including IL-1β and IL-6 in LPS-stimulated and TNF-α-stimulated RAW264.7 cells and peritoneal macrophages. Similar results were observed with the synthetic RXR agonist BEX. However, SPF3, which has a structure similar to SPF1 and SPF2 but no RXR ligand activity, did not exhibit such an effect. These results suggest that the anti-inflammatory effects of SPF1 and SPF2 may result from RXR activation. However, when we examined the effect of the RXR antagonist HX531 and PA452 on the anti-inflammatory effects of SPF1 and SPF2, both RXR antagonists by themselves markedly increased proinflamma- tory cytokine mRNA levels in the models employed. There- fore, whether the effects of SPF1 and PSF2 were mediated by RXR was not definitely assessed. SPF1 and T0901317 suppressed LPS-stimulated proinflammatory cytokine pro- duction, suggesting that SPF1 and T0901317 in combination activated RXR/LXR heterodimers, strengthening their anti- inflammatory effects. It is well known that NRs act broadly to suppress pro- inflammatory gene expression. Steroid receptors belonging to the type III NR subfamily such as glucocorticoid recep- tor [32], estrogen receptor [33], and androgen receptor [34] have been shown to inhibit NF-κB activity and physically interact with NF-κB in vitro. Similarly, NRs of the type 1 subfamily including PPARs and LXRs are known to transre- press proinflammatory gene induction [35]. LXRs become SUMOylated in response to LXR agonism, and the result- ing SUMOylated LXR monomer stabilizes corepressor complexes on the promoters of proinflammatory genes, ulti- mately resulting in inhibition of signal-dependent induction of proinflammatory genes by transcription factors such as NF-κB, activator protein-1, and signal transducer and activa- tor of transcription [36]. It has been reported that inflamma- tory mediators—but not RXR ligands—cause SUMOylation of RXRα, reducing RXR activity in hepatoma cells [37], but whether SUMOylated RXR plays an important role in tran- srepression of proinflammatory genes has not yet been deter- mined. It has also been reported that RXR ligand binds to NF-κB and blocks inflammatory signaling in macrophages and epithelial cells [38, 39], but the mechanisms involved were not adequately characterized. In the present study, we did not detect SUMOylation of RXR, thus we conclude that transrepression through RXR SUMOylation is not the main mechanism underlying the anti-inflammatory effects of SPF1 and SPF2. Therefore, the mechanism underlying the anti-inflammatory effects of RXR ligands has not been established, and the involvement of RXR heterodimer in anti-inflammatory effects also remains to be characterized. It has been hypothesized that LXRs transactivate gene expression via RXR/LXR heterodimers to regulate physiological functions, while LXRs may transrepress inflammatory gene expression by acting as LXR mono- mers. It was recently reported that knockdown of RXRα and RXRβ in mouse macrophages abrogated the ability of LXR agonist to repress LPS-induced expression of proin- flammatory genes [40], suggesting that LXR ligands exert anti-inflammatory effects via RXR/LXR heterodimer acti- vation. In the present study, the anti-inflammatory effects of T0901317 were also markedly attenuated in RAW264.7 cells, partially knocking down RXRα/β by only approxi- mately 40 % via small interfering RNA (siRNA), suggest- ing that T0901317 likely mainly exerted anti-inflammatory effects via RXR/LXR heterodimer, but not LXR monomer. The involvement of SUMOylated LXR in transrepression is not well established. In contrast, it has been reported that LXR suppresses TLR4 signaling by depleting cholesterol in lipid rafts of plasma membrane via adenosine triphos- phate-binding cassette transporter A1 (ABCA1) induction [40]. It has also been reported that the anti-inflammatory activity of LXR ligands remains present or is increased in ABCA1−/−ABCG1−/− macrophages [41]. In the present study, we evaluated the involvement of ABCA1 induction in the anti-inflammatory effect exerted by SPF1, but did not detect any significant positive correlations between ABCA1 levels and the anti-inflammatory effects of SPF1, T0901317, or PA452. Thus, the precise mechanism underlying the anti- inflammatory effects of LXR ligands is likely to be intricate, and remains to be established. The mechanism by which SPF1 and SPF2 exert anti- inflammatory effects seems unlikely to be associated with transrepression via RXR monomer. It is more likely to be associated with transactivation by RXR/LXR heterodimer. In the present study, SPF1 induced ATF3 mRNA and pro- tein expression, which were further enhanced by T0901317. This finding is presumably associated with the existence of two RXR binding sites in the ATF3 promoter [42]. How- ever, the synthetic RXR agonist BEX did not induce ATF3 mRNA. This result suggests that the mechanism underlying the anti-inflammatory activity of SPF1 may be distinct from that of BEX, or that the ability of SPF1 to induce ATF3 may not be shared by any of the other proteins investigated in the current study. ATF3 is induced by various TLR ligands, including LPS and zymosan [42]. It functions as a transcrip- tional repressor in macrophages, and plays a regulatory role in a negative feedback loop involved in the modulation of LPS-stimulated inflammatory responses [29]. ATF3 not only prevents binding of NF-κB to the promoters of proinflam- matory cytokine genes but also impedes translocation of NF-κB into the nucleus via direct physical association with the p65 subunit of NF-κB [30]. Therefore, ATF3 induced by pretreatment with SPF1 and/or T0901317 seems likely to be implicated in their anti-inflammatory effects, including those observed following LPS stimulation. Thus, we indicate for the first time that an RXR ligand exerts anti-inflammatory effects by inducing ATF3 expression via RXR/LXR heter- odimer, as evidenced by the observation that translocation of NF-κB was impeded and proinflammatory cytokine expres- sion was also transcriptionally suppressed by ATF3. Several previous studies have reported the anti-inflammatory activity of RXR ligands. 1,25-Dihydroxyvitamin D3 (VD3) and the retinoid X receptor agonist HX630 addi- tively inhibited LPS-induced expression of mRNAs encod- ing inducible nitric oxide synthase (iNOS) and IL-6, whereas expression of IL-1β mRNA was only inhibited by VD3 in microglial BV-2 cells [43]. The underlying mechanism may be at least in part due to inhibited LPS-induced acti- vation of extracellular signal-regulated kinase and nuclear translocation of NF-κB [43]. BEX and 9-cis-retinoic acid reportedly inhibit TNF-α-induced leukocyte/endothelial cell interaction through PPARγ/RXRα activation, but not RXRα monomer [44]. Additionally, the LXR agonist T0901317 and 9-cis-retinoic acid cooperatively suppressed LPS-induced nitric oxide production in microglial cells by preventing the degradation of inhibitor-κB-α protein by T0901317 [45], although the concentration of T0901317 used in that study was extremely high, thus nonspecific effects such as FXR activation may have been involved. The anti-inflammatory activities of NR ligands including RXR ligands have been reported, but the underlying mechanisms have not yet been clearly elucidated. In the present study, the RXR agonist SPF1 exhibited anti-inflammatory effects, and the gene prod- uct ATF3, induced by SPF1—mainly via RXR/LXR heter- odimer—was intimately involved in these anti-inflammatory effects. Sophora tonkinensis Gagnep. has been used traditionally in China and Korea for treatment of inflammatory condi- tions including pharyngitis, tonsillitis, asthma, and chronic hepatitis [46]. Recent studies indicate that the pterocarpan- type flavonoid sophotokin, maackiapterocarpan B, and 6,8-diprenyl-7,4′-dihydroxyflavanone, which have been iso- lated from Sophora tonkinensis, exhibit anti-inflammatory effects [47–49] and that the ethyl acetate extract of Sophora tonkinensis exerts a hypoglycemic effect in KK-Ay diabetic mice [50]. Thus, accumulating evidence may support the usefulness of Sophora tonkinensis for treatment of inflam- matory diseases and as a natural resource from which to derive therapeutic agents. Collectively, the results of the current study suggest that the prenylated flavanoids SPF1 and SPF2, which are selec- tive RXR agonists, exert anti-inflammatory effects mainly via RXR/LXR heterodimer. Although the inflammatory effects mediated by RXR/LXR heterodimer have not been previously thoroughly investigated, these results shed light on the mechanisms involved, and future therapeutic strat- egies for treatment of inflammation-associated conditions such as AD. References 1. Stephenson J, Nutma E, van der Valk P, Amor S (2018) Inflam- mation in CNS neurodegenerative diseases. Immunology 154:204–219 2. Perry VH, Cunningham C, Holmes C (2007) Systemic infections and inflammation affect chronic neurodegeneration. Nat Rev Immunol 7:161–167 3. Nimmo AJ, Cernak I, Heath DL, Hu X, Bennett CJ, Vink R (2004) Neurogenic inflammation is associated with development of edema and functional deficits following traumatic brain injury in rats. Neuropeptides 38:40–47 4. Tynan RJ, Naicker S, Hinwood M, Nalivaiko E, Buller KM, Pow DV, Day TA, Walker FR (2010) Chronic stress alters the density and morphology of microglia in a subset of stress-responsive brain regions. Brain Behav Immun 24:1058–1068 5. Brüünsgaard H, Pedersen BK (2003) Age-related inflammatory cytokines and disease. Immunol Allergy Clin North Am 23:15–39 6. Cuello AC (2017) Early and late CNS inflammation in Alzhei- mer’s disease: two extremes of a continuum? Trends Pharmacol Sci 38:956–966 7. Butchart J, Holmes C (2012) Systemic and central immunity in Alzheimer’s disease: therapeutic implications. CNS Neurosci Ther 18:64–76 8. Holmes C (2013) Systemic inflammation and Alzheimer’s disease. Neuropathol Appl Neurobiol 39:51–68 9. Evans RM (1988) The steroid and thyroid hormone receptor superfamily. Science 240:889–895 10. Glass CK, Saijo K (2010) Nuclear receptor transrepression path- ways that regulate inflammation in macrophages and T cells. Nat Rev Immunol 10:365–376 11. Moutinho M, Landreth GE (2017) Therapeutic potential of nuclear receptor agonists T0901317 in Alzheimer’s disease. J Lipid Res 58:1937–1949
12. Mandrekar-Colucci S, Karlo JC, Landreth GE (2012) Mechanisms underlying the rapid peroxisome proliferator-activated receptor- gamma-mediated amyloid clearance and reversal of cognitive deficits in a murine model of Alzheimer’s disease. J Neurosci 32:10117–10128
13. Malm T, Mariani M, Donovan LJ, Neilson L, Landreth GE (2015) Activation of the nuclear receptor PPARδ is neuroprotective in a transgenic mouse model of Alzheimer’s disease through inhibition of inflammation. J Neuroinflamm 12:7
14. Mariani MM, Malm T, Lamb R, Jay TR, Neilson L, Casali B, Medarametla L, Landreth GE (2017) Neuronally-directed effects of RXR activation in a mouse model of Alzheimer’s disease. Sci Rep 7:42270
15. Katsuki H, Kurimoto E, Takemori S, Kurauchi Y, Hisatsune A, Isohama Y, Izumi Y, Kume T, Shudo K, Akaike A (2009) Retinoic acid receptor stimulation protects midbrain dopaminergic neurons from inflammatory degeneration via BDNF-mediated signaling. J Neurochem 110:707–718
16. Zelcer N, Khanlou N, Clare R, Jiang Q, Reed-Geaghan EG, Lan- dreth GE, Vinters HV, Tontonoz P (2007) Attenuation of neu- roinflammation and Alzheimer’s disease pathology by liver X receptors. Proc Natl Acad Sci USA 104:10601–10606
17. Casali BT, Corona AW, Mariani MM, Karlo JC, Ghosal K, Lan- dreth GE (2015) Omega-3 fatty acids augment the actions of nuclear receptor agonists in a mouse model of Alzheimer’s dis- ease. J Neurosci 35:9173–9181
18. Corona AW, Kodoma N, Casali BT, Landreth GE (2015) ABCA1 is necessary for bexarotene-mediated clearance of soluble amyloid beta from the hippocampus of APP/PS1 mice. J Neuroimmune Pharmacol 11(1):61–72
19. Savage JC, Jay T, Goduni E, Quigley C, Mariani MM, Malm T, Ransohoff RM, Lamb BT, Landreth GE (2015) Nuclear receptors license phagocytosis by trem2+ myeloid cells in mouse models of Alzheimer’s disease. J Neurosci 35:6532–6543
20. Cramer PE, Cirrito JR, Wesson DW, Lee CYD, Kario JC, Zinn AE, Casali BT, Restivo JL, Goebel WD, James MJ, Brunden KR, Wilson DA, Landreth GE (2012) ApoE-directed therapeutics rap- idly clear β-amyloid and reverse deficits in AD mouse models. Science 335:1503–1506
21. Landreth GE, Cramer PE, Lakner MM, Cirrito JR, Wesson DW, Brunden KR, Wilson DA (2013) Response to comments on “ApoE-directed therapeutics rapidly clear β-amyloid and reverse deficits in AD mouse models”. Science 340:924-g
22. Mouier A, Georgiev D, Nam KN, Fitz NF, Castranio EL, Wolfe CM, Cronican AA, Schug J, Lefterov I, Koldamova R (2015) Bexarotene-activated retinoid X receptors regulate neuronal dif- ferentiation and dendritic complexity. J Neurosci 35:11862–11876
23. Pinaire JA, Reifel-Miller A (2007) Therapeutic potential of reti- noid X receptor modulators for the treatment of the metabolic syndrome. PPAR Res 2007:94156
24. Liu S, Ogilvie KM, Klausing K, Lawson MA, Jolley D, Li D, Bilakovics J, Pascual B, Hein N, Urcan M, Leibowitz MD (2002) Mechanism of selective retinoid X receptor agonist-induced hypo- thyroidism in the rat. Endocrinology 143:2880–2885
25. Lenhard JM, Lancaster ME, Paulik MA, Weiel JE, Binz JG, Sund- seth SS, Gaskill BA, Lightfoot RM, Brown HR (1999) The RXR agonist LG100268 causes hepatomegaly, improves glycaemic con- trol and decreases cardiovascular risk and cachexia in diabetic mice suffering from pancreatic beta-cell dysfunction. Diabetologia 42:545–554
26. Al Mamun Bhuyan A, Bissinger R, Cao H, Lang F (2016) Trig- gering of suicidal erythrocyte death by bexarotene. Cell Physiol Biochem 40:1239–1251
27. Inoue M, Tanabe H, Nakashima K, Ishida Y, Kotani H (2014) Rexinoids isolated from Sophora tonkinensis with a gene expres- sion profile distinct from the synthetic rexinoid bexarotene. J Nat Prod 77:1670–1677
28. Wang W, Nakashima K, Hirai T, Inoue M (2019) Neuroprotective effect of naturally occurring RXR agonists isolated from Sophora tonkinensis Gagnep. on amyloid-β-induced cytotoxicity in PC12 cells. J Nat Med 73:154–162
29. Gilchrist M, Thorsson V, Li B, Rust AG, Korb M, Kennedy K, Hai T, Bolouri H, Aderem A (2006) Systems biology approaches iden- tify ATF3 as a negative regulator of Toll-like receptor 4. Nature 441:173–178
30. Kwon JW, Kwon HK, Shin HJ, Choi YM, Anwar MA, Choi S (2015) Activating transcription factor 3 represses inflamma- tory responses by binding to the p65 subunit of NF-κB. Sci Rep 5:14470
31. Cameron B, Landreth GE (2010) Inflammation, microglia and Alzheimer’s disease. Neurobiol Dis 37:503–509
32. Ray A, Prefontaine KE (1994) Physical association and functional antagonism between the p65 subunit of transcription factor NF- kappa B and the glucocorticoid receptor. Proc Natl Acad Sci USA 91:752–756
33. Ray A, Prefontaine KE, Ray P (1994) Down-modulation of inter- leukin-6 gene expression by 17 beta-estradiol in the absence of high affinity DNA binding by the estrogen receptor. J Biol Chem 269:12940–12946
34. Palvimo JJ, Reinkainen P, Ikonen T, Kallio PJ, Moilanen A, Janne OA (1996) Mutual transcriptional interference between RelA and androgen receptor. J Biol Chem 271:24151–24156
35. Ghisletti S, Huang W, Ogawa S, Pascual G, Lin ME, Willson TM, Rosenfeld MG, Glass CK (2007) Parallel SUMOylation-depend- ent pathways mediate gene- and signal-specific transrepression by LXRs and PPAR. Mol Cell 25:57–70
36. Lee JH, Park SM, Kim OS, Lee GS, Woo JH, Park SJ, Joe EH, Jou I (2009) Differential SUMOylation of LXRα and LXRβ medi- ates transrepression of STAT1 inflammatory signaling in IFN-γ- stimulated brain astrocytes. Mol Cell 35:806–817
37. Aguirre RS, Karpen SJ (2013) Inflammatory mediators increase SUMOylation of retinoid X receptor α in a c-Jun N-terminal kinase-dependent manner in human hepatocellular carcinoma cells. Mol Pharmacol 84:218–226
38. Na SY, Kang BY, Chung SW, Han SJ, Ma X, Trinchieri G, Im SY, Lee JW, Kim TS (1999) Retinoids inhibit interleukin-12 production in macrophages through physical associations of reti- noid X receptor and NFκB. J Biol Chem 274:7674–7680
39. Lee HC, Headley MB, Iseki M, Ikuta K, Ziegler SF (2008) Cutting edge: inhibition of NF-κB-mediated TSLP expression by retinoid X receptor. J Immunol 181:5189–5193
40. Ito A, Hong C, Rong X, Zhu X, Tarling EJ, Hedde PN, Gratton E, Parks J, Tontonoz (2015) LXRs link metabolism to inflammation through Abca1-dependent regulation of membrane composition and TLR signaling. eLife 4:e08009
41. Kappus MS, Murphy AJ, Abramowicz S, Ntonga V, Welch CL, Tall AR, Westerterp M (2014) Activation of liver X receptor decreases atherosclerosis in Ldlr−/− mice in the absence of ATP- binding cassette transporters A1 and G1 in myeloid cells. Arte- rioscler Thromb Vasc Biol 34:279–284
42. Hai T, Wolfgang CD, Marsee DK, Allen AE, Sivaprasad U (1999) ATF3 and stress responses. Gene Expr 7:321–335
43. Dulla YAT, Kurauchi Y, Hisatsune A, Seki T, Shudo K, Katsuki H (2016) Regulatory mechanisms of vitamin D3 on production of nitric oxide and pro-inflammatory cytokines in microglial BV-2 cells. Neurochem Res 41:2848–2858
44. Sanz MJ, Albertos F, Otero E, Juez M, Morcillo EJ, Piqueras L (2012) Retinoid X receptor agonists impair arterial mononu- clear cell recruitment through peroxisome proliferator-activated receptor-γ activation. J Immunol 189:411–424
45. Zhang-Gandhi CX, Drew PD (2007) Liver X receptor and retinoid X receptor agonists inhibit inflammatory responses of microglia and astrocytes. J Neuroimmunol 183:50–59
46. Chen Y, Chen LM, Tong Y, You Y (2017) Pharmacological effect and toxicology of Sophorae Tonkinensis Radix et Rhizoma. Zhongguo Zhong Yao Za Zhi 42:2439–2442
47. Chae HS, Yoo H, Choi YH, Choi WJ, Chin YW (2016) Maacki- apterocarpan B from Sophora tonkinensis suppresses inflamma- tory mediators via nuclear factor-κB and mitogen-activated pro- tein kinase pathways. Biol Pharm Bull 39:259–266
48. Xia W, Luo P, Hua P, Ding P, Li C, Xu J, Zhou H, Gu Q (2018) Discovery of a new pterocarpan-type antineuroinflammatory compound from Sophora tonkinensis through suppression of the TLR4/NFκB/MAPK signaling pathway PU.1 as a potential tar- get. ACS Chem Neurosci. https://doi.org/10.1021/acschemneu ro.8b00243
49. Chae HS, Yoo H, Kim YM, Choi YH, Lee CH, Chin YW (2016) Anti-inflammatory effects of 6,8-diprenyl-7,4′-dihydroxyflavanone from Sophora tonkinensis on lipopolysaccharide-stimulated RAW264.7 cells. Molecules 21:E1049
50. Huang M, Deng S, Han Q, Zhao P, Zhou Q, Zheng S, Ma X, Xu C, Yang J, Yang X (2016) Hypoglycemic activity and the potential mechanism of the flavonoid rich extract from Sophora tonkinensis Gagnep. in KK-Ay mice. Front Pharmacol 7:288

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.