Background
Diabetes mellitus (DM) is a major public health concern that is approaching epidemic proportions globally. There are 425 million people worldwide who have DM, and this number is expected to increase to 629 million by 2045 [
1]. DM is associated with an extensive list of complications involving nearly every tissue in the body, including the heart, blood vessels, eyes, kidneys, and nerves. It has been reported that DM adversely affects bone health and is associated with reduced bone mineral density or reduced bone strength, resulting in a higher risk of fracture [
2]. Clinical studies have shown higher incidences of delayed union and nonunion, as well as a doubling of the time to healing of fractures in patients with DM, compared with healthy patients [
3,
4]. In patients with DM, fracture healing time is prolonged by up to 87% [
3]. In addition, Hernandez et al. reported higher odds of delayed union in patients with DM [
4]. Despite reports of associations between DM and impaired fracture healing, including delayed union and nonunion, only a few investigations have examined the molecular mechanisms by which DM affects the process of fracture healing [
5‐
7].
microRNAs (miRNAs) are short, single-stranded, non-coding RNA molecules that regulate gene expression [
8]. miRNAs play pivotal roles in biological processes, including cellular differentiation, cell growth, and organ development [
9]. In the field of skeletal biology, several studies showed that miRNAs contribute to the regulation of osteoblast, chondrocyte, and osteoclast differentiation and function, suggesting that miRNAs have important roles in bone formation, resorption, remodeling, and repair [
10‐
12]. In addition, the involvement of miRNAs in fracture healing and nonunion has been demonstrated [
10,
13‐
18].
Recently, some miRNAs were reported to be involved in the pathology of DM and its complications [
19‐
21]. Chronic inflammation is an important determinant of insulin resistance and contributes to microvascular complications of type 2 DM, such as nephropathy, neuropathy, and retinopathy. miR-146a has been reported to regulate genes involved in the pathogenesis of type 2 DM and its related complications by attenuating inflammatory cytokine production in macrophages [
20]. Grieco et al. reported hyperexpression of miR-148a and miR-21-5p in serum/plasma from patients with type I DM. miR-148a and miR-21-5p target pivotal genes involved in bone development, modeling, and remodeling (e.g., wingless-type MMTV integration site family, member 1 (WNT1), and transforming growth factor-β 2 (TGF-β2)), thus suggesting that miR-148a and miR-21-5p serve as contributing factors to bone fragility or as potential biomarkers of bone metabolic alterations in patients with type 1 DM [
21]. However, only a few studies have shown direct roles for miRNAs in impaired fracture healing in patients with DM or in animal models of DM [
22].
The elucidation of miRNA expression changes during fracture healing may reveal valuable information regarding the pathophysiology of the healing process in patients with DM and suggests potential therapeutic interventions to accelerate fracture healing. This study was conducted to examine miRNA expression profiles during fracture healing of the femur of rats with DM by using microarray analysis and to elucidate the dynamic expression patterns of expressed miRNAs during fracture healing by using real-time polymerase chain reaction (PCR) analysis.
Discussion
Recently, miRNAs have emerged as key regulators in the complex process of fracture healing [
13‐
15]. In addition, some miRNAs are reportedly involved in the pathology of DM and its complications [
19‐
21]. To better understand the underlying pathogenesis of impaired fracture healing in DM patients, we focused on miRNA expression profiles in fracture healing of the femur in diabetic rats. In the current study, we identified five miRNAs, miR-221-3p, miR-339-3p, miR-376a-3p, miR-379-5p, and miR-451-5p, which were expressed differentially with changing patterns during fracture healing in diabetic rats when compared with control rats. These findings provide new insights into the cellular processes involved in impaired fracture healing in patients with DM.
Bone fracture healing involves remarkably complex processes that require the coordination of a sequence of many biological events [
32]. The rationale for the selection of two time points, post-fracture days 5 and 11, for microarray analysis was that key cellular events of the fracture healing process in healthy rats take place on days 5 and 11. The process of fracture healing can be divided into three overlapping phases: inflammation, repair, and remodeling [
32]. Post-fracture day 5 corresponds to the transitional period from the inflammatory phase to the repair phase [
33]. Post-fracture day 11 corresponds to the repair phase, which involves the processes of intramembranous and endochondral ossifications, including osteogenesis, chondrogenesis, and vascular invasion [
33]. Imbalance in inflammatory responses, reduced proliferation and differentiation of osteoblasts and chondrocytes, and reduced function and alteration in vascularization have been implicated as plausible pathogenic mechanisms underlying impaired fracture healing in patients with DM [
2]. Previous studies showed that long-bone fractures of STZ-induced diabetic animals result in smaller calluses with reduced bone and cartilage formation, reduced proliferation and differentiation of osteoblasts and chondrocytes, and reduced mechanical strength [
6,
7,
34]. In agreement with these previous studies, radiographic assessment in the present study showed that the fracture healing process in the DM group was delayed at days 21 and 28, compared with healing in the control group. Histological evaluation also revealed a delayed fracture healing process in the DM group, which was characterized by a smaller cartilage callus and a delay in endochondral ossification (Supplemental materials).
Inflammation is a critical factor during fracture healing, with inflammatory cells and molecular factors appearing locally at the fracture site in a distinct spatial and temporal manner [
33]. Highly regulated inflammatory signaling during fracture healing is essential for priming bone regeneration. Disturbances in the finely tuned inflammatory responses at the fracture site lead to impaired vascularization, reduced bone formation, disturbed osteoclastic function, and subsequent delayed fracture healing or nonunion [
35]. miR-339-3p was reported to inactivate the Akt/mammalian target of rapamycin (mTOR) signaling pathway and to alleviate inflammation and edema caused by severe acute pancreatitis with acute lung injury in a mouse model by targeting annexin A3 (ANXA3) [
26]. In addition, rapamycin, an inhibitor of mTOR, inhibits osteogenesis both in vitro and in vivo [
36]. Inhibition of mTOR by rapamycin blocked insulin-like growth factor 1-induced osteoblast differentiation of mesenchymal stem cells and mineralization [
36]. miR-451-3p reduces inflammation by suppressing phosphorylation of p38 mitogen-activated protein kinase (MAPK) via 14-3-3ζ and Rab5a [
25]. p38 MAPK plays an important signaling role in orchestrating injury or stress-induced responses, as well as in bone formation [
37]. Pro-inflammatory cytokines, such as tumor necrosis factor-α and interleukin (IL)-1a, activate p38 MAPK; activation and signaling of p38 MAPK also lead to the production of these pro-inflammatory cytokines and their signal transduction. The interactions between p38 MAPK and pro-inflammatory cytokines are important for controlling life and death signaling cascades in osteoblasts and chondrocytes. Taken together, miR-339-3p and miR-451-3p may play important roles during fracture healing by regulating inflammation.
Chondrogenesis is an essential component of endochondral ossification in fracture healing [
32]. miR-376a-3p is one of the most important miRNAs to trigger chondrogenesis [
27]. Bakhshandeh et al. [
27] reported that the expression of miR-376a was downregulated during the early phases of chondrogenic differentiation in unrestricted somatic stem cells. Bone morphogenetic protein (BMP) receptor 2 and TGF-β receptor 1 are putative target genes of miR-376a [
27]. Many studies have demonstrated that BMP-2 induces chondrogenic differentiation in various types of stem cell [
38]. The recent study revealed that BMP-2 and TGF-β1 cooperate during chondrogenic differentiation of synovium-derived mesenchymal stem cells [
39]. In addition, miR-379-5p, which is predicted to target activating transcription factor 3 (ATF3), regulates the endochondral ossification process by modulating proliferation and hypertrophic differentiation of the growth plate in male rats [
29]. James et al. demonstrated the importance of ATF3 in the control of endochondral bone growth and skeletal development [
40]. Some fundamental aspects of fracture healing, particularly endochondral ossification, have clear parallels with long bone development [
41]. The altered expression of miR-376a-3p and miR-379-5p at the fracture site in rats with DM might dysregulate chondrogenesis and thus inhibit endochondral ossification, which could lead to impaired fracture healing.
Angiogenesis and bone formation are coupled during fracture healing [
42]. A lack of oxygen (hypoxia) in fractured bone and the subsequent generation of angiogenic factors are critical for achieving successful fracture healing [
43]. miR-451-5p was reported to inhibit cell growth, migration, and angiogenesis via downregulation of IL-6R [
31]. miR-221-3p regulates angiogenesis by directly reducing the levels of c-Kit [
28]. c-Kit and its ligand, stem cell factor (SCF), play an important role in the proliferation and differentiation of hematopoietic progenitor cells [
44]. In addition, miR-221-3p inhibits the SDF-1/CXCR4 signaling pathway in human chondrocytes [
30]. The SDF-1/CXCR4 axis plays a pivotal role in fracture healing by affecting the migration and differentiation of progenitor cells at fracture sites [
7,
45,
46]. Kitaori et al. [
45] reported that SDF-1 was induced in fractured bone and promoted endochondral bone formation by recruiting mesenchymal progenitor cells to the site of injury. Moreover, the SDF-1/CXCR4 axis acts as a key regulator of angiogenesis and contributes to the regulation of endothelial progenitor cell (EPC) recruitment in ischemic tissues. EPCs represent a small population of blood cells that can differentiate into endothelial cells and participate in postnatal angiogenesis [
47]. In addition, EPCs are involved in physiological and pathological angiogenesis/vasculogenesis [
48]. Recently, Arakura et al. [
7] revealed that gene expression levels and localizations of SDF-1 and CXCR4 at fracture sites were altered during fracture healing in animals with experimental DM, which may contribute to the impaired fracture healing associated with inhibition of endochondral ossification and angiogenesis. Taken together, the altered expression patterns of miR-221-3p and miR-451-5p in the DM group during fracture healing suggest that disturbed angiogenesis/vasculogenesis and endochondral ossification may occur at fracture sites, potentially leading to impaired fracture healing.
The present results have clinical implications. Due to the increasing number of patients with DM, the incidence of impaired fracture healing associated with DM is expected to increase. It is desirable to establish strategies to prevent impaired fracture healing and enhance bone healing in patients with DM. Many biological and biophysical interventions have been attempted to accelerate fracture healing in these patients, including the use of exogenous growth factors, such as bone morphogenetic proteins, low-intensity pulsed ultrasound, and stem/progenitor cell transplantation [
49]. Although most of these strategies have shown relatively satisfactory results, there are some notable limitations to their effectiveness and availability. The recent discovery of miRNAs, being powerful regulators in a variety of tissues and processes, suggests their potential as therapeutic target. Several therapeutic trials targeting miRNAs have been conducted [
50].
Notably, the use of miravirsen (Santaris Pharma A/S, Copenhagen, Denmark), an anti-miR-122 drug, in patients with chronic hepatitis C (HCV) infection showed prolonged dose-dependent reductions in HCV RNA levels without evidence of viral resistance [
51], which implies that miRNA-based therapeutics can indeed become a reality in clinical medicine. In the field of orthopedics, Murata et al. demonstrated that inhibition of miR-92a enhanced fracture healing in mice [
15]. The five miRNAs identified in our study might represent a tool for the establishment of therapeutic approaches, which either prevent impaired fracture healing or enhance fracture healing. Further in vivo functional analyses will be needed to clarify the exact role and therapeutic potential of each miRNA during fracture healing.
This study had a few limitations. First, we did not demonstrate specific regulatory activities exerted by these five miRNAs; hence, our results solely indicate associations with impaired fracture healing, rather than clear causative relationships. Further in vivo analyses, including the loss or gain of miRNA function tests, are needed to define the exact role of each miRNA during fracture healing. Second, our study did not validate target genes of the five selected miRNAs. Target identification of miRNAs is computationally difficult due to the relatively low homology between miRNAs and their targets. Further, miRNAs may target any of the listed target genes in one cell type but not in others (given the diversity of cell types found within the fracture callus), and as such many may not be related specifically to bone or cartilage cells [
17]. Further investigation is needed to clarify the contributions of each miRNA identified in this study to the downregulation of specific genes in the context of fracture healing process.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.