Epimedin C protects dexamethasone-induced osteoblasts through NRF1/RhoA pathway

Epimedin C enhances the migration/proliferation of OB after DXMS-treatment

The function of Epimedin C in stabilizing OB growth following DXMS treatment was revealed at present study. DXMS-treated cells exhibited a slow growth pattern compared to non-treated cells, whereas the cells treated with Epimedin C exhibited growth at an average speed (Figure 1A). After DXMS treatment, compared with cells treated with DXMS alone, cells cultured with Epimedin C showed better growth and proliferation rate (Figure 1B, Column 4, Figure 1C) indicating that Epimedin C enhances the migration and growth after DXMS treatment.

Figure 1. Epimedin C enhances migration of osteoblasts in dexamethasone treatment. (A) A Following scratch wound, application of Epimedin C accelerated migration of osteoblasts. (B) Change in area over time for osteoblasts at 48 h after treatment was quantified. n = 6 independent experiments. ^*p < 0.01. (C) Cell proliferation was measured by BrdU assay. OB cells were seeded in a 96-well plate with serum at basal concentrations. Finally, 10 μM BrdU was added to the plate and cells were incubated for 4 hr. Data are shown as mean ± sd.

DXMS suppresses RhoA expression by inhibiting its transcription in OBs

We then explored the potential molecular target in DXMS-induced OB. DXMS was added to the culture medium and the OBs were harvested to evaluate RhoA expression. The results revealed that DXMS treatment had a negative modulation effect on RhoA expression (Figure 2A). DXMS remarkably suppressed the mRNA level after treatment for 24 h (Figure 2A column 4). 6–24 h after DXMS treatment, the protein level also confirmed the low RhoA level, revealing the negative impact of DSMS on RhoA protein level (Figure 2B; columns 2, 3 and 4). The synchronous pattern of mRNA and protein levels is important evidence that DXMS inhibits RhoA expression at the beginning of transcription.

Figure 2. DXMS suppress the expression of RhoA. (A) Total RNAs were extracted from OB at 6, 12, 24 h post treatment DXMS. The relative transcripts level verified by qPCR. (B) Protein level of RhoA confirmed by immunoblot.

DXMS suppresses RhoA expression though Nrf1 modulation in OBs

To investigate upstream molecules that regulate the transcription of RhoA, Nrf1 was identified by the DNA pull-down assay and the coding region of Nrf1 was cloned and constructed in the pET28a (+) vector and then expressed in BL21 (DE3) for Nrf1 protein. Apparently, the protein exhibits a strengthen signal with a decreased concentration of the cold probe (Figure 3A).

Figure 3. DNA-binding and transcription activation activity of NRF1. (A) EMSA analysis of the binding of NRF1 to RhoA promoter. Proteins were incubated with biotin-labeled probe in the absence or presence of 8- to 2-fold of unlabeled probes for 30 min. (B) DLR assay of NRF1 transcription factor activity in OB. Asterisks indicate statistically significant differences as determined by Student's t-test (^*P < 0.05, ^**P < 0.01).

The transcription factor activity of Nrf1 was examined using a dual-luciferase reporter (DLR) assay system in cultured cells. The coding sequence of Nrf1 was fused to the DNA sequence encoding the GAL4 DNA-binding domain under the control of the cytomegalovirus (CMV) promoter in pcDNA3.1. As shown in Figure 3B, Nrf1 was able to activate the reporter expression in vivo compared with the negative control (CMV-BD). Results with the transient expression assay also showed that overexpression of Nrf1 in rodent long bone shaft presents a higher Luc activity (Figure 4A, column 2) in comparison with the non-transgenic ones, revealing that Nrf1 promotes the RhoA promoter drive-Luc signal.

Figure 4. Epimedin C antagonizes the effects of DXMS. (A) Transient expression assay. Reporter Luc drive by RhoA promoter. (B) DLR assay with treatment of DXMS and/or Epimedin C. (C) IF staining for the endogenous RhoA in rodent long bone slice.

Epimedin C antagonizes the effects of DXMS induced OBs

Epimedin C was added to the culture medium externally and cell growth was determined in this study. A straight gap was made manually by scratching the surface of the cells, and the gap area was calculated. In the Epimedin C-treated groups, the migration rate was normalized and antagonized the influence of DXMS (Figure 1A and 1B).

A similar phenomenon was also visualized at the molecular level using DLR assays. The results revealed that Nrf1 activation was also enhanced by Epimedin C (Figure 4B, column 4) co-cultured with DXMS, which obviously functioned as an antagonist in the cellular model.

A transient expression assay was used to determine the Epimedin C function and the long bone shafts of rodents were sampled as the target region to calculate Luc reporter activity. The bones from mice administered Epimedin C were found to exhibit more intense Luc activity compared to those obtained from DXMS-treated mice (Figure 4A). Effect of Epimedin C on endogenous RhoA was evaluated by IF staining. The protein level of RhoA was remarkably decreased by external DXMS treatment, whereas Epimedin C restored the expression of RhoA (Figure 4C). Thus, our results confirmed that Epimedin C antagonized the effect of DXMS both in vitro and in vivo.

RhoA expression level was down regulated in CRISPR nrf1 cells

Internal Nrf1 expression was verified in CRISPR-nrf1 and the Osteoblast cells were also selected as samples to detect RhoA promoter activation. The RhoA reading frame was replaced by reporter Luc in RhoA promoter assay experiment. The origin signal RhoA in living Osteoblast was examined by fluorescence microscope, the protein RhoA was probed by antibody (Abcam, ab271951), as shown in Figure 5A. The RhoA signal in WT cells was as same as described in DXMS and Epimedin C treatment, while the signal in CRISPR group was lower than group DXMS/Epimedin C and Epimedin C treatment. These two groups can also be read but the Epimedin C exhibited lost its salvage function that the RhoA expression level was still lower in front of DXMS treatment in CRISPR-nrf1 cells.

Figure 5. Epimedin C lost its function without Nrf1 expression. (A) OB cells probed by RhoA antibody, WT and CRISPR-Nrf1 treated with Epimedin C and DXMS in different dosage. (B) The Luc signal captured in OB cells. WT and CRISPRE-Nrf1 group treated with Epimedin C and DXMS, the Luc ORF drive by RhoA promoter exhibits the positive signal in different level. (C) Compared with WT group, the CRISPR-Nrf1 OB cells got a lower expression level of RhoA protein, even in the Epimedin C treatment condition.

RhoA promoter activation was also assayed in these Osteoblast cells, the reporter Luc activity was read and shown that the CRISPR-nrf1 Osteoblast was not responding to the Epimedin C, and did not salvage the RhoA promoter activation in DXMS treatment (Figure 5B).

The native RhoA expression level was also examined in WT and CRISPRE-nrf1 OB cells, with and without the expression of Nrf1, the native RhoA level is different. The results can be verified as shown in Figure 5C, that with the comparison of WT cells, CRISPR-nrf1 cells got lower density of RhoA signal, although treated with Epimedin C only.

Cell isolation and culture

Mus musculus OBs were isolated from the long bones of 6-week-old mice, according to the protocol described by Bakker and Klein [42]. Cells were washed with PBS (Gibco, 10010-023, China) and centrifuged at 300 × g for 5 min, then inoculated into 25 cm² flasks and cultured at 37°C. Dulbecco’s modified eagle medium (DMEM, Gibco, 11885-092, China) supplemented with L-glutamine, and pyruvate was used for culture OBs.

An in vitro scratch assay was designed based on the method reported by Mouritzen [43]. OBs were isolated from rodents for 4 weeks, inoculated in flasks, and cultured with or without Epimedin C and DXMS. A manual scratch was made by the pipette tip in the middle of the culture flask when the cell was grown up.

Animals

Wildtype C57BL/6N (Jackson Laboratory, Bar Harbor, USA) male mice aged 4 weeks with body weights ranging from 20–25 g were used for all experiments. The mice were housed in a temperature-controlled room (20–24°C) and subjected to a 12 h light/dark cycle and provided free access to standard rodent chow and drinking water.

DNA delivery

JetPEI reagent (VWR, 89129-920, USA) was used to transiently express the Luc reporter gene driven by the Rho1 promoter in a rodent model. Primers used in promoter cloning were shown in Table 1. The modified pcDNA3.1-RhoA pro-Luc (50 μg) vector was mixed with in vivo-jetPEI reagent in 400 μL of 5% glucose solution and injected into the tail vein of mice for in vivo experiments. Bioluminescence imaging of luciferase expression in bone tissue was performed using the NightOWL II LB 983 imaging system (Berthold Technologies, Bad Wildbad, Germany).

DXMS and Epimedin C treatment

DMEM medium was mixed with DXMS at a concentration of 10 μM, pre-warmed at 37°C, and loaded into the culture flask. Cells were sampled at 0, 1, 6, 12, and 24 h. Epimedin C was dissolved in H2O, and the final concentration was 10 μM. The liquid was orally administered by dropping it directly into the mouth at a specified dosage. DXMS was dissolved in drinking water at a concentration of 10 μM and administered at a final dose of 1 × 10⁻⁴ μmol/mouse/day. Oral administration was performed for 4 weeks. A similar procedure was used for the administration of Epimedin C at a final dose of 1 × 10⁻³ μmol/mouse/day.

BrdU (Bromodeoxyuridine) cell proliferation assay

Detection of BrdU incorporation was performed by ELISA (Cat. No. #6813, Cell Signaling Technology, Danvers, MA, USA) to investigate the proliferation of OBs cell lines according to the manufacturer’s instructions. 5000 cells were seeded in 100 μl starvation medium (DMEM, 0.1% FBS, 100 U/ml penicillin and 100 mg/ml streptomycin) and incubated at 37°C for 24 h hours to synchronize the cell cycle. Medium was then changed to 1x BrdU solution prepared in regular growth medium (DMEM, 5% FBS, 100 U/ml penicillin and 100 mg/ml streptomycin) cultured for 6 h at 37°C to induce proliferation. 10 μM DXMS and 10 μM Epimedin C were used in cell treatment for 0–48 h. Subsequent procedure was performed according to the manufacturer’s instructions. The BrdU incorporation was measured at 450 nm with the TECAN GENios V4.62-07/01 microplate reader (Tecan, Reading, UK) and XFLUOR4 Version V 4.51 software (Tecan).

Real time-PCR

Total RNA was extracted using the PureLink^™ Total RNA Kit (Invitrogen, IS10006) according to the manufacturer’s instructions, and cDNA was synthesized using SuperScript^™ III First-Strand Synthesis SuperMix (Invitrogen, 11752050). The marker gene primers synthesized by Integrated DNA Technologies, Inc. (IDT) and the sequences are shown in Table 1. The results were quantified using the 2^−ΔΔC method by Ish-Shalom and Lichter [44]. Primers used in qPCR were shown in Table 1.

Immunoblotting

Whole protein lysates were separated using SDS-PAGE and transferred to PVDF membranes. The blots were incubated with antibodies (Cell Signaling Technology, Inc) directed against RhoA (26C4), GAPDH (D16H11), NRF1 (D9K6P), followed by incubation with the corresponding HRP-conjugated secondary antibody and detected using the Pierce^™ Fast Western Blot Kit (35050). The band intensities were quantified using ImageJ software (National Institutes of Health, USA).

DNA pull-down

The promoter region was amplified according to the NCBI Reference Sequence: NG_051308.1 and a 2,100 bp target DNA was acquired with a 5′ biotin label. The procedure for DNA pull-down screening was optimized using the procedure reported by Iwasaki [45] and Larsen [46]. The mixture of the DNA probe and total protein extracted from rodent OBs were incubated overnight at 4°C. The DNA-protein complexes were washed with Binding/Wash Buffer (Tris-buffered saline containing 0.1% Tween-20 detergent) and harvested using Pierce^™ streptavidin magnetic beads (88816).

The RhoA promoter was cloned and labeled with biotin adhered to streptavidin magnetic beads used as bait. Total protein extracted from cultured OBs was mixed with the bait and incubated. Candidate proteins in the elution of the DNA pull-down were identified using MS. Nrf1 was identified among the candidate binding proteins. Next, the binding ability was confirmed using EMSA.

MS sample processing and analytical methods

The samples prepared in this study were modified based on the method reported by Hou [47]. The sample was precipitated by adding five volumes of cold acetone and incubating overnight at −20°C. The precipitated protein was washed twice with 80% cold acetone. Then, 10 μL each of 6 M urea, 2 M thiourea, and 100 mM ammonium bicarbonate were added to the pellet. The solubilized protein was reduced by DTT and alkylated by IAA. Next, trypsin was added for digestion and allowed to react overnight. The digested peptide was C18 zip-tip desalted, lyophilized, and suspended in 10 μL of 5/0.1% acetonitrile/formic acid.

LC-MS+MSMS (TimsTOF pro)

Suspended peptide (1 microliter) was loaded onto a C18 column and eluted using a stepwise gradient with acetonitrile at a flow rate 400 nL/min. A Bruker nanoElute system linked to a time TOF Pro mass spectrometer was used. The LC gradient conditions were as follows: the initial conditions were 2% B (A: 0.1% formic acid in water, B: 99.9% acetonitrile, 0.1% formic acid), followed by a 4.5 min ramp to 17% B. 17–25% B over 8.5 min, 25–37% B over 4.5 min, a gradient of 37–80% B over 2 min, hold at 80% B for 5.5 min. The total run time was 25 min. MS data were collected over an m/z range of 100 to 1700. During MS/MS data collection, each TIMS cycle included 1 MS + an average of 10 PASEF MS/MS scans. The raw data were searched using PEAKS with the Uniport Mus/Homo protein database. Data were searched with trypsin as an enzyme; two missed cleavages were allowed, carbamidomethyl cysteine was a fixed modification, oxidized methionine, acetylation on protein N terminus variable mod, 15 ppm mass tolerance on precursor ions, and 0.1 Da on-fragment ions.

Electrophoretic mobility shift assay (EMSA)

EMSA was carried out using biotin-labeled probes and the Pierce Light Shift Chemiluminescent EMSA kit (Thermo Scientific, 89880, Rockford, IL, USA). His-tagged Nrf1 proteins were expressed and purified. Promoter sequences and synthesized DNA fragments were used as probes. Unlabeled DNA fragments were used as cold probes. The binding reaction was carried out in a 20 μl reaction mixture at room temperature for 30 min and then separated by 6% native polyacrylamide gel in 0.53 Tris-borate/EDTA buffer. The bands were detected according to the instructions provided with the EMSA kit.

Dual-luciferase reporter (DLR) assay

To verify transcription factor activity in OBs, the coding region of the nfe2l1 gene was cloned into the expression vector pcDNA3.1-BD to generate the BD-nfe2l1 effector plasmid. Gal4-Luc was used as a reporter in the same backbone. The Renilla Luc gene served as an internal control. The DLR assay was conducted according to the method described by Ohta [48]. OBs were transfected using a Lipofectamine 3000 (Thermofisher, L3000015, Carlsbad, CA, USA) -mediated transient expression system. The co-transformed cultures were kept in the dark for 16 h at 37°C. Luciferase assay was performed using the Promega DLR assay system and values were measured using a GloMax 20–20 luminometer (Bio-Rad, Hercules, CA, USA).

CRISPR vector construction and cell verification

Original vector containing Cas91.1 expression frame and gRNA editing region purchased from Addgene (#71814), gRNA designed online and cloned into vector Cas91.1, verified by sequencing. The final sequence of Nrf1 (XM_027775873.1) gRNA is 5′-GGATTAGACTCAAACACGTG-3′ according to its score calculated on https://www.genscript.com/gRNA-design-tool. Purified plasmid of the final version was transfected into cells by JetPEI reagent as described in above chapter.

The positive signal, e.g., EGFP was verified by fluorescence microscope. Also, the signal in living cells e.g., Luc activity can be examined in the 96-well culture plate by adding additional Luciferin substrate, and the signal was read by NightOWL II LB 983 system (Berthold Technologies).

Promoter isolation and transient expression assay

An in vivo experiment was also designed to verify the Nrf1 function in the RhoA promoter. A transient expression assay was performed using the luciferase coding region as a reporter. Luc fluorescence signals were captured using a low-light cooled CCD camera. The promoter sequence of the RhoA gene was cloned by PCR-based genome walking. The infusion system was used following the instruction manual (Clontech, San Jose, CA, USA) for the construction of vectors. The 1,193 bp promoter sequence was cloned into the pcDNA3.1 vector containing a Luc reporter gene to generate RhoA Pro: Luc as reporters. The coding sequence of nfe2l1 gene was cloned into normal pcDNA3.1 to generate the constructs as effectors. For the expression assay in bone tissue, a JetPEI-mediated transfection system was used for rodent transfection and the bone tissue was isolated as samples to verify the expression of the Luc reporter. The detached bone shafts were sprayed with 1 mM D-luciferin (Promega, P1042), the Luc signal was captured using a low-light cooled CCD camera and the relative Luc activity was measured. Quantitative analysis was performed using IndiGo software (Berthold Technologies, Oak Ridge, TN, USA).

Rodent long bone shaft immunofluorescence staining

Rodent long bone shafts were dissected and sliced according to the process described by Amend [49]. The middle of the tissue with the highest density was sliced as samples for observation. The primary antibody (Cell Signaling Technology, 2117) and the secondary antibody conjugated with Alexa Fluor 488 conjugated (Cell Signaling Technology, 4412) for IF staining observation.

The figures shown in this study, error bars represent the SD of three biological replicates with three technical repeats each. Asterisks indicate statistically significant differences as determined by Student’s t-test (^*P < 0.05, ^**P < 0.01).