Transcriptional Profiling of Muscle in Females With Distal Radius Fracture and Functional Sarcopenia

Abstract

Skeletal muscle and bone interact with each other in mechanical and biochemical ways. This study aimed to investigate the molecular mechanisms of interaction between muscle and bone by analyzing the transcriptional profiles of total RNA from the muscle tissue of females with distal radius fractures. A total of 30 female participants (mean age 71.1 ± 8.9 years) with distal radius fractures were recruited. Participants were categorized into 2 groups: the NORM group consisted of participants with T score of the areal bone mineral density (aBMD) of the femoral neck higher than –1.0, handgrip strength greater than 18 kg, and gait speed faster than 1.0 m/s (n = 10). Otherwise, participants with T score of the aBMD of the femoral neck equal to or less than –1.0, handgrip strength lower than 18 kg, and gait speed slower than 1.0 m/s (n = 20) were categorized into EXP group. Pronator quadratus muscle samples were obtained from all participants. Total RNA was extracted from frozen muscle samples and sequenced. The gene ontology analysis demonstrated that the potential interactions between attached muscle function and the density of the associated bone would be linked with collagen biosynthetic activity and maintenance of extracellular matrix structures. The analysis of the pathway, network, and protein class exhibited that integrin signaling, inflammatory reactions, matrix metalloproteinase (MMP) activity, and extracellular matrix protein structure had possible associations with the molecular background of muscle–bone interaction. Through integrin signaling, MMP activity, inflammatory reactions, and collagen biosynthesis, muscle and bone may mutually interact with one another.

Aging accompanies immunological, neuromuscular, and endocrinological body changes and inevitably has influences on biological and functional alterations in skeletal muscle and bone (1,2). Several studies have focused on the shared risk factors and pathophysiological pathways of osteoporosis and sarcopenia, the well-known geriatric diseases, and pathophysiological pathways of osteoporosis and sarcopenia, the well-known geriatric diseases, and the new concept “osteosarcopenia” has been proposed to define simultaneous muscle weakness and reduced bone quality and the coexistence of sarcopenia and osteoporosis (3,4). The current therapeutic approaches to reduce the incidence of osteoporotic fractures and falls, which may cause frailty and mortality, have placed a strong emphasis on the reciprocal relationships between bone and muscle as well as increasing bone mineral contents and strengthening muscular force and function.

To adapt to environmental requirements and maintain metabolic homeostasis, skeletal muscle and bone interact with each other constantly (1–8). These are the 2 basic components that primarily comprise the musculoskeletal system establishing a mechanical pulley–lever system and having chemical interactions. To create the motion needed to produce torque, skeletal muscle is attached to the associated bone close to the axis of motion (9). Myokines, osteokines, adiponectin, senescence-associated secretory phenotype, micro-RNA, and other elements mediate the biochemical interactions between bone and muscles (1–8). Although the specific interactions connecting them are still unclear, it appears that the interaction of mechanical loading and biochemical signaling between them produces more potent influences than either stimulus alone.

The RNA sequencing (RNA-seq) technology has been utilized as a formidable tool to identify potential molecular backgrounds or therapeutic targets of certain diseases or conditions (10). Few studies exist to discover the muscle–bone interaction using RNA-seq technology. Hence, the aim of the current study was to elucidate the molecular mechanisms of interaction between muscle and bone, and by analyzing the transcriptional profiles of total RNA from muscle tissue of females with distal radius fractures.

Materials and Methods

Participants

In total, 30 female participants (mean age 71.1 ± 8.9 years, age range 55–93 years) with distal radius fractures requiring surgical operation were enlisted from a tertiary care hospital. All participants were postmenopausal females. The participants were subsequently classified into 2 groups. NORM group comprised participants without osteopenia/osteoporosis and functional sarcopenia who showed a T score of the areal bone mineral density (aBMD) of the femoral neck higher than –1.0, handgrip strength greater than 18 kg, and gait higher than 1.0 m/s (n = 10). Participants with a T score of the aBMD of the femoral neck equal to or less than –1.0, handgrip strength lower than 18 kg, and gait speed equal or lower than 1.0 m/s (n = 20) were grouped in the EXP group (11,12). All surgical procedures including open reduction, and internal fixation with a volar plate fixation device, were performed by a single orthopedic surgeon. Participants with disorders which might have influences on bone and muscle metabolism such as malignancy, thyroid disorder, uncontrolled diabetes mellitus, neurodegenerative diseases, rheumatoid arthritis, liver disease, and kidney failures or history of taking bone active medications including hormone modulating agents, anabolic agent, bisphosphonates, and denosumab were excluded from the study, as were participants who were uncommunicative. Each participant’s height, weight, and circumferences of the calf and upper arm were measured by a qualified nurse.

The research protocol was approved by the Institutional Review Board of the University Hospital (KBSMC-2022-06-044). Informed consents were obtained from all participants.

Assessment of Handgrip Strength and Gait Speed

A single orthopedic surgeon (S.W.H.) followed the technique outlined in the previous study to measure handgrip strength using a hand dynamometer (Jamar 5030J1 Hydraulic Hand Dynamometer, Sammons Preston Rolyan, Bolingbrook, USA) at the noninjured contralateral hand (13). Each participant performed 3 times of handgrip measurements and the mean value was then computed. If the dominant hand was injured, the 10% rule was used to estimate handgrip strength (14). Walking speed over 6 m was measured. The average values were obtained from 3 independent measurements (11).

Measurement of Bone Mineral Density

Dual-energy x-ray absorptiometry (DEXA) with a Hologic device (Horizon-W; Hologic Inc., Bedford, MA) was used to assess the aBMDs (g/cm²) of the total hip, femoral neck, and lumbar spine (15). T scores of the aBMDs of the total hip, femoral neck, and lumbar spine were evaluated based on the value of aBMDs according to the standard technique of the Korean Society of Osteoporosis and Hologic Discovery software.

Reconstruction of the 3D Bone Model and Bone Quality of Distal Radius

A 256-slice multidetector computed tomography (CT) scanner (Brilliance iCT 256, Philips Medical Systems, Amsterdam, the Netherlands) was used to capture wrist CT images just after the closed reduction of fracture at the Department of Emergency Medicine and prior to open reduction surgery. The scanning protocol and the bone reconstruction methodology for wrist model and the method used to measure the cortical and trabecular bone density and cortical bone thickness of the distal radius bone were in accordance with the previous reports (16–19). The average cortical and trabecular bone model Hounsfield units (HU), total HU, and cortical thickness of the wrist bone models were calculated automatically using the Mimics 22.0 software and 3 matics 14.0 software (Materialise, Antwerp, Belgium).

Muscle Specimen Harvesting

During the open reduction and internal fixation procedures for the distal radius fracture, a single orthopedic surgeon (S.W.H.) harvested all pronator quadratus muscle samples. A 4-cm longitudinal incision was made along the radial border of the flexor carpi radialis (FCR) tendon. The FCR tendon was gently moved toward the ulnar side, and the dorsal surface of the FCR sheath was incised to expose the flexor pollicis longus tendon. The flexor pollicis longus tendon was also moved toward the ulnar side to expose the pronator quadratus muscle. A small piece of the pronator quadratus muscle (1 × 1 × 2 mm³) was taken from all participants while elevating the muscle from the cortical surface of distal radius. The collected muscle samples were placed in cryotubes, immediately frozen using liquid nitrogen, and stored at –80°C until total RNA extraction.

Total RNA Isolation and RNA Sequencing

Frozen pronator quadratus muscle samples were homogenized and total RNA was extracted using Qiazol Lysis Reagent (Qiagen, Germany). Genomic DNA contamination was removed using RNase-free DNase I (Qiagen, Germany). Using NanoDrop 2000 (Thermo Fisher Scientific Invitrogen Inc., Waltham, MA, USA), the amount of total RNA was determined. All samples passed quality control RNA integrity analysis (RIN ≥ 7). The TruSeq Stranded Total RNA Library Prep Gold Kit (Illumina Inc, San Diego, CA, USA) was used to construct cDNA libraries in accordance with the manufacturer’s protocol. To produce 101 bases paired-end reads, the total RNA was sequenced using the Illumina NovaSeq 6000 system (Macrogen, Seoul, Korea). After sequencing, the indexed samples were demultiplexed before the generation of FASTQ files for analysis and assessed by FastQC version 0.11.7.

All libraries were aligned to the GRCh37 human genomic reference using HISAT2 version 2.1.0 with the best score matches reported for each read. The mapped reads were assembled utilizing the STRING Tie version 2.1.3b. Based on estimated transcripts per kilobase of transcript per million (TPM), inter-gene expression comparisons were made. The expression level was normalized by calculating TPM-mapped reads. For differentially expressed gene (DEG) analysis, the value of log₂ (fold change) was calculated. The DEGs with an adjusted p ≤ .05 by independent t-test and log₂ (fold changes) ≤ –1 or ≥ 1 were identified.

Bioinformatics Analysis

For bioinformatic analysis, transcripts with a log₂ (fold changes) ≤ –1 or ≥ 1 with an adjusted p ≤ .05 were used. Unsupervised hierarchical clustering was used to build a heatmap for significant DEGs using MeV software (https://sourceforge.net/projects/mev-tm4/). DAVID (https://david.ncifcrf.gov/home.jsp) was used to assess gene ontology (GO), biological pathways including biological process, molecular function, and cellular components. PANTHER (http://pantherdb.org) determined the comprehensive information about the evolution of protein-coding gene families. The significant DEGs were imported to ingenuity pathway analysis (IPA; Ingenuity System, Inc., Redwood City, CA, USA) software for canonical pathway and network analysis.

Statistical Analysis

The Shapiro–Wilk normality test revealed that all of the data were normally distributed, hence parametric tests were applied. Independent t-tests were used to assess the differences in demographic features, handgrip strength, gait speed, and BMD between the NORM group and EXP group. The level of significance was set at p < .05 on clinical parameters. Adjusted p values were applied for comparison of fold change of RNA expression between groups.

Real-Time Polymerase Chain Reaction for Validation of Gene Expression

To validate the differential expression levels of the 6 hub genes from bioinformatic analysis, gene expression levels were examined using quantitative real-time polymerase chain reaction (RT-PCR). The 6 network hub genes including COL1A1, COL1A2, COL3A1, MMP2, MMP14, and SERPINH1, which showed significant enrichment values in PANTHER GO process, pathway and protein class and IPA network analysis were selected. RT-PCR reactions were performed with an ABI 7500 (Applied Biosystem, Foster City, CA) using a SYBR Green PCR kit (Takara Bio, Shiga, Japan), and cDNA was synthesized using 100 µg of total RNA (Thermo Fisher Scientific Invitrogen Inc.). The target genes and their primer sequences were as follows: COL1A1: 5’-GGGTGACCGTGGTGAGA-3’, 5’-CCAGGAGAGCCAGAGGTCC-3’; COL1A2: 5’-TCCAAGGACAAGAAACAC-3’, 5’- GCAGCCATCTACAAGAAC-3’; COL3A1: 5’-TGGTCTGCAAGGAATGCCTGGA-3’, 5’- TCTTTCCCTGGGACACCATCAG-3’; MMP2: 5’- GGCACCCATTTACACCTACACCAA-3’, 5’- GCTTCCAAACTTCACGCTCTTCAG-3’; and MMP14: 5’- CTCCTGCTCCCCCTGCTCACG-3’, 5’- CTCACCCCCATAAAGTTGCTG-3’; and SERPINH1: 5’- CCACTGTCGCCCAGATTATTTA-3’, 5’- GTCTCCCGCCCCTCACCT-3’. Each sample was assessed in triplicate and normalized against the expression level of the housekeeping gene GAPDH (5’- CCATCTTCCAGGAGCGAGATC-3’, 5’- GCCTTCTCCATGGTGGTGAA-3’). The gene expression results were obtained using the formula 2^–(ΔCt), and the fold change was calculated by the formula 2^–(ΔΔCt).

Human Bone Marrow Stem Cell Osteogenic Differentiation Induction

Immortalized human bone marrow stem cells (hBMSC; hMSC-BM-c) were cultured at 37°C in Dulbecco’s Modified Eagle Medium with 10% fetal bovine serum and 1% penicillin/streptomycin (HyClone, Logan, UT). For osteogenic differentiation, hBMSCs were cultured in α Modified Eagle Medium containing 0.1 μM dexamethasone, 50 μg/mL ascorbic acid, and 10 mM β-glycerophosphate for 14 days. The medium was replaced every 3 days. RT-PCR of ALP, RUNX2, Osteocalcin, and Osterix were performed on the 14th day of osteogenic differentiation. Expression levels were compared using Kruskal–Wallis test or Mann–Whitney U-tests. The level of significance was set at p < .05.

Cell Treatment

Recombinant human pro-collagen COL1A1 (R&D System, Minneapolis, MN, USA, 6220-CL) 75 μg/mL and recombinant human pro-collagen COL3A1 (R&D System, 7294-CL) 75 μg/mL were incubated with 5 μg/mL recombinant human BMP-1 (R&D, 1927-ZN) for 2 hours. Final concentration of pro-collagen COL1A1 0.1%, pro-collagen COL3A1 0.01%, BMP-1 0.150 μg/mL, pro-collagen COL1A1 treated with BMP-1, and pro-collagen COL1A1 treated with BMP-1 were individually administered to hBMC, each in triplicate.

Recombinant human MMP-2 (Peprotech, 420-02) was diluted to 50 μg in TCNB buffer (50 mM Tris, 10 mM CaCl2, 0.15M NaCl, 0.05% Brij 35) and treated with 1 mM APMA at 37°C for 2 hours for activation. MMP-14 (R&D System, 918-MP) was diluted to 40 μg/mL. Recombinant human MMP-2 and MMP-14 were administered to hBMC, each in triplicate.

Results

Demographic Features of Participants and BMD

Differences between the NORM and EXP groups in age (p = .066), body mass index (BMI) (p = .376), circumferences of the upper arm (p = .958) and calf (p = .347), and time period between fracture and surgery (p = .145) were not statistically significant. On the other hand, handgrip strength (p = .002), gait speed (p < .001), aBMDs of the total hip (p < .001), and femoral neck (p < .001) displayed significant differences between the 2 groups. The parameters of local bone quality of the distal radius such as total bone density (p = .035), cortical bone density (p = .016), and cortical thickness (p = .047) of the distal radius bone showed significant differences between the groups (Table 1).

Differentially Expressed Genes in the Muscle of Patients

There were 35 993 transcripts in total, but after removing transcripts with a TPM value less than 5, 16 919 transcripts remained to be analyzed. A total of 115 genes were differentially expressed in the EXP compared with those in the NORM (|log2 (fold change) | ≥ 1, p < .05) when the TPM baseline genes were excluded. 102 genes were upregulated and 13 genes were downregulated (Figure 1A and B). The EXP was found to be clustered in a heatmap from unsupervised hierarchical clustering analysis (obtained from MeV; Figure 1C) on the other hand, the multidimensional scaling plot did not reveal clear clustering patterns of NORM and EXP (Supplementary Figure 1). The genes with the top 5 highest upregulated fold changes in EXP were COL1A1, COL3A1, COL1A2, SFRP4, and COL5A1, and those with the top 5 downregulated fold changes were MYL3, ATP2A1, CALML6, PYGM, and LINC01405 (Table 2).

GO Enrichment Analysis

GO analysis was performed to analyze the biological process, molecular function, and cellular component. The results from DAVID demonstrated that biological processes of the DEGs identified by GO enrichment analysis comprised collagen fibril organization, collagen biosynthetic process, cell adhesion, osteoblast differentiation, ossification, bone development, wound healing, and skeletal system development (Figure 2A). Likewise, cellular component and molecular function involved mechanisms related to the maintenance of extracellular matrix structures and collagen synthesis (Figure 2B and C).

The GO-Slim biological process analyzed by PANTHER demonstrated the significance of extracellular matrix structure (Supplementary Table 1). In the PANTHER database, significant GO-Slim molecular function and cellular components were absent. The role of attached muscle in the maintenance of BMD and proper muscle mass and function would be linked with collagen biosynthetic activity, collagen structure integrity, cell adhesion, signaling pathway, and preservation of extracellular matrix structures in muscle fibers.

Pathways, Disease, Function, Protein Coding, and Network Analysis

Three pathways and 5 protein classes were identified by PANTHER with statistical significance (adjusted p < .01). The pathway from PANTHER suggested the importance of integrin signaling pathway and inflammatory reaction in muscle–bone interaction (Supplementary Table 2). The protein class related to extracellular matrix protein, protein degradation, and cell adhesion appeared to have potential roles in muscle–bone interactions (Supplementary Table 3).

Seven canonical pathways were differentially regulated according to IPA annotation (|z-score| ≥ 2.0, adjusted p < .001) with statistical significance. Wound healing signaling pathway, matrix metalloprotein (MMP), and collagen degradation in muscle have all been identified as potential contributors to the molecular background of muscle–bone interaction by the canonical pathway from IPA (Supplementary Table 4). Also, the importance of connective tissue development and function was highlighted by network analysis (Supplementary Table 5, Figure 3A). Inflammatory reaction, MMP activity, and integrin signaling could play a part in muscle–bone interaction as demonstrated by the PANTHER pathway and by the IPA network analysis (Supplementary Table 5, Figure 3B and C). There were found to be 4 significant categories of disease or function (|z-score| ≥ 2.0, p < .001) including abnormal bone density and osteoporosis (Supplementary Table 6). The fragile bone structure would be affected by abnormal collagen degradation signaling process and wound healing capacity of attached muscle tissue.

Validation Real-Time Polymerase Chain Reaction

The gene expression patterns of 6 hub genes, including COL1A1, COL1A2, COL3A1, MMP2, MMP14, and SERPINH1 transcripts which showed significant results in both GO enrichment analysis and IPA network analysis were tested via real-time PCR. MMP2, MMP14, and SERPINH1 were related to maintaining extracellular matrix organization, and COL1A1, COL3A1 seemed to have an influence on integrin signaling pathway. COL1A2, COL3A1, and SERPINH also seemed to have associations with type III collagen activity, muscle fibrosis, and wound healing capacity, and COL1A1 was involved in inflammatory signaling mechanisms. Significant differences were observed in COL1A1 (p = .021), COL1A2 (p = .014), COL3A1 (p = .016), MMP2 (p = .031), and MMP14 (p = .017) transcripts (Supplementary Figure 2).

Human Bone Marrow Stem Cell Osteogenic Differentiation Induction

The hBMSC treated with MMP-2 or MMP-14 exhibited significantly lower mRNA expression levels in ALP. Conversely, hBMSC treated with pro-collagen COL1A1 with BMP-1 or pro-collagen COL3A1 with BMP-1, showed significantly lower mRNA expression levels in RUNX2. The hBMC administered with pro-collagen COL1A1 treated with BMP-1, pro-collagen COL1A1 with BMP-1, MMP-2, and MMP-14 showed decreased levels of mRNA expression levels in Osteocalcin (Supplementary Figure 3). Hence, type I collagen, type III collagen, peptide MMP-2, and MMP-14 appeared to have an impact on osteoblastogenesis activity of hBMSC.

Discussion

Bone and skeletal muscle continuously exhibit physical interplay to adapt to environmental requirements and maintain metabolic homeostasis (1–8). Previous studies have attempted to reveal the mechanical and biochemical interactions between 2 main components of the musculoskeletal systems, skeletal muscle, and bone. Numerous myokines, osteokines, adiponectin, senescence-associated secretory phenotypes, and micro-RNA have been proposed as mediators between them (1–8) but the clear molecular interactive mechanisms between these 2 endocrinological organs have not been thoroughly identified. By comparing the transcriptional profiles of total RNA from muscle tissue of females with distal radius fractures and functional sarcopenia to those of healthy controls, the current study sought to clarify the molecular basis of muscle–bone interaction in patients with fall and fracture history.

The probable significance of integrin signaling in muscle–bone interaction was the novel discovery of the present study. The importance of integrin signaling was proven by the pathway analysis by IPA and PANTHER. Integrins are the family of cell-adhesion molecules, which mediate adhesive interactions with cells in the extracellular matrix (20). Prior research has suggested that integrin and irisin, a myokine that can improve bone formation action interact with each other (21,22). Integrin may play a part in irisin’s anti-apoptic impact on osteocytes by acting as the receptor for irisin on osteocyte (22). Abnormal integrin expression and signaling in muscle tissue might hinder the ability of irisin to promote bone formation. As a consequence, this inhibition could lead to a decrease in bone density at sites where muscle and bone are closely attached.

The interactions between MMP and collagen fiber integrity in muscle appear to be crucial in the complex relationship between muscle and bone. The collagen biosynthesis and activities of MMP, particularly MMP-2 and MMP-14, have been identified as important players in the muscle–bone interaction. Both the canonical pathway and network analysis from IPA and the pathway and protein class from PANTHER have supported the significance of these factors in the interactions between muscle and bone. The elastolytic activity of MMP-2 (23) and negative correlations between serum levels of MMP-2 and proximal femur BMD have been proposed in previous reports (24). Similarly, the role of MMP-14 on skeletal muscle remodeling and osteoclastic activity associated with receptor activator of NF-kappa B ligand shedding in osteoblast (25) also have been suggested. MMP-14 and MMP-2 exhibited complementary activities, with MMP-14 having type I collagenolytic activity, whereas MMP-2 had gelatinolytic activity (26). They are involved in extracellular matrix deposition, fibrosis, and progressive muscular atrophy, as previously suggested (27). Although MMP activity is recognized to play a role in injured muscle healing (28), the lack of statistical significance in the time period between fracture to muscle harvesting between the groups suggests that the variation in of mRNA expression patterns associated with MMP activities cannot be solely attributed to muscle injuries and bone fractures. Hence, proteolytic activity of MMP-2 and MMP-14 can potentially alter the integrity of collagen structure and muscle mass and function. Their detrimental effects on ossification may cause decreased bone quality in muscle attachment sites by paracrinological pathways.

The importance of inflammatory reactions in muscle–bone interaction was demonstrated by the findings of GO, IPA, and PANTHER analysis. The muscle tissue isolated from a fracture site is expected to generate inflammatory responses. However, as mentioned earlier, the lack of statistical significance in the time period between fracture to muscle harvesting between 2 groups suggested that the differences in gene expression patterns observed in patients with low BMD and sarcopenia may be result of complex interactions between bone and muscle beyond the immediate effects of bone and muscle healing. Innate immune responses are the primary component of pro-inflammatory mediators. The pro-inflammatory cytokines including interleukin-6, TNF-α, and interleukin-1β on pathogenesis of sarcopenia and osteoporosis have been demonstrated previously (29–32). Among them, IL-6 and TNF-α appear to play a part in preserving muscle–bone homeostasis as myokines (18,33). Hence, the results from RNA sequencing data validated the importance of inflammatory response in molecular background muscle–bone interaction, as suggested in previous reports.

Interestingly, the aforementioned results showed increased expression levels of genes, including COL1A1, COL3A1, and COL1A2 which are associated with muscle fibrosis in patients with low mineral density and functional sarcopenia. Skeletal muscle fibrosis is one of the significant pathological manifestations related with sarcopenia (34,35). Irisin, one of the well-known myokines that can improve bone formation, also can attenuate muscle cell senescence and skeletal muscle fibrosis by amelioration of redox imbalance (36). Therefore, altered irisin activity may have multiple associations and not only be linked to abnormal integrin signaling but also to other factors, such as decreased muscle mass, muscle fibrosis, and lower muscle mass and function.

Type III collagen synthesis is generally increased during the healing process in a variety of tissues including tendon, ligament, skin, and bone owing to its ability to form rapid crosslinks and stabilization of the repair sites (37). There is a significant amount of type III collagen in the fracture callus, but type I collagen eventually replace it (38). The role of COL1A1 and COL1A2 which encode the 2 subunits of type I collagen in fracture healing and muscle injury also have been proposed, previously (39). The females with lower aBMD of the distal radius bone may have a risk of more severe bone destruction and attached muscle injury compared to those with normal aBMD after bone fracture. Therefore, the increased expression levels of genes related with collagen type I and III integrity such as COL1A1, COL1A2, and COL3A1 genes partially influenced by compensatory reaction from more severe bone destructions and attached muscle injuries in participants.

The significant novelty of this study lies in the integration of standard RNA sequencing data obtained during the reduction of distal radius fracture into a stratified patient cohort characterized by clinical parameters including decreased bone mass, handgrip strength, and gait speed. However, there are several limitations. First of all, due to female preponderance of osteoporosis and sarcopenia, only female participants were included in this study; therefore, limited information about muscle–bone interaction could be provided. Second, because pronator quadratus muscle is a relatively minor nonweight-bearing wrist muscle, proving a clear signaling pathway and mechanism between bone and muscle would be difficult. Furthermore, the absence of analysis of other associated musculoskeletal tissue, such as bone, enthesis, and tendon tissue, revealing complex mechanisms of interplay between bone and muscle at molecular muscle would be hard to be expected. Finally, because pronator quadratus muscle sample without fracture history and surgical procedure are impractical, acquiring muscle samples from healthy individuals is quite challenging. Consequently, there is a notable gap in studies addressing transcriptional data in healthy subjects without fracture of osteopenia/osteoporosis. Nonetheless, this study holds significance as a meaningful step toward exploring the molecular background of the muscle–bone interaction in patients with osteopenia and sarcopenia.

Associations between alterations in the transcriptional profiles of the attached muscle and local bone quality may provide insights into the molecular background of the muscle–bone interaction. The transcriptional profiles of patients’ muscle samples utilized in this study exhibits impaired integrin signaling pathway, inflammatory reaction, and fibrotic changes. Potential therapeutic target for geriatric conditions such as falls, fractures, osteoporosis, sarcopenia, and even osteosarcopenia would focus on the above molecules and pathways to delay musculoskeletal aging and overcome frailty.

Funding

This research was supported by the two National Research Foundation of Korea (NRF) grant funded by the Korea government (No. 2020R1I1A1A01071537).

Conflict of Interest

None.

Data Availability

RNA-seq data in this study are available from the corresponding author on reasonable request.

Author Contributions

Using contributor roles taxonomy (CRediT) terms, all authors participated in multiple contributing roles. J.-H.K.: data curation, formal analysis, funding acquisition, investigation, methodology, construction of original draft. J.-H.B.: data curation, investigation, critically reviewed the manuscript. J.K.L.: data curation, methodology. S.-W.H.: conceptualization, data curation, investigation, methodology, supervision, validation, critically reviewed the manuscript.

References