Neuropeptide Y1 receptor antagonist promotes osteoporosis and microdamage repair and enhances osteogenic differentiation of bone marrow stem cells via cAMP/PKA/CREB pathway

OP group showed deteriorated bone microstructure and more microdamage than OA

There was no statistical difference with regard to the age of the patients between OA group and OP group (69.00±5.31 vs. 68.30±5.95 years, respectively, P=0.7845). The specimens were analyzed to perform the 3-D reconstruction of trabecular bone via micro-CT. As shown in Figure 1B, the OP group represented a marked deterioration in the microstructure of trabeculae. Compared to OA group, OP group showed significant decreases in BV/TV, Tb.Th, Tb. N and Conn.D but significant increase in Tb.Sp (Table 1).

Figure 1. OP group showed deteriorated bone microstructure and more microdamage than OA. (A) The position of the samples for micro-CT and microdamage analysis within the femoral head. (B) Representative micro-CT images of tibias, and comparison of the microstructure parameters between two groups. (C) Fluorescent images and comparison of microdamage (white arrow head) in the OA group and OP group (x 200). The data are expressed as the means ± SD (n = 10 in each group). *P<0.05; ***P<0.001; ****P<0.0001 vs. SHAM by the unpaired two-tailed Student’s t-test.

The microcracks of the specimens were imaged by red fluorescence and purple fluorescence. As shown in Figure 1C, significant increases were observed in the Cr.Le, Cr.Dn and Cr.S.Dn in OP group.

NPY and Y1R were upregulated in subchondral cancellous bone of OP patients and OVX rats

The expression of NPY and Y1R in bone samples of patients and rats were evaluated by immunohistochemical analysis (Figure 2). Compared to OA group, the subchondral cancellous bone of OP patients exhibited stronger immunoreactivity for both NPY and Y1R (Figure 2A). Similarly, NPY and Y1R immunoreactivity in bone were stronger in the OVX rats than that in the SHAM rats (Figure 2C). In the OP patients and OVX rats, correlation analysis showed that the MOD of NPY and Y1R were negatively correlated with BV/TV, respectively (Figure 2B, 2D). It was suggested that the expressions of NPY and Y1R in bone tissue were associated with the bone metabolism in osteoporosis.

Figure 2. NPY and Y1R were upregulated in subchondral cancellous bone of OP patients and OVX rats. Immunohistochemical analysis of bone tissue between OA and OP group (A), and between SHAM and OVX group (C) for NPY and Y1R (x 400). Correlation between BV/TV and the expression of NPY and Y1R in OP group (B) and OVX group (D). The data are expressed as the means ± SD. r=correlation coefficient assessed by Pearson correlation analysis. **P<0.01; ***P<0.001; ****P<0.0001 by the unpaired two-tailed Student’s t-test.

Y1R antagonist treatment improved bone microstructure and decrease microdamage in OVX rats

The micro-CT images of right tibias of rats clearly displayed the differences of microstructure among the groups (Figure 3A). As expected, the OVX rats exhibited remarkable reductions of BMD, BV/TV, Tb.Th, Tb.N and Conn.D. And the osteoporotic changes of bone microstructure in OVX rats were greatly improved after Y1R antagonist BIBO3304 treatment (Figure 3B) (Table 2).

Figure 3. Y1R antagonist treatment improved bone microstructure and decrease microdamage in OVX rats. Representative micro-CT images of rat tibias (A) and quantitative analysis of bone microstructure (B). (C) In vivo fatigue loading model and the position of the sample for microdamage analysis in the rat tibia. Fluorescent images (D) and comparison (E) of microdamage (white arrow head) among groups (x 200). The data are expressed as the means ± SD (n = 6 in each group). *P<0.05; ***P<0.001; ****P<0.0001 vs. SHAM, and ^##P<0.01; ^####P<0.0001 vs. OVX by one-way ANOVA and Tukey’s post hoc test.

After in vivo fatigue loading, the middle segment of tibia specimens was obtained to evaluate the bone microdamage (Figure 3C). Cyclic loading induced microdamage and leaded to a significant increase in Cr.Le, Cr.Dn and Cr.S.Dn in OVX rats compared to the SHAM rats. In group treated with BIBO3304, the Cr.Le, Cr.Dn and Cr.S.Dn in OVX+ BIBO3304 rats were significantly decreased compared to OVX rats, suggesting that Y1R antagonist treatment prevented the load-induced increase in bone microdamage (Figure 3D–3E).

Y1R antagonist treatment promoted bone formation and inhibit bone resorption in OVX rats

The Alizarin Red S and TRAP staining were performed to evaluate the bone formation and resorption activities among groups. The calcified nodules in bone tissues would be stained by Alizarin Red S with the color of deep red, which reflects the activity of osteoblasts and bone formation [22]. As shown in Figure 4A, it is apparently that the area of calcified nodules was higher in BIBO3304 treatment groups than OVX group, which indicated that bone mineralization was improved with BIBO3304 treatment. TRAP as a marker enzyme of osteoclasts, was significantly down-regulated in the OVX rats with BIBO3304 treatment when compared with group OVX. The immunohistochemical analysis showed that OVX+BIBO3304 group exhibited stronger immunoreactivity for RUNX2 and lower for immunoreactivity for MMP9 than those in OVX group (Figure 4B).

Figure 4. Y1R antagonist treatment promoted bone formation and inhibit bone resorption in OVX rats. (A) The Alizarin Red S (yellow arrow head) and TRAP staining (red arrow head) of bone tissues in groups. (B) Immunohistochemical analysis of bone tissue among groups for MMP9 and RUNX2 (x 400). (C, D) Western blotting results of MMP, RUNX2, Cath-K, OPG and RANKL expression in bone marrow from rat femurs. The data are expressed as the means ± SD (n = 6 in each group). ***P<0.001; ****P<0.0001 vs. SHAM, and ^#P<0.05; ^####P<0.0001 vs. OVX by one-way ANOVA and Tukey’s post hoc test.

The bone marrow from rat femurs was extracted and analyzed by western blotting. The results showed that the expression level of RUNX2 and OPG enhanced, and MMP9 and Cath-K decreased apparently in OVX+BIBO3304 group than those in the OVX group. Besides, the RANKL/OPG ratio in OVX+BIBO3304 group was markedly decreased in comparison with that in OVX rats (Figure 4C, 4D). The results indicated that treatment with BIBO3304 inhibited the bone resorption, and promoted bone formation meanwhile in OVX rats.

Characterization and identification of BMSCs

BMSCs at passage 3 were characterized by flow cytometry to detect the expression of the surface markers. As shown in Figure 5A, a high percentage of cells expressed the mesenchymal stem cell markers CD29, CD44, and CD90, and cells barely expressed the hematopoietic markers CD11b/C, CD45 and CD34. The Primary rat BMSCs displayed fibroblast-like morphologies under the microscope. And the BMSCs were successfully differentiated into osteogenic, adipogenic, and chondrogenic lineages in vitro under suitable differentiation conditions (Figure 5B). The results indicated that the cells we cultured were BMSCs with multilineage differentiation potential.

Figure 5. Y1R antagonist treatment reversed NPY-mediated suppression of osteogenesis and cAMP/PKA/CREB pathway in BMSCs. (A) Flow cytometric analysis of BMSC surface markers. (B) The multidifferentiation potential of BMSCs in vitro (Alizarin red S staining for osteogenic differentiation, Alcian blue staining for chondrogenic differentiation, and oil Red O staining for adipogenic differentiation). (C) Western blotting results of OPG, RANKL, RUNX2, OCN, p-PKAs and p-CREB expression in BMSCs after 3 days of incubation in osteogenic medium with NPY or NPY+BIBO3304. (D, E) The cAMP concentrations and ALP activity were measured on days 3 of incubation in osteogenic medium with NPY or NPY+BIBO3304. The data are expressed as the means ± SD. *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001 vs. CON, and ^#P<0.05; ^##P<0.01; ^####P<0.0001 vs. NPY by one-way ANOVA and Tukey’s post hoc test.

Y1R antagonist treatment reversed NPY-mediated suppression of osteogenesis and cAMP/PKA/CREB pathway in BMSCs

With regard to the effects of NPY on BMSCs osteogenesis, BMSCs were cultured under osteogenic medium for 3d with and without the treatment of NPY or Y1R antagonist BIBO3304. The ALP activity was assessed among groups. And the protein expression levels of osteogenic marker RUNX2 and OCN, and the regulator of osteoclastogenesis RANKL/OPG were determined by western blotting. As shown in Figure 5C, 5E, the ALP activity was decreased, and the expressions of RUNX2 and OCN were significantly downregulated in NPY-treated BMSCs compared with the control group. Conversely, the RANKL/OPG ratio was higher in NPY-treated BMSCs. With the Y1R antagonist BIBO3304 treatment, the effects of NPY on BMSCs were reversed.

In addition, the expression of cAMP, p-PKAs and p-CREB were determined. The results indicated that NPY treatment clearly decreased the expression levels of cAMP, p-PKAs and p-CREB, and the decreases were partially attenuated by treatment with Y1R antagonist (Figure 5C, 5D). The results showed that NPY inhibited the BMSCs osteogenesis through activation of Y1R, which might be involved with the cAMP/PKA/CREB pathway.

NPY inhibited osteogenesis and elevated RANKL/OPG ratio through Y1R via suppression of cAMP/PKA/CREB pathway in BMSCs

To investigate the relationship between the cAMP/PKA/CREB pathway and the Y1R in BMSCs, cells were treated with either the Y1R antagonist BIBO3304 or the PKA activator 8-Bromo-cAMP for 7d and 21d with the addition of NPY. Firstly, we evaluated the expression of RUNX2 and OCN, RANKL/OPG, and p-PKAs and p-CREB by western blotting. As shown in Figure 6A, 6C, NPY downregulated the expression levels of osteogenic marker RUNX2 and OCN, and p-PKAs and p-CREB at days 7 and 21, and increased the RANKL/OPG ratio. Both the BIBO3304 and 8-Bromo-cAMP reduced the effects of NPY. And then we evaluated the osteogenesis and mineralization of BMSCs through ALP and Alizarin Red S staining. Similarly, treatment with both the BIBO3304 and 8-Bromo-cAMP significantly attenuated the inhibitory effects of NPY on the osteogenesis of BMSCs at day 7, and mineralization at day 21 (Figure 6B).

Figure 6. NPY inhibited osteogenesis and elevated RANKL/OPG ratio through Y1R via suppression of cAMP/PKA/CREB pathway in BMSCs. (A, C) Western blotting results of OPG, RANKL, RUNX2, OCN, p-PKAs and p-CREB expression in BMSCs after day 7 and 21 of incubation in osteogenic medium with NPY, NPY+BIBO3304 or NPY+cAMP. (B) ALP staining on day 7, and Alizarin red S staining of BMSCs on day 21 of incubation in osteogenic medium with NPY, NPY+BIBO3304 or NPY+cAMP. The data are expressed as the means ± SD. *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001 vs. CON, and ^#P<0.05; ^##P<0.01; ^###P<0.001; ^####P<0.0001 vs. NPY by one-way ANOVA and Tukey’s post hoc test.

Clinical specimens

The subchondral cancellous bone specimens were obtained from 10 patients who underwent total hip arthroplasty for primary osteoarthritis OA (mean age 69, range 63–78 years) and 10 patients who underwent total hip arthroplasty for hip osteoporotic fracture (mean age 68.3, range 61–77 years). The sample size estimation was calculated on the basis of the previously published data [23] comparing BV/TV and trabecular thickness (Tb.Th) in OA compared with OP. Setting the power at 0.90, the significance level at 0.05 with two-tailed testing, the maximum calculated sample size was 9. Thus we set a sample size of 10 patients in each group. To avoid the confounding effect of age and sex, postmenopausal women with a history of menopause for more than 5 years were included. The inclusion and exclusion criteria in this study followed as published previously [7, 23, 29]. Briefly, OA patients with grade III-IV disease according to the Outerbridge classification [48] and OP patients with femoral neck fracture were included. Patients with other osteoarticular diseases like congenital or acquired dysplasia and rheumatoid arthritis were excluded. And patients with diseases and therapy that influence bone metabolism were excluded as well. Informed consent was obtained from all individual participants included in the study. This study was approved by the Institutional Review Board of Shanghai Renji Hospital.

The specimen preparations were carried out as described previously [7, 23, 29]. Briefly, the femoral head was obtained from each included patient underwent total hip arthroplasty, and two bigger samples (15×15mm in cross-section and 10mm long) of subchondral cancellous bone were obtained from the load-bearing area (Figure 1A) for micro-computed tomography (micro-CT) imaging and microdamage evaluation. And one smaller sample (5×5mm in cross-section and 10mm long) was obtained for immunohistochemical analysis.

Animal groups

The animal experiments were ethically approved by the Institutional Animal Care and Use Committee of Shanghai Jiaotong University and was conducted by the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. Eighteen 3-month-old female virgin SD rats were randomly divided into 3 groups (n=6 per group): the SHAM group was a control group containing sham-operation rats; the rats in OVX group were ovariectomized and treated with 0.9% saline; the OVX+BIBO3304 group was a treatment group in which OVX rats were treated with Y1R antagonist BIBO3304 (Tocris Bioscience, Bristol, UK) (1 mg/kg/day, i.p.). For OVX rats, bilateral ovaries were excised, and for SHAM rats, adipose tissue of the equal weight as the ovaries was removed. After 8 weeks following surgery, drugs were administered to rats daily for 4 weeks.

In vivo fatigue loading

In vivo fatigue loading was performed to induce bone microdamage in all animals. In 12 weeks after operation, the left tibias of all rats were subjected to fatigue loading as previously described [49]. In brief, left knees and ankle joints were placed in concaved cups of the mechanical testing machine (Instron 8874; Instron Limited). Dynamic compressive loads (peak loads of 70 N, 4 Hz, sinusoidal wave) were applied axially to the tibias. The load was applied 1,4400 times per day for 2 d. All rats were euthanized 3d after fatigue loading. The left tibias were fixed in 70% ethanol for microdamage evaluation. The right tibias were collected for micro-CT imaging. The right femurs were fixed in 4% paraformaldehyde for immunohistochemical analysis. Besides, the bone marrow of left femurs was collected. After erythrolysis by Red Blood Cell Lysis Buffer (Sigma Aldrich), the total protein was extracted using RIPA lysis buffer (Beyotime, Shanghai, China). The protein of bone marrow was frozen at −80°C for western blot.

Bone micro-computed tomography (micro-CT) imaging

The clinical bone specimens and right tibia of rats were examined using a micro-CT system (SkyScan-1176, Bruker microCT, Belgium) at a voxel size of 18 μm. After scanning, a region of interest (ROI) centered over the specimen was selected for analysis. The three-dimensional (3-D) reconstruction of trabecular bone was performed according to the ROI. The quantitative analysis was carried out and the following parameters were calculated by the software provided with the instrument: Bone mineral density (BMD, g/cm³), bone volume fraction (BV/TV; %), trabecular number (Tb.N; mm⁻¹), trabecular thickness (Tb.Th; μm), trabecular separation (Tb.Sp; μm), connectivity density (Conn.D; mm⁻³) and structure Model Index (SMI).

Microdamage evaluation

The microdamage of clinical bone specimens and left tibia of rats was evaluated according to the method reported previously [50]. Briefly, the specimens were fixed in 70% ethanol for more than 48h and bulk stained in 1% basic fuchsin in a graded series of ethanol under vacuum. And then the specimens were embedded in polymethylmethacrylate. The middle segment of tibia specimens was longitudinally sectioned into 60 μm-thick slices. Slides were observed using green incident light (λ=594 nm) and microdamage identified by red fluorescence. And under ultraviolet incident light (λ=365 nm), microdamage identified by purple fluorescence. The crack length (Cr.Le; μm) was initially examined, and then crack numerical density (Cr.Dn; #/mm²) and crack surface density (Cr.S.Dn; μm/mm²) were calculated to evaluate the microdamage of the specimens.

Immunohistochemistry

The clinical bone specimens and right distal femurs of rats were embedded with paraffin after decalcification and were sectioned into 5-mm thickness. The immunohistochemistry was assessed according to the method reported previously [29]. Briefly, the sections were incubated with the primary antibody at 4°C overnight, and then with a biotinylated secondary antibody for 15 min at 37°C. The primary antibodies for the clinical bone specimens included rabbit anti-NPY (1:200, Abcam, Cambridge, USA), rabbit anti-NPY1R (1:200, Abcam, Cambridge, USA) or rabbit anti-NPY2R (1:200, Abcam, Cambridge, USA) antibody. And the sections of rat femurs were stained with rabbit anti-Runx2 (1:200, Affinity Biosciences, Inc. USA) or rabbit anti-matrix metalloproteinase-9 (MMP9, 1:200, Affinity Biosciences, Inc. USA). Peroxidase reaction was visualized using a solution of diaminobenzidine. The images were obtained with a Spot Advanced digital-imaging system (Diagnostic Instruments, Inc., Sterling Heights, MI, USA), and the mean optical density (MOD) was measured using the software package Image Pro Plus (version 5.0.1, Media Cybernetics, Silver Spring, MD, USA).

Histological examination

The sections of distal rat femurs were stained with hematoxylin and eosin (HE, Beyotime Biotech, Shanghai, China), tartrate-resistant acid phosphatase (TRAP, Sigma-Aldrich, St. Louis, MO, USA) and Alizarin Red (Sigma-Aldrich) staining. The staining was performed according to manufacturer's protocol and finally observed using a digital microscope (Olympus, Japan).

Isolation, culture, identification and differentiation of rat BMSCs

Primary BMSCs were harvested from 4-week-old SD rats as described previously [51]. Briefly, bone marrow was flushed out from the femurs with DMEM/F12 (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, Grand Island, NY, USA), 100 U/ml penicillin, and 100 μg/ml streptomycin (Gibco, Grand Island, NY, USA). After being centrifuged at 800 rpm for 5 min, the cells were cultured in 25 cm² culture flasks (Corning, NY, USA) at 37°C with 5% CO2. The medium was changed every 2 days, and the non-adherent cells were removed. When the confluence reached approximately 80%, the adherent cells were harvested by incubation with 0.25% trypsin-EDTA (Gibco, Grand Island, NY, USA) and were passaged. BMSCs at passage 3 were identified by flow cytometry to detect the expression of the surface markers CD29, CD90, CD44, CD34, CD45 and CD11b. To confirm the multilineage differentiation potential, the BMSCs were cultured under adipogenic, chondrogenic, and osteogenic differentiation conditions, and stained with oil Red O (Sigma-Aldrich, USA), Alcian blue (Sigma-Aldrich, USA), and Alizarin Red (Sigma-Aldrich, USA), respectively. The multilineage differentiation examination were performed described previously [51]. BMSCs at passage 3-5 were used for experiments.

Cell treatment and groups

BMSCs were seeded in 6-well plates (Corning, NY, USA) at 1 × 10⁵ cells/well. we divided the cells into 4 groups to investigate the role of NPY in the cAMP/PKA signaling pathway and the Y1R relationship. The CON group was treated a volume of phosphate-buffered saline (PBS; Gibco, Grand Island, NY, USA) equal to that used for treatments; The NPY group was treated with 10^-10 M NPY (Abcam, Cambridge, USA); The NPY+BIBO3304 group was supplemented with 10^-10 M NPY and 10^-6 M Y1R antagonist (BIBO3304, Tocris Bioscience, UK). The NPY+ cAMP group was treated with a mixture of NPY and 10^-4 M 8-bromoadenosine-3',5'-cyclic monophosphate (8-Bromo-cAMP, Selleck Chemicals, USA). 8-Bromo-cAMP is a small molecule cAMP analogue, which is an activator of PKA. For osteogenic induction, the medium was changed to the osteogenic medium containing with 10^-7 M dexamethasone (Sigma-Aldrich, USA), 10^-2 M sodium β-glycerophosphate (Sigma-Aldrich, USA) and 50 μg/mL L-ascorbic acids (Sigma-Aldrich, USA).

ALP activity assay

BMSCs were cultured with osteogenic medium for 3d. ALP activity was quantified in cell lysates using an ALP assay kit (Beyotime Biotech, Shanghai, China) according to the manufacturer’s instructions.

Enzyme linked immunosorbent assay (ELISA)

The cAMP concentrations in the media were measured. BMSCs were cultured with osteogenic medium for 3d. The culture media were collected and centrifuged at 1000rpm for 10min, and the supernatants were stored at -80°C until use. The cAMP concentration was detected using Rat cAMP ELISA kit (Jianglai biotech, Shanghai, China), according to the manufacturer’s instructions. The limit of detection was 0.5-20 nmol/L, and the coefficient of variation (CV) was less than 9%. The absorbance was read at 450 nm using a microplate reader. The concentrations were calculated according to the optical density (OD) and standard curve.

Alizarin Red S and ALP staining

After osteogenic induction for 7d, cells in 6-well plates were stained by Leukocyte Alkaline Phosphatase Kits (Sigma-Aldrich, USA) for ALP staining. And Alizarin Red S staining was performed to detect the mineralized nodule formation after BMSCs were incubated with osteogenic medium for 21d. All the staining was performed according to the manufacturer’s instructions.

Western blot

BMSCs were cultured with osteogenic medium for 3d, 7d and 21d. Total protein of BMSCs or rat bone marrow was extracted using RIPA lysis buffer (Beyotime, Shanghai, China). The protein concentration was measured using the bicinchoninic acid protein assay kit (Thermo Scientific, Rockford, IL, USA). Proteins were separated using SDS-PAGE gel and then transferred to 0.22 μm polyvinylidine diluoride membrane (Millipore, USA). Membranes were incubated at 4°C overnight with a 1:1000 dilution primary antibodies for RANKL (1:1000; Santa Cruz, Biotechnology, Inc. USA), OPG (1:1000; Affinity Biosciences, Inc. USA), Osteocalcin (OCN, 1:1000; Affinity Biosciences, Inc. USA), RUNX2 (1:1000; Affinity Biosciences, Inc. USA), MMP9 (1:1000; Affinity Biosciences, Inc. USA), Cathepsin K (Cath-K, 1:1000; Affinity Biosciences, Inc. USA), GAPDH (1:1000; Proteintech, Chicago, IL, United States) or β-tubulin (1:1000; Proteintech, Chicago, IL, United States), and then incubated for 1 h at room temperature by secondary antibody. The immunoblots were visualized using enhanced chemiluminescence (Thermo Scientific) and images were obtained using ChemiDoc XRS (Bio-Rad, USA).

Statistical analysis

Data were analyzed using GraphPad Prism software (GraphPad Prism 8.00, San Diego, USA). All data were presented as the means ± standard deviation (SD). The difference between two groups was evaluated by the Student’s t-test, and One-way analysis of variance was used for the statistical comparisons between multiple groups. Pearson correlation analysis was performed to evaluate the correlation between variables. The significance level was set at P < 0.05.