Let-7f-5p regulates TGFBR1 in glucocorticoid-inhibited osteoblast differentiation and ameliorates glucocorticoid-induced bone loss

Osteogenic differentiation and treatment

For the osteogenic differentiation, unattached cells were removed and replaced with osteogenic induction medium (10% FBS in a‐MEM containing 25mg/mL Vit C and 5mM β‐Glycerophosphate) with or without Dex (0.1 μM) in 12-well cell culture plates as previously described ¹⁸, which were then replaced with the osteogenic induction medium and Dex every 2-3 days. After 5 to 7 days in osteogenic induction medium, the cells were used for alkaline phosphatase (ALP) staining and ALP activity assay. Mineralization typically occurs after 10 to 14 days in culture, and the cells were stained with the Von Kossa method for measurement of mineralized nodule formation.

ALP staining, ALP activity, and mineralization assay

After osteogenic induction for 7 days, we fixed the cells with 4% formaldehyde (Sigma, Shanghai, China) for 15 min at room temperature, and incubated them with ALP substrate, BCIP/NBT (Thermo Scientific Waltham, MA). To test the ALP activity, we lysed the cells by using a radio immunoprecipitation assay (RIPA, Beyotime, Shanghai, China) and then determined the ALP activity through performing an ALP Activity Assay (Beyotime). For measurement of mineralized nodule formation, cells were fixed with 4% formaldehyde and washed with PBS for 3 times. Then, we incubated the cells with a 5% silver nitrate solution and exposed them under the light for 30 min. Lastly, we used a 5% sodium thiosulfate to remove the nonspecific staining for 5 min.

Prediction of let-7f-5p target genes

Let-7f-5p target genes and the binding sites were predicted by using diverse bioinformatic platforms, such as TargetScan 7.2 (http://targetscan.org), miRBase, miRDB, miRanda, etc.

MiRNAs and reporter vectors construction

Let-7f-5p, mutation constructs, and reporter gene construction were performed according to previous studies ²⁰. Briefly, we used genomic DNA from mouse as the template and the genomic fragments of let-7f-5p precursors were amplified by reverse transcription PCR (RT-PCR). Next, we cloned the amplified products into the restriction sites (NotI and XhoI) of pLAS2-RFP vector. Then BMSCs was virally infected with the modified vector and let-7f-5p expression was detected by quantitative RT-PCR (qRT-PCR). Additionally, we cloned the let-7f-5p's binding site in TGFRB1 and the whole TGFRB1 3′UTR sequence into the restriction sites (PmeI and XhoI) of pmirGLO luciferase vector. Also, a pair of primers with mutant sequence were used to generate the mutation constructs of TGFRB1 3′UTR.

Transfection of let-7f-5p mimics and antagomiR

MiRIDIAN miRNA mimics were used to design the let-7f-5p overexpression. Anti-let-7f-5p miRNA inhibitors (AntagomiR) were purchased from Dharmacon (Denver, CO). BMSCs were transfected for 24 h with let-7f-5p mimics (100 nM), let-7f-5p antagomiR (100 nM) or miR-NC (negative control, 100 nM) through Lipofectamine 2000 reagent. Then the cells were used for the following experiments.

RT-PCR and qRT-PCR

RT-PCR and qRT-PCR were performed as described elsewhere ²⁰. Firstly, total RNA was extracted from BMSCs or bone tissues with Trizol (Sigma). Then, to obtain cDNA, we diluted 1 μg of RNA with 10 ml of nuclease-free water. Then we added into 1 μl of 50 mM hexamer primers. Next, the denatured process of the solution was performed with respective temperature and time point sequentially (65°C, 5 min; and 4°C, 60 min). Lastly, the solution was incubated with at 25°C for 10 min, 45°C for 60 min and and 75°C for 5 min. The primers for qRT-PCR were listed at Table S1. Additionally, for miRNA qRT-PCR assay, total RNA was isolated and the small RNA fractions were enriched using mirVana miRNA Isolation Kit. Next, the cDNA was prepared with the TaqMan miRNA Reverse Transcription Kit and RT-PCR was performed. Lastly, let-7f-5p expression level was quantified by TaqMan U6 snRNA assay.

Western blot analysis

To investigate protein expression levels in cells and bones, western blot was performed. The protein samples were subjected with equal protein concentration to SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to PVDF membranes (Milipore, Darmstadt, Germany). Next, membranes were blotted with 5% skim milk-PBS-Tween 20 for 1 h at room temperature and incubated with primary antibodies overnight at 4°C. Rabbit anti-TGFBR1 (1:1,000, sc-101574, Santa Cruz) and rabbit anti-GR (1:1000, sc-56851, Santa Cruz) were used. After that, the blots were washed with PBS-Tween 20 (PBST) and incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (1:3000) for 1 h at room temperature. Lastly, the blots were washed again and incubated with enhanced chemilumescent (ECL) substrates (Bio-rad) for 1 min and the Image J software was applied to analysing the blots.

Gene overexpression and knockdown assay

For the overexpression of TGFBR1, TGFBR1 CDS region was amplified from mouse cDNA and cloned into pCS2+ plasmid vector which included the CMV promoter. For the silence of TGFBR1 and GR, small interfere RNA (siRNA), composed of several target-specific 19-25-nucleotide siRNAs, was used (Santa Cruz, Dallas, Texas) and BMSCs were transfected with Lipofectamine RNAimax (Invitrogen) according to the manufacturer's instruction. The overexpression and knockdown efficiency of TGFBR1 and GR were examined using qRT-PCR.

Dual-luciferase reporter assay

Luciferase activity of the reporter constructs were evaluated by performing the dual-luciferase reporter assay (Promega, USA). In brief, BMSCs (80% confluence) were seeded in the 6-well plates and were induced the differentiation for 24 h. Then the cells were cotransfected with 100 nM of let-7f-5p mimics, 1 μg of TGFBR1 3′UTR or the control, and miR-NC per well by using Lipofectamine 2000. After 24 h of transfection, cell extracts were prepared for measuring the luciferase activity through the Dual-Luciferase Reporter Assay System (Promega, Biotech Co., Ltd., China).

ChIP assay

Chromatin immunoprecipitation (ChIP) assay was carried out in line with the manufacturer's instructions (Millipore, Darmstadt, Germany). After the intervention process, BMSCs were fixed through 1% paraformaldehyde and then lysed with SDS lysis buffer. Followed by which, the lysate was sonicated (16 rounds of 20 pulses with 2 minutes between rounds) and the protein-DNA mixture were immunoprecipitated by using IgG or anti-GR antibody at 4°C overnight. Then the products were collected after incubation on protein A + G-coated magnetic beads. The beads were washed and ChIP elution buffer was used to elute the bound chromatin. Proteinase K was applied to digest the protein for 4 h at 45°C. DNA purification kit was used to purify the DNA. The DNA fragments of the GR binding sites in the let-7f-5p promoter were designed and synthesized by RiboBio (Guangzhou, China). Next, qPCR was applied to test the GR binding site. The total chromatin was regarded as input and IgG was used as controls.

GC-induced bone loss

For GIOP animals model, 8-week-old male C57BL/6 mice (n=6 per group) (Experimental Animal Center of Guangzhou University of Chinese Medicine, Guangzhou, China) were intraperitoneally administered with Dex (the First Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, China) daily at a dose of 1 mg/kg body weight as described ^34-36. Animals were housed in cages under pathogen-free conditions under a 12-hlight and 12-h dark cycle in the animal room of the First Affiliated Hospital of Guangzhou University of Chinese Medicine (approval no. SYXK [Yue] 2018-0001) under a 12-h/12-h light/dark cycle (lightson at 08:00) with free access to a standard rodent diet and water. After 8 weeks of treatment, vertebrae were dissected and used for micro-CT and histomorphometric analysis.

Micro-CT analysis

Micro-CT examination was accomplished as described previously ³⁷ through a Skyscan 1172 micro-CT imaging system (Skyscan, Kontich, Belgium) with a 12 mm spatial resolution (X-ray source 80 kV/100μA). Volume reconstruction was achieved via the built-in software NRecon 1.6 and CTAn 1.8, respectively. To analyze the bone remodeling, the volume of interest (VOI) was characterized as a cylindrical space wrapping the cancellous bone. Trabecular bone volume (BV/TV, %), trabecular number (Tb.N, /mm), trabecular thickness (Tb.Th, mm), and trabecular separation (Tb.Sp, mm) were calculated within the delimited VOI.

Bone histomorphometry

For histological analysis, the mice were injected with 25 mg/kg calcein at 8 and 2 days before euthanasia. Preparation of undecalcified vertebrae samples and trabecular bone histomorphometry were executed in accordance with previous studies ³⁸. Bone samples were fixed with 10% paraformaldehyde overnight and reserved in 70% ethanol at 4℃, and then were embedded with methylmethacrylate. Five-mm-thick pieces were processed through the microtome. Histomorphometric analysis of the L4 vertebrae mainly contained bone formation parameters bone formation rate (BFR/BS, μm³ μm^-2 per day), mineral apposition rate (MAR, μm per day), mineralizing surface (MS/BS, %), and osteoblast number (N.Ob/B.Pm, /mm). All above parameters were measured and expressed as stated in ASBMR nomenclature committee ³⁹.

ELISA

The concentration of serum procollagen type 1 N-terminal propeptide (P1NP) and osteocalcin (OCN) were measured using commercial enzyme-linked immunosorbent assay (ELISA) kits from IDS (Fountain Hills, AZ). Blood was collected by puncturing the mice cheekpouch. And the mice were fasted for 4 h before collecting blood.

Let-7f-5p negatively regulated TGFBR1 and promoted osteogenic differentiation of Dex-treated BMSCs

To investigate the regulatory effect of let-7f-5p in Dex-inhibited osteogenic differentiation of BMSCs, we carried out gain-of-function experiments through transfecting let-7f-5p mimics into BMSCs prior to Dex-inhibited osteogenic differentiation of BMSCs (Figure ​Figure22A). Osteogenic differentiation and mineralisation activity of BMSCs as demonstrated by ALP staining (Figure ​Figure22B), Von Kossa staining (Figure ​Figure22C), and quantitative ALP activity (Figure ​Figure22D), were promoted by let-7f-5p mimics. And the marker genes (Runx2, Osx, Alp, and Ocn) of osteoblast differentiation exhibited increased expression after transfection of let-7f-5p mimics (Figure ​Figure22E). Further, overexpression of let-7f-5p markedly decreased TGFBR1 expression at the mRNA and protein levels (Figure ​Figure22F-H). Our data indicated that let-7f-5p restores Dex-inhibited osteogenic differentiation of BMSCs by targeting the 3′-UTR of TGFBR1. In addition, let-7f-5p was also shown to downregulate TGFBR1 expression and promoted promote the development of osteogenic differentiation phenotypes of BMSCs without Dex treatment (Figure S2).

Open in a separate window

Figure 2

Let-7f-5p promotes osteogenic differentiation of BMSCs by targeting TGFBR1 under Dex conditions. (A) Increased let-7f-5p expression in BMSCs transfected with let-7f-5p mimics. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^***P < 0.001 by t test. (B-C) ALP staining and Von Kossa staining of BMSCs transfected with let-7f-5p mimics with Dex treatment. (D) Quantitative analysis of the ALP activity in BMSCs transfected with let-7f-5p mimics with Dex treatment. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^***P < 0.001 by t test. (E) Increased mRNA expression of Runx2, Osx, Alp, and Ocn in BMSCs transfected with let-7f-5p mimics with Dex treatment. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^*P < 0.05 and ^**P < 0.01 by t test. (F-H) let-7f-5p inhibited mRNA expression and protein expression of TGFBR1 in BMSCs under Dex conditions. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^**P < 0.01 and ^***P < 0.001 by t test.

Overexpression of TGFBR1 reversed the effects of let-7f-5p upregulation on osteogenic differentiation of Dex-treated BMSCs

To examine whether TGFBR1 gain-of-function can reverse the promotion of Dex-treated osteogenic differentiation by let-7f-5p mimics, we co-transfected the TGFBR1 overexpression plasmid and let-7f-5p mimics into BMSCs. TGFBR1 expression was markedly upregulated in comparison to the controls (Figure ​Figure33A-C). We next evaluated the effects of overexpressing TGFBR1 together with let-7f-5p mimics on osteogenic differentiation and mineralisation activity of Dex-treated BMSCs. Compared with the control group, reduced levels of ALP staining, fewer mineralization nodules, and decreased ALP activity were observed in the group with TGFBR1 overexpression and let-7f-5p mimics (Figure ​Figure33D-F). In addition, the upregulation of osteospecific genes including Runx2, Osx, Alp, and Ocn messenger RNA induced by let-7f-5p mimics in BMSCs was also reversed by a combination of TGFBR1 overexpression and let-7f-5p mimics in BMSCs (Figure ​Figure33G).

Open in a separate window

Figure 3

Overexpression of TGFBR1 reverses the effect of the upregulation of let-7f-5p on osteogenic differentiation of BMSCs under Dex conditions. (A-C) Increased mRNA and protein expression of TGFBR1 in BMSCs transfected with let-7f-5p mimics with overexpression of TGFBR1. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^**P < 0.01 and ^***P < 0.001 by t test. (D-E) ALP staining and Von Kossa staining of BMSCs transfected with let-7f-5p mimics with overexpression of TGFBR1 under Dex conditions. (F) Quantitative analysis of the ALP activity in BMSCs transfected with let-7f-5p mimics with overexpression of TGFBR1 under Dex conditions. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^**P < 0.01 by t test. (G) Inhibited mRNA expression of Runx2, Osx, Alp, and Ocn in BMSCs transfected with let-7f-5p mimics with overexpression of TGFBR1 under Dex conditions. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^*P < 0.05 by t test.

Knockdown of TGFBR1 reversed the effects of downregulation of let-7f-5p on osteogenic differentiation of Dex-treated BMSCs

Next, we examined the role of TGFBR1 loss-of-function in reversing the inhibition of osteogenic differentiation of Dex-treated BMSCs with let-7f-5p antagomiR by co-transfecting TGFBR1 siRNA and let-7f-5p antagomirs into BMSCs. Let-7f-5p and TGFBR1 expression levels were verified after transfection (Figure ​Figure44A-C). The effects of TGFBR1 knockdown combined with let-7f-5p antagomiR on osteogenic differentiation and mineralisation activity of Dex-treated BMSCs were examined. The let-7f-5p antagomiR significantly inhibited ALP positive staining, mineralisation nodule formation, and ALP activity. The group treated with let-7f-5p antagomir and TGFBR1 siRNA exhibited increased ALP staining, more mineralisation nodules, and increased ALP activity relative to the controls (Figure ​Figure44D-E). In addition, expression trends of osteospecific genes (Runx2, Osx, Alp, and Ocn) were consistent with the results of ALP staining and the mineralisation assay (Figure ​Figure44F-I). These results indicated that TGFBR1 acted as a negative regulator for the promotion of osteogenic differentiation of Dex-treated BMSCs by let-7f-5p.

Open in a separate window

Figure 4

Knockdown of TGFBR1 reverses the effect of the downregulation of let-7f-5p on osteogenic differentiation of BMSCs under Dex conditions. (A) mRNA expression of let-7f-5p in BMSCs. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^***P < 0.001 by ANOVA with Tukey's post hoc test. (B-C) mRNA and protein expression of TGFBR1 in BMSCs. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^***P < 0.001 by ANOVA with Tukey's post hoc test. (D) ALP staining and Von Kossa staining of BMSCs transfected with let-7f-5p antagomiR with siTGFBR1 under Dex conditions. (E) Quantitative analysis of the ALP activity in BMSCs transfected with let-7f-5p antagomiR with siTGFBR1 under Dex conditions. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^*P < 0.05 and ^**P < 0.01 by ANOVA with Tukey's post hoc test. (F-I) mRNA expression of Runx2, Osx, Alp, and Ocn in BMSCs transfected with let-7f-5p antagomiR with siTGFBR1 under Dex conditions. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^*P < 0.05 and ^**P < 0.01 by ANOVA with Tukey's post hoc test.

GR suppressed let-7f-5p expression under Dex treatment

Most roles of GC are modulated by the GR, an extensively expressed nuclear receptor. GCs suppress bone formation by attenuating osteoblast differentiation via monomeric GR ². To determine whether the GR directly regulated let-7f-5p expression, we performed loss-of-function experiments using siRNAs to knock down GR expression in the BMSCs (siGR), and then confirmed the knockdown effect (Figure ​Figure5H5H and J) . The control group was termed “siCtrl”. Next, we verified transcriptional regulation through ChIP experiment using an anti-GR antibody on the let-7f-5p promoter. The results indicated that GR binding was significantly reduced at the let-7f-5p promoter in the siGR cells compared to siCtrl cells (Figure ​Figure55A). In addition, GR knockdown markedly increased the level of cellular let-7f-5p expression (Figure ​Figure55B).

Open in a separate window

Figure 5

GR mediates transcriptional silencing of let-7f-5p during Dex-inhibited osteogenic differentiation of BMSCs. (A) The predicted GR binding site in let-7f-5p promoter and the mutant versions are shown. (B) ChIP assay revealed the significant reduction in binding of GR with the let-7f-5p promoter in BMSCs transfected with siGR under Dex conditions. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^**P < 0.01 by ANOVA with Tukey's post hoc test. (C) siGR promoted the mRNA expression of let-7f-5p under Dex conditions. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^***P < 0.001 by ANOVA with Tukey's post hoc test. (D) Quantitative analysis of the ALP activity in BMSCs transfected with siGR under Dex conditions. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^**P < 0.01 by t test. (E) ALP staining and Von Kossa staining of BMSCs transfected with siGR under Dex conditions. (F) Increased messenger RNA expression of Runx2, Osx, Alp, and Ocn in BMSCs transfected with siGR under Dex conditions. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^*P < 0.05 and ^**P < 0.01 by t test. (G-J) Reduced mRNA and protein expression of TGFBR1 and GR in BMSCs transfected with siGR under Dex conditions. Results are shown as mean ± s.d. of three independent experiments performed in triplicates, ^*P < 0.05, ^**P < 0.01, and ^***P < 0.001 by t test.

To elucidate the effect of GR-mediated let-7f-5p in Dex-inhibited osteogenic differentiation of BMSCs, we performed downstream functional analysis, including assays of osteogenic differentation, mineralisation activity, osteogenic differentation marker expression level, and TGFBR1 expression level, by transfecting siGR into BMSCs. Levels of ALP activity (Figure ​Figure55C), ALP staining (Figure ​Figure55D), Von Kossa staining (Figure ​Figure55E), and expression of Runx2, Osx, Alp, and Ocn mRNA increased in the GR-knockdown cells (Figure ​Figure55F). Further, GR loss-of-function markedly inhibited the mRNA and protein expression levels of TGFBR1 (Figure ​Figure55G-I). Taken together, these observations suggest that GR may suppress let-7f-5p expression at the transcriptional level during Dex-inhibited osteogenic differentiation of BMSCs.

This work was generously supported by Guangdong Province Universities and Colleges Pearl River Scholar Funded Scheme (2018), National Natural Science Foundation of China (81674000, 81774338), The youth scientific research training project of GZUCM (2019QNPY04), Key Project of Basic Research and Applied Basic Research of Department of Education of Guangdong Province (2018KZDXM021), Excellent Young Scholars Project of China Association of Traditional Chinese Medicine (CACM-2017-QNRC1-01), Science and Technology New Star of Guangzhou Pearl river (201710010078), Scientific Research Project of Guangdong Traditional Chinese Medicine Bureau (20180330134046), Guangdong Medical Research Fund (A2019167), Youth Research Fund of the First Affiliated Hospital of Guangzhou University of Traditional Chinese Medicine (2019QN17), Natural Science Foundation of Xinjiang Autonomous Region (2019D01C013, 2018D01C011), Excellent Doctoral Dissertation Incubation Grant of First Clinical School of Guangzhou University of Chinese Medicine (YB201602, YB201702), Excellent Young Scholars Project of the First Affiliated Hospital of Guangzhou University of Chinese Medicine (2017QN08, 2017TD08).