Tumour necrosis factor α regulates the miR‐27a‐3p–Sfrp1 axis in a mouse model of osteoporosis

Abstract Osteoporosis is a metabolic bone disease that involves gradual loss of bone density and mass, thus resulting in increased fragility and risk of fracture. Inflammatory cytokines, such as tumour necrosis factor α (TNF‐α), inhibit osteogenic differentiation of bone marrow mesenchymal stem cells (BMSCs), and several microRNAs are implicated in osteoporosis development. This study aimed to explore the correlation between TNF‐α treatment and miR‐27a‐3p expression in BMSC osteogenesis and further understand their roles in osteoporosis. An osteoporosis animal model was established using ovariectomized (OVX) mice. Compared with Sham mice, the OVX mice had a significantly elevated level of serum TNF‐α and decreased level of bone miR‐27a‐3p, and in vitro TNF‐α treatment inhibited miR‐27a‐3p expression in BMSCs. In addition, miR‐27a‐3p promoted osteogenic differentiation of mouse BMSCs in vitro, as evidenced by alkaline phosphatase staining and Alizarin Red‐S staining, as well as enhanced expression of the osteogenic markers Runx2 and Osterix. Subsequent bioinformatics analysis combined with experimental validation identified secreted frizzled‐related protein 1 (Sfrp1) as a downstream target of miR‐27a‐3p. Sfrp1 overexpression significantly inhibited the osteogenic differentiation of BMSCs in vitro and additional TNF‐α treatment augmented this inhibition. Moreover, Sfrp1 overexpression abrogated the promotive effect of miR‐27a‐3p on the osteogenic differentiation of BMSCs. Furthermore, the miR‐27a‐3p–Sfrp1 axis was found to exert its regulatory function in BMSC osteogenic differentiation via regulating Wnt3a–β‐catenin signalling. In summary, this study revealed that TNF‐α regulated a novel miR‐27a‐3p–Sfrp1 axis in osteogenic differentiation of BMSCs. The data provide new insights into the development of novel therapeutic strategies for osteoporosis.


INTRODUCTION
Osteoporosis is a progressive metabolic bone disease involving the gradual loss of bone density and mass, resulting in increased fragility and risk of fracture (Akkawi & Zmerly, 2018;Lorentzon, 2019).Even though current therapeutic strategies can reduce the risk of developing osteoporotic fractures, osteoporosis often develops without diagnosis until a fracture occurs (Mauck & Clarke, 2006).Accumulating studies have shown that high levels of inflammatory cytokines could increase the activity of osteoclasts and impair bone formation in osteoporosis (Khosla et al., 2008;Lacativa & Farias, 2010).Specifically, the inflammatory cytokine tumour necrosis factor α (TNF-α) has been demonstrated to play a critical role in osteogenesis via the modulation of bone loss and bone formation (Yang et al., 2013;Zhao et al., 2011).
In accord with these findings, several studies recently suggested that inhibition of TNF-α-mediated impairments in osteogenesis, such as using resveratrol, quercetin and artemisinin, can help to promote the osteogenesis of mesenchymal stem cells thereby potentially improved osteoporosis (Hu et al., 2021;Yuan et al., 2018Yuan et al., , 2020)).However, the underlying mechanism of impaired osteogenic differentiation by inflammatory cytokines remains unclear.
MicroRNAs (miRNAs) are a group of short non-coding RNAs which participate in various biological processes via post-transcriptional regulation of target genes (O'Brien et al., 2018).An increasing number of studies have suggested that miRNAs play important regulatory roles in bone marrow mesenchymal stem cell (BMSC) proliferation, differentiation and function (Lei et al., 2011;Shao, 2017;Teng et al., 2016;Zhang et al., 2020).MicroRNA profiling of the entire trabecular bone has identified multiple dysregulated miRNAs associated with osteoporosis (De-Ugarte et al., 2015).For example, miR-133a was reported to be upregulated during osteoclastogenesis and promoted osteoclast differentiation of THP-1 and RAW264.7 cells (Li et al., 2018).In another study, miR-7b-5p was found to inhibit the adipogenic differentiation of human BMSCs via insulin receptor substrate 2 (IRS2) targeting, thus alleviating the development of osteoporosis (Li et al., 2019b).In contrast, Cheng et al. reported that miR-365-5p could promote the development and progression of osteoporosis via the negative regulation of Runx2 and inhibition of osteogenic differentiation (Cheng et al., 2019).Thus, further dissection of the functional roles of miRNAs in osteoporosis is an important endeavour.
Among the miRNAs involved in osteogenesis, miR-27a-3p was found to be associated negatively with the osteogenic differentiation of MC3T3-E1 pre-osteoblasts by targeting Osterix (Xu et al., 2020).
However, it is unclear whether miR-27a-3p functions as a downstream regulator of inflammatory signals and participates in the TNFα-mediated osteoporosis.Therefore, we focused on the correlation between TNF-α stimulation and miR-27a-3p expression to clarify the role of miR-27a-3p in the development of osteoporosis.

Highlights
• What is the central question of this study?
What is the mechanism by which TNF-α plays its critical role in osteogenesis via the modulation of bone loss and formation?
• What is the main finding and its importance?
Secreted frizzled-related protein 1 (Sfrp1) is a member of the Sfrp family regulating the Wnt signalling pathway (Bovolenta et al., 2008).Downregulation of Sfrp1 has been reported in colorectal cancer, and low expression levels of Sfrp1 have been associated with poor prognosis in colorectal cancer patients (Huang et al., 2014).In addition, Sfrp1 has been reported to negatively regulate BMSC proliferation and differentiation (Hu et al., 2019;Tang et al., 2019).However, the exact mechanisms underlying the regulation of Sfrp1 expression and the role that Sfrp1 plays in regulating osteogenic differentiation of BMSCs remain unknown.
In this study, we used the ovariectomized (OVX) mouse model and demonstrated that elevated levels of TNF-α suppressed the expression of miR-27a-3p during osteogenic differentiation of BMSCs.MiR-27a-3p promoted osteogenic differentiation and exerted its function

OVX model procedures
Wild-type C57BL/6 female mice were purchased from Vital River Laboratory (Beijing, China) and were randomly divided into two groups: the Sham group (n = 5) and OVX group (n = 5).The OVX group underwent bilateral ovariectomy by the dorsal approach.After adaptive housing for 1 week, the OVX modelling was performed as previously reported (Li et al., 2019a).Briefly, anaesthesia was performed using 1.5% isoflurane at a flow rate in the range of 0.5-1.0L/min.Then, hair in the dorsal mid-lumbar area was shaved and the skin was cleaned to make a 20-mm midline dorsal skin incision.Following this, the paired ovaries were excised and the wound was closed using suturing.Mice in the Sham group received the same procedure except ovary excision.
The post-operative analgesia was performed by adding 1.7 μg/mL meloxicam into the drinking water.Since the entire operation and animal maintenance was performed in a pathogen-free animal facility, no antibiotic was used in this study.Eight weeks later, mice were killed with an overdose of isoflurane (5% isoflurane at a rate of 1.0 L/min for 20 min).The killing was confirmed with cardiac and respiratory arrest.
Subsequently, the serum and leg samples were harvested and stored at -80 • C until utilized.

BMSC isolation, characterization and treatment
Mouse BMSCs were obtained from long bones of the indicated C57BL/6J mice.Mice were killed, and bone marrow cells were collected by flushing nucleate cells out of the femurs and tibiae with cold phosphate buffered saline (PBS).Cells were cultured (2 × 10 6 nucleated cells/10 cm dish) in α-Minimum Essential Medium (Thermo Fisher Scientific) supplemented with 10% FBS (Thermo Fisher Scientific), 1% penicillin-streptomycin in an incubator at 37 • C with 5% CO 2 .Three days later, non-adherent cells were removed.The adhering cells were cultured in intact medium and changed every 2 days.
After one passage, the primary BMSCs were cryopreserved in liquid nitrogen using a cryopreservation/recovery kit (AccuRef Scientific, Xi'an, China).When recovered, the BMSCs were passaged at least once and phenotypically confirmed by flow cytometry before an experiment.
Mouse IgG isotype was used as the control (cat.no.565832; BD Biosciences).A BMSC purity >97% was considered appropriate, and the purity-confirmed BMSCs were used in the following investigations.
To determine the effect of cytokines on the expression of miR-27a-3p, BMSCs from wild-type mice were treated with the cytokines interleukin (IL)-1β (1 ng/mL), interferon γ (IFN-γ; 50 ng/mL) or TNFα (10 ng/mL) for 2 weeks, which was followed by determination of miR-27a-3p expression using RT-quantitative PCR (RT-qPCR).During treatment, the culture medium was replaced every 2 days with fresh medium supplemented with the same concentrations of cytokines.No obvious toxic effects induced by these cytokines were identified.

Western blot
Total protein was purified from BMSCs derived from the Sham or OVX mice, or cultured BMSCs using radioimmunoprecipitation assay (RIPA) buffer (AccuRef Scientific), and the protein concentration was determined using a BCA kit (Exprecision, Xi'an, China) according to the manufacturer's protocol.An equal amount of protein was separated using SDS-PAGE and transferred onto a nitrocellulose membrane (Millipore, Billerica, MA, USA) for western blot analysis.
The membranes were probed with the following primary antibodies: and developed with an ECL kit (AccuRef Scientific).

Enzyme-linked immunosorbent assay
Mouse serum samples from the Sham or OVX group were analysed by Mouse TNF-α and IL-1β enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems, Minneapolis, MN, USA) according to the manufacturer's instructions.

Haematoxylin and eosin staining
Mouse bone samples from the Sham or OVX group were fixed with 4% paraformaldehyde for 2 days and decalcified with 10% EDTA at room temperature for 4 weeks.Samples were then dehydrated with gradient ethanol, degreased in xylene, embedded in paraffin, and cut into the sequential 5 μm-thick slices.The slices were then stained with haematoxylin and eosin (H&E) according to the manufacturer's instruction (Beyotime, China) to confirm the occurrence of osteoporosis in mice.

Luciferase reporter assay
Luciferase reporter vector containing wild type or a mutated version of 3ʹ-UTR of Sfrp1 was constructed using pGL3-Luc vector (Promega, Madison, WI, USA).HEK293T cells were transfected with the indicated reporter vector, together with miR-27a-3p mimics or negative control.
Relative luciferase activity was analysed with a Dual-Luciferase reporter assay kit (Promega) 48 h later.

Alkaline phosphatase staining and Alizarin Red-S staining
BMSCs from wild type mice were isolated and cultured in vitro for 14 days.Alkaline phosphatase (ALP) staining was performed to detect the differentiation of BMSCs using an alkaline phosphatase live stain kit (cat.no.ab242286, Abcam, Waltham, MA, USA) following the manufacturer's protocol.Differentiated BMSCs and control cells were also stained with aqueous 0.5% (v/v) Alizarin Red-S (Sigma-Aldrich, St Louis, MO, USA).The mineralized nodule formation was analysed and recorded under an inverted microscope (Olympus, Tokyo, Japan).
The next day, the RIP mixture was washed with RIP Wash Buffer, and proteinase K was used to digest the antibody.The pulled-down RNA was purified by TRIzol for subsequent analysis.

Statistical analysis
All data obtained from the in vitro experiments (n = 3) or the in vivo experiments (n = 5) were presented as mean ± standard deviation.
The statistical significance was analysed by two-tailed Student's ttest or one-way analysis of variance (ANOVA) with post-hoc Tukey's multiple comparisons test wherever necessary.GraphPad Prism v. 8.0 (GraphPad Software, Inc., San Diego, CA, USA) was used for statistical analysis.A P < 0.05 was considered statistically significant.

TNF-α suppresses miR-27a-3p expression in OVX mice
Previous studies have demonstrated that TNF-α plays a critical role in osteogenesis (Yang et al., 2013).To confirm the function and explore the underlying mechanism of TNF-α in osteogenesis, we examined the expression level of TNF-α in the OVX animal model.Wild type mice were subjected to either a sham operation or a bilateral ovariectomy to establish the OVX mouse model.Osteoporosis in OVX mice was confirmed by H&E (Figure 1a), Masson (Figure 1b) and Goldner staining (Figure 1c).The TNF-α and IL-1β levels in OVX mouse serum were significantly higher than those in the Sham group (Figure 1d; P = 0.0106 and 0.0147 for TNF-α and IL-1β, respectively; n = 5 for each group).Significantly increased expression of TNF-α mRNA was also confirmed in the bone tissues of the OVX mice (Figure 1e; P = 0.003; n = 5 for each group).Although an obvious upregulation of IL-1β mRNA was observed in the OVX group, no significant difference was identified between the OVX and Sham group (Figure 1e; P = 0.07; n = 5 for each group).To understand the functional role of TNF-α, we performed nextgeneration sequencing using the total RNA from BMSCs isolated from the bone marrow of the Sham and OVX mice and found that multiple miRNAs were markedly dysregulated in expression between the Sham and OVX group (data not shown).MiR-27a-3p, which regulates cell proliferation in various tumours (Chen et al., 2021;Su et al., 2019;Yang et al., 2021), was found significantly downregulated in OVX mice.

MiR-27a-3p promotes osteogenic differentiation of BMSCs in vitro
To further characterize the function of miR-27a-3p in osteogenesis, BMSCs were isolated from wild type mice and cultured in vitro for osteogenic differentiation.The expression of miR-27a-3p was analysed at different time points (Figure 2a).We found that miR-27a-3p expression reached the peak level at around day 7 and decreased gradually, thus indicating a promotive role of miR-27a-3p in osteogenic differentiation (Figure 2a).
Following this observation, we transfected BMSCs with miR-27a-3p mimics and treated them with TNF-α for further analyses.

MiR-27a-3p negatively regulates Sfrp1 expression by binding to the 3ʹ-UTR of its mRNA
To further reveal the molecular mechanisms under which TNF-α regulates miR-27a-3p expression in BMSCs, we next predicted the downstream targets of miR-27a-3p with online tools.In total, 45 genes were predicted by microT and 63 genes were predicted by miRDB, with five common targets predicted by both online tools (Figure 3a, Table S1).Thus, five genes (Pkia, Gfpt2, Gmps, Car10 and Sfrp1) were considered as the potential targets of miR-27a-3p, and the relative expression of these genes in BMSCs transfected with miR-NC or miR-27a-3p mimics was examined by qPCR.As shown in Figure 3b (miR-NC vs. miR-27a-3p mimic, P = 0.14 for Pkia; P = 0.21 for Gfpt; P = 0.1611 for Gmps; P = 0.2623 for Car10; P = 0.04 for Sfrp1; n = 3 for each group), Sfrp1 was the only gene whose mRNA expression was significantly inhibited by miR-27a-3p mimics.Further analysis indicated that miR-27a-3p had complementary binding sequences that targeted the 3ʹ-UTR of wild type Sfrp1 mRNA (Figure 3c).The luciferase reporter assay confirmed that miR-27a-3p mimics suppressed the relative luciferase activity in HEK293T cells and BMSCs transfected with luciferase reporter containing the wild type (WT) 3ʹ-UTR of Sfrp1 mRNA, but not with the mutated 3ʹ-UTR of Sfrp1 mRNA (Figure 3d; miR-NC vs. miR-27a-3p mimic, in 293T cells, P = 0.032 for Sfrp1 WT; P = 0.5612 for Sfrp1 MUT; in BMSCs, P = 0.002 for Sfrp1 WT; P = 0.9943 for Sfrp1 MUT; n = 3 for each group).In addition, the interaction between Sfrp1 and miR-27a-3p was validated by the RIP assay in BMSCs.Anti-AGO2 immunoprecipitation was observed to markedly enhance the enrichment of Sfrp1 and miR-27a-3p, compared to the immunoprecipitation with the IgG control, suggesting that Sfrp1 could directly interact with miR-27a-3p (Figure 3e; IgG vs. AGO2, P = 0.004 for Sfrp1; P = 0.0022 for miR-27a-3p mimic; n = 3 for each group).Additionally, BMSCs from OVX mice had significantly more Sfrp1 expression at both mRNA and protein levels than that from the mice in the Sham group (Figure 3f; Sham vs. OVX, P = 0.0256 for Sfrp1 mRNA; P = 0.0287 for Sfrp1 protein; n = 3 for each group).

DISCUSSION
The imbalance between osteoblastic bone formation and resorption leads to osteoporosis, which is a common disease among elderly people (Boyle et al., 2003;Teitelbaum, 2000).This study demonstrated that TNF-α regulates the miR-27a-3p-Sfrp1 axis in osteoporosis.Overexpression of miR-27a-3p or knockdown of Sfrp1 was observed to promote osteogenic differentiation of BMSCs in vitro, thus providing a potential therapeutic strategy targeting miR-27a-3p/Sfrp1 for regenerating the bone mass.
Inflammatory cytokines such as TNF-α, IL-1β and IL-6 are elevated in postmenopausal osteoporosis and have been shown to play a role in the development of the disease (Zheng et al., 1997).This study employed the OVX mouse model and confirmed that TNFα expression was elevated in OVX mice compared to the Sham group.Multiple mechanisms have illustrated how TNF-α suppresses osteogenic differentiation, such as enhancing purinergic P2Y2 receptor expression, inducing semaphorin3D expression, or inhibiting Foxo1 via miR-705 upregulation (Du et al., 2018;Liao et al., 2016;Sang et al., 2017).The expression levels of various miRNAs were dysregulated in BMSCs from the OVX mice based on the results of microarray analysis for miRNA profiling (data not shown).Particularly, we focused our study on miR-27a-3p.Of note, the OVX mice also had a higher serum level of IL-1β than the Sham mice, but the treatment with TNF-α but not IL-1β significantly downregulated miR-27a-3p level in BMSCs.Therefore, we further focused on the relationship between TNF-α treatment and miR-27a-3p expression during osteogenic differentiation of BMSCs in vitro.
Studies have demonstrated that miR-27 is involved and plays critical roles in adipogenic and osteogenic differentiation (Lin et al., 2009;Wang & Xu, 2010).Given the fact that osteoporosis develops from the imbalance between adipogenic and osteogenic differentiation, we set out to investigate the expression pattern and functional role of miR-27a-3p in osteoporosis.Overexpression of miR-27a-3p was observed to significantly enhance osteoblast differentiation with upregulated ALP expression and increased formation of mineralized nodules, while inhibition of miR-27a-3p had the opposite effect.We also found that overexpression of miR-27a-3p enhanced consistently the expressions of osteoblast-specific markers Runx2 and Osterix.
Furthermore, miR-27a-3p expression peaked at day 7 during the 28-day osteogenic differentiation of BMSCs in vitro, implicating the potential indispensable role of miR-27a-3p in osteoblastic bone formation.Liu et al. showed that overexpression of lncRNA MEG3 promoted osteogenic differentiation of periodontal ligament stem cells by enhancing expression of IGF1 and suppressing expression of miR-27a-3p (Liu et al., 2019b), which indicates a negative effect of miR-27a-3p on osteogenesis.Our findings are consistent with a previous report by Xu et al., showing that exosomes from C2C12 myoblasts can promote pre-osteoblast MC3T3-E1 differentiation to osteoblasts, and the effects of exosomes depended on the miR-27a-3p component, as C2C12 exosomes failed to exert these effects when miR-27a-3p was absent (Xu et al., 2018).These results uncover a novel regulatory pathway involving miR-27a-3p-Sfrp1 modulating osteogenesis in the BMSC, opening new perspectives into the pathophysiology and treatment of osteoporosis.
Bioinformatics analysis combined with experimental validation was conducted to understand the functional mechanism of miR-27a-3p in osteogenic differentiation and revealed that Sfrp1 was the direct target of miR-27a-3p.Sfrp1 was identified as being associated with bone strength and body composition (Boudin et al., 2012).Interestingly, Sfrp1 has been demonstrated to regulate cell proliferation in multiple tumours such as myeloid leukaemia, ovarian cancer and colorectal cancer (Hu et al., 2019;Huang et al., 2014;Liu et al., 2019a;Pehlivan et al., 2017).Consistent with previous findings reported by Tang et al. (2019), we demonstrated that downregulation of Sfrp1 promoted the proliferation and differentiation of BMSCs.
However, other studies indicate that the expression of Sfrp1 can be regulated by other miRNAs such as miR-1180 and miR-144 (Hu et al., 2019;Tang et al., 2019).Similarly, Wu et al. reported that miR-27a targeted DKK2 and Sfrp1 to promote re-osseointegration in the regenerative treatment of peri-implantitis (Wu et al., 2019).Thus, further investigations on the miRNAs involved in the inflammatory responses and the regulation of Sfrp1 expression during osteogenic differentiation of BMSCs are warranted to fully elucidate the roles of Sfp1 in osteoporosis development.
To illustrate the complicated regulation of Sfrp1 by different miRNAs in osteoporosis, we focused on Wnt-β-catenin signalling.Wnt signalling is reported to play crucial roles in osteogenic differentiation, and disrupted Wnt signalling in osteoblastic-lineage cells can result in bone formation deficiency in osteoporosis (Jing et al., 2018;Tu et al., 2012).Multiples studies have reported that the Wnt signalling suppresses osteogenic differentiation of mesenchymal stem cells (Boland et al., 2004;De Boer et al., 2004;Yuan et al., 2016).In agreement with the previous finding that Sfrp1 is an antagonist of Wnt signalling (Kawano & Kypta, 2003), we showed that the miR-27a-3p-Sfrp1 axis regulated the Wnt3a-β-catenin signalling pathway in osteogenic differentiation of mouse BMSCs.However, since miR-27-3p might have multiple downstream targets, additional effort is needed to explore other possible signalling pathways that are regulated by the miR-27a-3p.
In conclusion, we demonstrated that miR-27a-3p was suppressed by TNF-α in osteogenic differentiation of BMSCs and miR-27a- through the negative regulation of Sfrp1 and enhancement of βcatenin-C-Myc-Cyclin D1 signalling.These results suggest a novel regulatory axis of miR-27a-3p and Sfrp1 in modulating the osteogenic differentiation of BMSCs, which provides critical insights into the clinical diagnosis and treatment of osteoporosis.

F
I G U R E 1 TNF-α suppresses miR-27a-3p expression in ovariectomized (OVX) mice.(a-e) Wild-type (WT) C57BL/6 mice underwent bilateral ovariectomy surgeries to establish the OVX mouse model, and mice that underwent sham operation were used as controls.(a-c) Representative haematoxylin and eosin (H&E) staining (a), Masson staining (b) and Goldner staining (c) of bone tissues from the Sham or OVX group to confirm osteoporosis (arrow indicates osteoporosis).Scale bar, 100 μm.(d) The tumour necrosis factor α (TNF-α) and interleukin (IL)-1β levels in the serum were analysed by ELISA at 60 days after modelling.(e) The relative mRNA expression levels of TNF-α and IL-1β in bone tissue were analysed by qPCR.(f, g) Representative flow cytometrical histogram plots show that bone marrow mesenchymal cells (BMSCs) isolated from mice bones were positive for BMSC markers CD73, CD105 and CD90.1 (f), and were negative for haematopoietic markers CD34 and CD45 (g).(h) The relative expression of miR-27a-3p in BMSCs from the Sham or OVX group was analysed by qPCR.(i) BMSCs from WT mice were treated with IL-1β(1 ng/mL), interferon γ (IFN-γ; 50 ng/mL), or TNF-α (10 ng/mL) for 2 weeks, and the relative expression of miR-27a-3p was analysed by qPCR.*P < 0.05, **P < 0.01, between the indicated groups.

F
I G U R E 2 MiR-27a-3p promotes osteogenic differentiation of BMSCs in vitro.(a-g) BMSCs were isolated from WT mice and cultured in vitro.(a) The expression of miR-27a-3p was analysed by qPCR at the indicated time points.(b) BMSCs were transfected with miR-NC, miR-27a-3p mimics followed by TNF-α treatment (10 ng/mL).The expression of miR-27a-3p was analysed by qPCR 48 h later.(c) Transfected BMSCs were cultured in vitro for 7 days, and osteogenic differentiation of BMSCs was subsequently analysed by Alizarin Red staining.Scale bar, 100 μm.(d) Quantification result for Alizarin Red staining.(e) The expression of ALP was analysed by ALP staining.(f) Transfected BMSCs were cultured in vitro for 7 days and the mRNA expression of Runx2 and Osterix was analysed by qPCR.(g) Transfected BMSCs were cultured in vitro for 7 days and the protein expression of Runx2 and Osterix was analysed by western blot.*P < 0.05, **P < 0.01, ***P < 0.001, between the indicated groups.
3p promoted BMSC osteogenesis via negative regulation of Sfrp1 and enhancement of the Wnt3a-β-catenin signalling cascade.Overexpression of Sfrp1 abrogated the promotive effect of miR-27a-3p in osteogenic differentiation of BMSCs.These results suggest a novel regulatory axis of miR-27a-3p-Sfrp1 in modulating BMSC osteogenesis, which sheds new light in the clinical diagnosis and treatment of osteoporosis.