The key royal jelly component 10-hydroxy-2-decenoic acid protects against bone loss by inhibiting NF- k B signaling downstream of FFAR4

The supplementation of royal jelly (RJ) is known to provide a variety of health benefits, including anti-inflammatory and anti-obesity effects. RJ treatment also reportedly protects against bone loss, but no single factor in RJ has yet been identified as an anti-osteoporosis agent. Here we fractionated RJ and identified 10-hydroxy-2-decenoic acid (10H2DA) as a key component involved in inhibiting osteoclastogenesis based on mass spectro-metric analysis. We further demonstrated free fatty acid receptor 4 (FFAR4) as directly interacting with 10H2DA; binding of 10H2DA to FFAR4 on osteoclasts inhibited receptor activator of nuclear factor- k B (NF- k B) ligand (RANKL)-induced activation of NF- k B signaling, thereby attenuating the induction of nuclear factor of activated T cells (NFAT) c1, a key transcription factor for osteoclastogenesis. Oral administration of 10H2DA attenuated bone resorption in ovariectomized mice. These results suggest a potential therapeutic approach of targeting osteoclast differentiation by the supplementation of RJ, and specifically 10H2DA, in cases of pathological bone loss such as occur in postmenopausal osteoporosis.

The supplementation of royal jelly (RJ) is known to provide a variety of health benefits, including anti-inflammatory and antiobesity effects.RJ treatment also reportedly protects against bone loss, but no single factor in RJ has yet been identified as an anti-osteoporosis agent.Here we fractionated RJ and identified 10-hydroxy-2-decenoic acid (10H2DA) as a key component involved in inhibiting osteoclastogenesis based on mass spectrometric analysis.We further demonstrated free fatty acid receptor 4 (FFAR4) as directly interacting with 10H2DA; binding of 10H2DA to FFAR4 on osteoclasts inhibited receptor activator of nuclear factor-kB (NF-kB) ligand (RANKL)-induced activation of NF-kB signaling, thereby attenuating the induction of nuclear factor of activated T cells (NFAT) c1, a key transcription factor for osteoclastogenesis.Oral administration of 10H2DA attenuated bone resorption in ovariectomized mice.These results suggest a potential therapeutic approach of targeting osteoclast differentiation by the supplementation of RJ, and specifically 10H2DA, in cases of pathological bone loss such as occur in postmenopausal osteoporosis.
Royal jelly (RJ), an essential food required by the queen honeybee, is a product secreted from the hypopharyngeal and mandibular glands of worker honeybees.RJ has received considerable attention due to its pharmacological activities such as immunomodulatory, antitumor, antimicrobial, and vasoactive effects (1,2).Raw RJ consists of water, lipids, proteins, carbohydrates, and other minor constituents.The lipid composition of RJ is comprised of 90-95% fatty acids (1).RJ contains certain unique constituents, such as major royal jelly proteins.
The requisite homeostasis of adult bone is maintained by a continuous physiological process, termed bone remodeling.This process requires a finely balanced activity of specialized groups of cells, including osteoclasts, which are multinucleated cells that resorb bone, and osteoblasts, which refill the resorption cavities created by osteoclasts.An imbalance between osteoclasts and osteoblasts can lead to metabolic bone diseases, such as osteoporosis.Osteoclasts are derived from monocyte/ macrophage lineage precursor cells and their differentiation depends on receptor activator of NF-kB ligand (RANKL) (3,4) RANKL binding to RANK, a transmembrane molecule expressed on osteoclast precursor cells and mature osteoclasts, activates NF-kB, mitogen-activated protein kinases, and AP-1, consequently leading to the activation of nuclear factor of activated T cell (NFAT) c1, a key transcription factor of osteoclast differentiation (3,4).
Although a previous study reported a significant effect of the administration of RJ on the mineral content of bone ( 5), the molecular mechanisms by which RJ mediates the anti-osteoporotic effects were still unknown.In this study, we show that RJ partially prevents ovariectomy (OVX)-induced bone loss through the suppression of osteoclastogenesis in vivo.Furthermore, we show that 10-hydroxy-2-decenoic acid (10H2DA), one of a group of unique medium-chain fatty acids present in RJ, is a potent inhibitor of RANKL-induced osteoclastogenesis.These findings contribute to the understanding of the molecular basis for the beneficial effects of bone health elicited by RJ intake.

Oral administration of RJ protected against OVX-induced bone loss in mice
A mouse model of postmenopausal osteoporosis due to estrogen deficiency was used to examine the effects of oral administration of RJ on bone remodeling.Ovariectomized 9week-old mice were treated with RJ using oral gavage once a day for 4 weeks.Microcomputed tomography (mCT) analysis revealed that ovariectomy-induced significant decreases in bone volume/tissue volume (BV/TV) and trabecular number (Tb.N) in mice, but RJ administration protected against bone loss as measured by BV/TV and Tb.N (Fig. 1, A and B).Similarly, OVX-induced decrease in bone mineral content per tissue volume, increases in trabecular separation, and marrow space star volume were blunted by RJ administration (Fig. 1A and Fig. S1A).A significantly lower osteoclast number was found in RJtreated ovariectomized mice when compared with vehicletreated controls using histological analysis (Fig. 1, C and D).‡ These authors contributed equally to this work.* For correspondence: Tomoki Nakashima, naka.csi@tmd.ac.jp.
Eroded surface and osteoclast surface also showed a tendency to decrease compared with vehicle-treated control (Fig. S1B).In contrast, RJ did not significantly induce the modulation of osteoblast surface and osteoblastic bone formation (Fig. 1, C and D, and Fig. S1B).These results suggest that RJ administration prevented OVX-induced bone loss by inhibiting osteoclastic bone resorption.

Identification of 10H2DA as an anti-osteoclastogenic molecule contained in RJ
These results prompted us to investigate the cellular and molecular mechanisms that contribute to the RJ-induced protection of bone loss induced by estrogen deficiency.The addition of RJ significantly inhibited the osteoclast differentiation induced by stimulation with macrophage-colony stimulating factor (M-CSF)/RANKL (Fig. 2A) and co-culture of bone marrow cells with calvarial osteoblastic cells (Fig. 2B) in a dose-dependent manner.Interestingly, RJ treatment had no effect on the expression of Tnfsf11 (encodes RANKL) or Tnfrsf11b (encodes OPG) in calvarial cells, both of which are involved in the supporting activity of osteoclastogenesis (Fig. 2C), indicating that RJ inhibited osteoclastogenesis directly.
To gain insight into the RJ constituents that suppress osteoclastogenesis, we first examined whether any of the putative anti-osteoclastogenic molecules in RJ are resistant to heating.The inhibitory effect of RJ on osteoclastogenesis was not affected by heating (Fig. 3A).Next, RJ was divided into a methanol-soluble portion and a methanol-insoluble residue (Fig. 3B).We confirmed that the methanol-soluble fraction hardly contained any protein (Table S1) and this deproteinized fraction had similar inhibitory activity toward osteoclastogenesis as RJ itself (Fig. 3C).The lipid-soluble constituents in the deproteinized fraction, which were extracted in ethyl acetate exhibited anti-osteoclastogenic activity (Fig. 3C), were fractionated using silica gel column chromatography (Fig. 3B).An anti-osteoclastogenic factor was eluted in chloroform-methanol (9:1, v/v).Among 30 fractions, we found that the pooled fractions of 16-20 had a similar inhibitory activity on osteoclastogenesis as before purification (Fig. 3D).To isolate the anti-osteoclastogenic molecule, the pooled fraction was then separated using reversed phase HPLC (Fig. 3E).Among the peaks we detected, the purified peak C still exhibited a similar inhibitory effect on osteoclastogenesis (Fig. 3E).To obtain information on the molecules contained in peak C, we utilized liquid chromatography-tandem MS (LC-MS/MS) (Fig. 3F).Mass analysis at the 10.7-min retention time of peak C (Fig. S2A) indicated the presence of a molecule that showed a strong molecular ion peak at m/z 185.1, which represented 10H2DA (Fig. S2B).The MS/MS spectra of these fractions were consistent with that of 10H2DA (Fig. 3F).Indeed, the LC-MS chromatograms of peak C exhibited a retention time that was similar to that of 10H2DA (Fig. S2A).Consistent with this, the treatment with 10H2DA remarkably decreased the formation of osteoclasts in a dose-dependent manner (Fig. 3G).

The molecular basis of 10H2DA-induced suppression of osteoclastogenesis
We analyzed the expression of the genes implicated in osteoclastogenesis and bone resorption to elucidate the molecular mechanism of the 10H2DA-mediated inhibitory effect.Neither 10H2DA nor RJ had any influence on the expression of Tnfrsf11a (encodes RANK), Csf1r (encodes M-CSFR), Fos, and Mitf (Fig. 4, A and B).In contrast, the expression of the key transcription factor of osteoclastogenesis Nfatc1 and its downstream elements was significantly decreased in osteoclasts treated with 10H2DA or RJ, without affecting the expression of Irf8 and Mafb, factors that are known to be involved in the inhibition of Nfatc1 expression (Fig. 4, A and B).Protein levels of cathepsin K (CtsK), tartrate-resistant acid phosphatase (TRAP), V-type proton ATPase subunit D2 (V-ATPase D2), and matrix metalloproteinase 9 (MMP9), all of which are downstream molecules of the RANK signaling pathway and involved in bone resorption, were decreased in osteoclasts treated with 10H2DA (Fig. 4, C and D).Indeed, the areas of hydroxyapatite resorption in 10H2DAtreated osteoclasts were significantly decreased (Fig. 4E).
To confirm 10H2DA as an inhibitor of bone resorption in vivo, OVX mice were treated with 10H2DA using oral gavage once a day for 4 weeks.Serum levels of the N-terminal propeptide of type I collagen (PINP), which is used to measure bone formation activity, was not changed by 10H2DA treatment.However, serum C-terminal telopeptides of type I collagen (CTX-I) level, which reflects osteoclastic activity, was significantly inhibited by 10H2DA treatment (Fig. 4F).These results showed that treatment with 10H2DA significantly inhibited OVX-induced osteoclastic bone resorption without affecting bone formation.Thus, 10H2DA is the molecule responsible for the inhibitory activity of RJ on OVX-induced bone loss.

FFAR4 is a receptor for 10H2DA in mature osteoclasts
We next searched for a putative receptor expressed in osteoclasts to determine the molecular mechanisms underlying 10H2DA-induced inhibition of osteoclastogenesis.Because 10H2DA is classified as a medium-chain fatty acid, we exam-ined the expression of free fatty acid receptors (6,7).Although the Ffar1, Ffar2, Ffar3, and Gpr84 expression patterns were greatly reduced, Ffar4 expression was strikingly increased after RANKL stimulation (Fig. 5A).Therefore, we tested whether FFAR4 serves as a receptor for 10H2DA.In a reporter assay, 10H2DA dose-dependently activated FFAR4 with EC 50 of 1.025 mM (Fig. 5B), indicating that 10H2DA had a moderate activity for FFAR4 activation.It has been demonstrated that the FFAR4 expressed in macrophages mediates the anti-inflammatory effects of fatty acids via inhibiting the NF-kB signaling pathway (8).Because NF-kB signaling is crucial for RANKL-induced osteoclastogenesis, we examined the effect of 10H2DA treatment on NF-kB signaling.RANKL-induced phosphorylation and the degradation of IkBa were both diminished in 10H2DA-treated osteoclasts (Fig. 5, C and D).When Ffar4 expression was knocked down with a lentiviral vector expressing a short hairpin RNA (shRNA) targeting Ffar4 (Fig. 5E), the inhibitory effect of 10H2DA on osteoclastogenesis was abrogated (Fig. 5F).In addition, 10H2DA-induced inhibition of IkBa phosphorylation and degradation was abrogated in Ffar4-silenced osteoclasts (Fig. 5, G  and H).These results collectively indicate that a single factor contained in RJ, 10H2DA, suppressed the NF-kB signaling pathway through the FFAR4 receptor so as to inhibit osteoclastogenesis.

Discussion
RJ is increasingly used as a dietary supplement for its healthpromoting effects (1).Thus, the characterization of the molecular mechanisms underlying these activities in normal and pathological contexts will help not only to better understand the molecular basis of these health-promoting effects of RJ, but also highlight potential pathways that can be targeted for therapy.In this study, we found that RJ administration to ovariectomized mice ameliorated bone loss through the suppression of OVX-induced activation of osteoclastogenesis.The anti-osteoclastogenic factor 10H2DA, which we identified in this study, is a medium-chain fatty acid specifically found in RJ.Although it is known to be one of the molecules responsible for the various pharmacological effects of RJ (9, 10), there is no report of a 10H2DA effect on osteoclastogenesis.Moreover, we speculate that RJ contains other anti-osteoclastogenic molecule (s) in addition to 10H2DA, because at concentrations of 10H2DA equivalent to those found in RJ (1.54%) (9) osteoclastogenesis was inhibited, but to a lesser extent.
FFAR4 is a member of the rhodopsin-like G protein-coupled receptor family, also known as G protein-coupled receptor 120 (GPR120).FFAR4, along with FFAR1, functions as a receptor for saturated and unsaturated fatty acids of medium-to longchain length (11, 12), including docosahexaenoic acid, whereas The key royal jelly component protects bone loss via FFAR4 FFAR2 and -3 are receptors for short-chain fatty acids.Ligation of docosahexaenoic acid to FFAR4 on macrophage induces the recruitment of TAB1 to FFAR4, thereby sequestering it from TAK1 and suppressing IKKa/b-NF-kB activation (8).In addition to natural ligands, several synthetic FFAR4 agonists have been developed and characterized (13).In particular, it has been reported that GW9508, a synthetic agonist of FFAR4 and FFAR1, inhibited osteoclast differentiation (14,15).Consistent with these reports, we found that 10H2DA has an inhibitory effect on osteoclast differentiation and function by suppressing the NF-kB signaling pathway and its downstream molecules including NFATc1, CtsK, TRAP, V-ATPase D2, and MMP9, via FFAR4 (Fig. 6).Moreover, it has been shown that FFAR4 is implicated in diverse physiological and pathological activities, including inflammation (16), secretion of glucagon-like peptide-1 (17), adipocyte differentiation (18), insulin sensitization, regulation of appetite (19), and tumor progression (20).Most of the beneficial effects of RJ are related to it being an FFAR4 ago-nist.Therefore, FFAR4 represents a promising target for the treatment of obesity-related metabolic disorders, such as inflammation and cancer, in addition to osteoporosis, due to its involvement in the regulation of adipogenesis, inflammation, insulin resistance and bone resorption.
In summary, we have demonstrated that 10H2DA, a key RJ constituent, suppresses osteoclastogenesis by inhibiting NF-kB signaling through its receptor, FFAR4.Because RJ is one of the most widely used health-promoting foods, RJ supplementation, as well as 10H2DA administration, comprise a potential therapeutic approach against various metabolic bone diseases, including osteoporosis.

Experimental procedures
Reagents RJ was a gift from Masaki Kamakura or provided by MORIKAWA KENKODO.10H2DA was purchased from the Nagara Science Corporation.

The key royal jelly component protects bone loss via FFAR4
Oral administration of RJ to ovariectomized mice Eight-week-old female C57BL6/J mice were purchased from CLEA Japan.Animals were housed in groups of 4-6 per cage under a 12-h light/dark cycle and specific pathogen-free conditions.Animals were provided sterile water and food (ORIENTAL YEAST CO., LTD).After acclimatization for a week, OVX or sham operation was performed as previously described (21).Mice were given daily oral injections of 1.0 g/kg of body weight of RJ (dissolved in saline) or 40 mg/kg of body weight of 10H2DA (dissolved in 5% etha-nol in corn oil) with a feeding needle for 4 weeks.Severe adverse events did not occur during observation.Uterine weights were recorded to confirm the success of OVX.Mice were injected with calcein (16 mg/kg) 4 days and 1 day prior to sacrifice.One day after the last injection of RJ, all mice were sacrificed.Femurs and tibiae were excised and fixed with 70% ethanol for 1 week prior to mCT and bone histomorphometric analyses, respectively.All animal experiments were approved by the Institutional Animal Care and Use Committee of Tokyo Medical and Dental University.Three-dimensional mCT scanning was performed using the ScanXmate-D090S105 Scanner System (Comscantecno).Images were acquired using the following settings: 80 kV voltage, 100 mA current, 14.848 mM voxel resolution, a frame rate of 8000 ms, 360°r otation with no frame averaging.Three-dimensional mCT images were reconstructed and the structural indices calculated using a three-dimensional image analysis system (TRI/3D-BON, RATOC System Engineering) (22)(23)(24).

Bone histomorphometric analysis
Tibiae were embedded in glycol methacrylate, sectioned (3 mM), and stained with toluidine blue and TRAP.Parameters for osteoblast and osteoclast were assessed in microscopic fields from two sections per mice by moving the sections in equally sized steps along the x and y axes.Images were taken using a light microscope (Axio Imager 2; Zeiss) and all histological analyses were performed using WinROOF 2013 software (Mitani) (22,23).

Enzyme-linked immunoassay
Serum carboxyl-terminal CTX-I and PINP levels were measured with the RatLaps and Rat/Mouse PINP EIA kits (Immunodiagnostic Systems), respectively.

In vitro osteoclast differentiation
In vitro osteoclast differentiation in a monoculture system was performed as described previously, with certain minor modifications (22,25).Primary bone marrow cells (1 3 10 5 cells/cm 2 ) were maintained in a culture medium (a-minimal essential medium containing penicillin, streptomycin, and 10% fetal bovine serum) supplemented with 10 ng/ml of M-CSF (R&D systems) for 2 days to obtain BMMs.Then BMMs were cultured in medium supplemented with 10 ng/ml of M-CSF, 12.5-50 ng/ml of RANKL and RJ or RJ components for 3 days.The culture medium was changed every second day.For the generation of osteoclasts in vitro in the coculture system, pri-mary bone marrow cells (5 3 10 4 cells/cm 2 ) and calvarial cells (5 3 10 3 cells/cm 2 ) were cultured in the presence of 10 nM 1a,25-dihydroxyvitamin D 3 (1a,25(OH) 2 D 3 ) and 1 mM prostaglandin E 2 for 4-6 days.Osteoclastogenesis was evaluated by TRAP staining.TRAP 1 MNCs (more than 3 nuclei) were counted.Raw RJ was dissolved in 200 mg/ml DMSO and centrifuged.Supernatants were sterilized with a 20-mm syringe filter (Millipore) and used for in vitro analysis.Heat treatment of RJ was performed at 100 °C for 10 min.

LC-MS/MS analysis
LC-MS/MS analysis was performed on a Xevo QT of MS system (Waters) coupled with an ACQUITY UPLC system (Waters).Chromatographic separation was performed on an ACQUITY UPLC HSS T3 column (1.8 mM, 2.1 mm inner diameter 3 100 mm, Waters), equilibrated at 40 °C, with a binary gradient of mobile phase A (0.1% acetate in water) and B (methanol).The gradient was started at 5% B, increased to 100% B at 20 min, and maintained for 5 min at 100% B at a constant flow rate of 0.4 ml/min.Mass spectra were acquired in the negative ion mode with a capillary voltage at 3.0 kV and cone voltage at 15 V, over the mass range m/z 50-1000.To confirm the identity of the 10H2DA ion, the synthetic compound was analyzed under identical conditions.

Quantitative RT-PCR
Total RNA was extracted by ISOGEN (NIPPON GENE) following the manufacturer's protocol (22).First-strand cDNAs were synthesized from 0.5 mg of total RNA using SuperScript III reverse transcriptase (Thermo Fisher Scientific).Quantitative RT-PCR analysis was performed with CFX384 Touch (Bio-Rad) using SYBR Green Real-time PCR Master Mix (TOYOBO).All of the specific primer sequences are listed to Table S2.The level of mRNA expression was normalized with Gapdh expression.

Hydroxyapatite resorption assay
Mature osteoclasts were generated on collagen-coated dishes.These cells were harvested by treatment with trypsin-EDTA solution, and adherent cells (2.0 3 10 4 cells) were plated on 96-well Osteo Assay Surface plate (Corning) and cultured in the presence of 10 ng/ml of M-CSF and 50 ng/ml of RANKL with or without 500 mM 10H2DA for 3 days.Resorption areas were stained with 0.5% toluidine blue and quantified using ImageJ (NIH).

Western blotting analysis
Cells were washed with cold PBS and then lysed with lysis buffer containing Complete protease inhibitor mixture (Roche Bioscience) and PhosSTOP (Roche Bioscience).After incubation on ice for 10 min, lysates were clarified by centrifugation at 15,000 rpm for 10 min.Extracted proteins were subjected to SDS-PAGE and transferred to polyvinylidene difluoride membranes (Millipore).The membranes were blocked with Blocking One P or Bullet Blocking One (Nacalai Tesque) and incubated with primary antibodies at room temperature for 2 h followed by secondary antibodies.The blots were visualized using Chemi-Lumi One Ultra (Nacalai Tesque).Primary antibodies were purchased from Cell Signaling Technology (phospho-IkBa and IkBa), Abcam (CtsK and MMP9), Abnova (TRAP), Santa Cruz Biotechnology (NFATc1), and Sigma-Aldrich (V-ATPase D2 and b-actin).

FFAR4 reporter assay
FFAR4 reporter assay was performed using the FFAR4 (GPR120) Reporter Assay Kit (Cayman Chemical) per the manufacturer's instructions.Luminescence was detected with an EnSight multimode plate reader (PerkinElmer Life Sciences).

Knockdown analysis
The shRNAs targeting FFAR4 mRNA (shFfar4) and control shRNA were purchased from Sigma-Aldrich (14,26).The lentiviral particles were obtained by transfecting the lentivirus into HEK293T cells with a lentiviral packaging mix (Sigma-Aldrich) using the FuGENE HD transfection reagent (Promega).BMMs were infected with virus for 24 h before RANKL stimulation.

Statistical analysis
Statistical analysis was performed using the unpaired twotailed Student's t test or analysis of variance with Tukey's multiple-comparison tests or Dunnett's multiple-comparison tests (*, p , 0.05; **, p , 0.01; ***, p , 0.001; ****, p , 0.0001; throughout the paper).All data are expressed as the mean 6 S.E.Results are respective samples of more than three independent experiments.All statistical analysis was performed using Graph-Pad Prism 6 (GraphPad software).