The skeletal system consists mostly of extracellular matrix and mineral salts. Osteoblasts and chondrocytes create enormous amounts of extracellular matrix proteins for skeletal systems to grow and sustain. Many skeletal metabolism related proteins are produced by endoplasmic reticulum (ER) and secrete outside cell membrane then form to extracellular oligometric matrix proteins and collagens. All is known that the ER is involved in a variety of important biological activities, including protein secretion, lipid synthesis, and calcium regulation [1,2]. As a result, an efficient regulation of ER activity is important to the survival of the organism in response to changes in the physiological environment. Conditions that disrupt ER function can decrease the ER's protein folding ability, resulting in the accumulation and aggregation of unfolded protein [3-6]. Cells engage a biochemical program termed an Unfolded Protein Response (UPR) route to counteract the detrimental consequences of ER stress. The UPR is involved in regulating the process of differentiation and growth of osteoblast and chondrocyte within the skeletal system [7-9]. The ER of osteoblast and chondrocyte has a well-established mechanism, which can improve their ability for UPR protein production during ER stress response.

The UPR comprises three signaling pathways triggered by three distinct transmembrane sensors– inositol-requiring enzyme 1 (IRE1,encoded by ERN1), activating transcription factor 6 (ATF6), and protein kinase R–like endoplasmic reticulum kinase (PERK, encoded by EIF2AK3) [10-12]. IRE1α is the UPR evolution's most conservative branch [13]. After the activation of IRE1α under ER stress, it clues specific exon intron sites into an mRNA, encrypting the transcription factor X-box-batches protein 1 (XBP1). This enhances the protein-folding capacity of the endoplasmic reticule and reduces the misfolded pro-ER1. This results in the production of active transcription factors and induction of different adaptive genes [14]. At the same time, the active IRE1α may also decrease the quantity of protein entering an ER via particular ER-localized cytoplasmic mRNA. This mechanism is known as IRE1-dependent decay control (RIDD)[13]. The JNK, ERK, and NF-κB path can also be activated by IeR1α through interaction with several adapters and regulatory proteins [15-17]. ATF6 is an endoplasmic transmembrane reticulum protein having in its cytoplasmic field bZIP transcription factor [18]. ATF6 translates under ER pressure into the Golgi apparatus, where it is split into two sites with the proteases site 1 (S1P) and site cleavage-2 (S2P). The ATF6 (ATF6-N) released amino terminus migrates to the nucleus and activates the chaperons for ER encoding, including those genes that increase the ERAD pathway. ATF6 can also be used to promote gene-expression patterns between Heterodimers and XBP1 [19]. PERK is a serine/threonine kinase that oligomerizes and trans-autophosphorylates under ER stress conditions, inhibiting general protein translation via phosphorylation of eukaryotic translation initiator factor-2 (eIF2) at serine, while selectively increasing ATF4 translation, and then reducing protein load into the ER [19]. ATF4 then activates a set of genes involved in regulating of antioxidant responses, ER folding capacity, amino acid metabolism, antioxidant responses, and macro-autophagy [13,18].

Obviously, eukaryotic cells have the above-mentioned signaling pathways from the ER to entire cell, to avoid overabundant cumulation of misfolded and unfolded proteins in the ER and maintain the ER homeostasis. Based on the literature, we assessed the bone metabolism-related proteins involved in the three pathways of UPR, and the role of these UPR signaling molecules in regulating bone metabolism biomarkers, providing a basis for the subsequent development of biologics and diagnosis and treatment.

Overexpression of IRE1α suppresses the differentiation of collagen II (ColII), SOX9, collagen X (ColX), metalloproteinase matrix 13 (MMP-13), hedgehog indian (IHH), RUNX2, and increased parathyroid-related peptide expression (PTHrP) [20]. The deficiency of IRE1α in chondrocytes downregulates prosurvival factors XBP1S and Bcl-2, which enhances the apoptosis of chondrocytes through increasing proapoptotic factors caspase-3, p-JNK, and CHOP. Meanwhile, the activation of IRE1α increases chondrocyte viability and reduces cell apoptosis [21]. The in vitro expression of cartilage formation indicators such as ColII, ColX and RUNX2, SOX9, and XBP1S can promote efficient cartilage development. Furthermore, XBP1S can also activate the precursor of granulin-epithelin (GEP, which increases the development of cartilages), which favorably regulates chondrocyte hypertrophy, mineralization and bone formation [22,23]. Besides, Chondrocyte apoptosis and proliferation are affected by the deletion of XBP1 in mice, and XBP1 signaling pathway is essential for alleviating mutant protein aggregation in ER-stress related skeletal disease, it presents a therapeutic target for aggregation related conditions in cells undergoing proliferation [24]. A high-throughput sequencing result found that there are multiple ncRNAs in cartilage XBP1 knockout mice, and these differentially expressed ncRNAs help to clarify the occurrence and development of bone development-related diseases and help identify prospective clinical markers [25].

OA is a degenerative common ailment that has an important impact on the quality of life of aged people, causing annoyance, pain, and sometimes disability [26]. The breakdown of cartilage in OA is characterized by aberrant cartilage cell metabolism and extracellular matrix degradation [27]. Chondrocyte metabolic balance and OA symptoms can be influenced by altering ER stress and autophagy by mechanical loading [28]. In the lumen of ER, excessive mechanical stress induces Ca²⁺ overflow, accumulation of malfunctioning proteins in the ER lumen and ER stress induction [28]. In addition, molecular chaperone disorders can also cause severe ER stress, which will lead to severe forms of OA, accompanied by increased cell death of chondrocytes [29]. ER stress was observed in chondrocytes in osteoarthritis (OA) patients, and ER stress and apoptosis increased throughout OA development [30]. In moderate OA cartilage, but not in mild or severe cartilage, XBP1 mRNA splicing has been enhanced [31]. Researchers also thought that the signal axis of IRE1-mTOR-PERK is used to control OA-derived inhibition and activation of Apoptosis [32]. The IRE1 driven autophagy flow is stopped by mTOR, and PERK blocks the subsequent fusion of automotive phagosomes and lysosomes. mTOR promotes PERK to enhance caspase-12 and CHOP production and to induce apoptosis of ER-stressed cells. Chain protein (BiP) expression has dramatically enhanced following medial meniscus (DMM) surgery in wild-type mouses and shows clearly that ER stress and UPR are implicated in OA development [33]. After DMM surgery in wild-type mice, BiP expression increased significantly, clearly indicating that ER stress and UPR are involved in the pathogenesis of OA. Although, a misfolded version of thyroglobulin (Tg cog) was overexpressed under the Col2a1 promoter (ColIITg cog, a model of cartilage-specific ER stress [34]) in transgenic mice following the DMM operation, BiP levels have also been increased but OA ontake have been delayed, and a delay in chondrocyte apoptosis and damage two weeks after surgical operation were also found in comparison with wild-type control. Before DMM surgery, ColIITg cog mice increased its immunoglobulin heavy BiP level and the ColIITg cog mice pro-survival XBP1 network of signals was activated in ColIITg cog mice after DMM [34], indicating that the ER stress pre-exposure or a specific pro-survival UPR response was activated. Chondroprotective impact may occur in the XBP1S signal. Besides, XBP1S inhibits ER stress-mediated apoptosis and nitrite production in OA cartilage, and XBP1S also acts as a negative regulator of apoptosis in osteoarthritis by affecting caspase 3, caspase 9, caspase 12, p-JNK1, and the pro-apoptotic ER stress marker C/EBP homologous protein (CHOP) [35]. Based on detailed analysis of gene expression patterns some researchers proposed that XBP1-independent UPR driven disruption of CCAAT/enhancer-binding protein beta (CEBPβ or CEBPB, a master regulator of chondrocyte differentiation) is important for the skeletal disease pathophysiology [36].

We sent human chondrocytes overexpressing IRE1α to Kangcheng Biotech for differential expression profiling chip (DEPC) and ChIP-chip transcription factor microarray (CTFM), Raw signal intensities were normalized in the RMA method by NimbleScan v2.5, and low-intensity genes were filtered (Genes with Intensity ≥ 100.0 in at least 2 out of 3 samples were chosen for further analysis). Differentially expressed genes that passed Fold Change filtering (Fold Change ≥ 2.0). The DEPC results uncovered that after overexpressing IRE1a, the expression profile of 532 genes changed, 19 of which are related to bone development or bone-related diseases (Figures 1A and 1B). Then, Gene Ontology (GO), kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis was performed on these 19 differential genes. These genes are mainly involved in the following four aspects (Figure 2A): Osteogenesis imperfecta, bone trabecula formation, bone marrow development, and negative regulation of chondrocyte differentiation, and avascular necrosis of femoral head. Furthermore, the CTFM results show that XBP1S can directly regulate the development of bone through the pathway of ossification, osteoclast fusion, proliferation, differentiation etc. (Figure 3A). In these signal pathways, XBP1S can bind to the growth factor of connective tissue (CTGF) to control chondrocyte differentiation and proliferation or in combination with Oxytocin (OXT), to regulate ossification involved in bone maturation by bind to ras homolog gene family member A (RHOA) or in combination with TC cell leukemia (TCTA) to regulate osteoclast fusion (Figure 4).

XBP1 is critical for bone morphogenetic protein 2 (BMP2) induced osteoblast differentiation. If without IRE1, BMP2 therapy failed to augment the activity of alkaline phosphatase or expression of the osteoblast markers Osterix (a bone forming transcription factor required) and osteocalcin [8].The IRE1α-XBP1 pathway can activate Osterix transcript, whereas XBP1 binds directly to the promoter region of the Osterix gene [37]. Signaling XBP1 inhibition may severely damage osteogenic differentiation caused by proteasomic inhibitors (PIs). In addition, XBP1S able to transcript the expression of ATF4, while XBP1S' overexpression can induce the expression of ATF4 and other osteogenic differentiation genes including Col1a1, Osteocalcin (Ocn), Bone morphogenetic protein 2 (Bmp2, RUN2), and Osteopontin (Opn), thus driving different types of osteoblasts. The results show that XBP1 may be late in the process of differentiation of the osteoblast [38].

The removal of IRE1 from bone marrow cells caused an increase in bone mass was unexpected [39]. This is because XBP1 contacts the promoter directly and promotes gene transcription that encodes the nuclear coding of cytoplasmic 1 activated t-cells (NFATc1), a master regulator for bone absorbing osteoclasts [39]. Therefore, IRE1α deletion is linked to the poor differentiation of the osteoclastic. The xbp1 mRNA processing is considerably decreased by pharmacological inhibition or detection of these receptors and may be transiently activated during osteoclastogenesis, partially dependent upon Ca²⁺ oscillation mediated by inositol 1,4,5-trisphosphate receptors 2 and 3, (ITPR2 and ITPR3) in the reticulum of the endoplasma. The NF-κB ligand (RANK1) gene encodes an osteoclast factor secreted by osteoblasts, and the receptor activator of Rank1 can be induced to express by PTH signal [40]. BMP2-mediated Pth1r transcription increase in mouse embryonic fibroblasts can reverse silences of the Ire1α and XBP1 genes, as XBP1 is capability to adjust Pth1r transcription directly. In addition, PTH-induced Rank1 expression in primary osteoblasts can be suppressed by silencing of XBP1 gene, thus osteoclastic development in an in vitro osteoclastogenesis model is eliminated [41].

Enhanced chondrocyte hypertrophy, mineralization, and endochondral bone by overexpression of ATF6 in mice [42], and knock-down of ATF6 prevents differentiation of chondrocytes hypertrophy [43]. BMP2 enhances the differentiation of osteoblast and mineralization by expressing ATF6 triggered by RUNX2, which in turn promotes osteocalcin transcription [44]. ATF6α also improves the chondrocyte hypertrophy of RUNX2 and changes the pathway to in vitro production of the crucial Indian hedgehog and parathyroid hormone cartilage [43]. Interestingly, osteocalcin expression has risen following co-expression of ATF6, RUNX2, and ATF4, indicate that ATF6 may have a synergistic impact on expression of osteocalcin [45]. Furthermore, ATF6 overexpression in human dental pulp cells shows that distinction between dentistry and matrix mineralization may be regulated by it [46]. ATF6 aggravates angiogenesis-osteogenesis coupling during ankylosing spondylitis by mediating fibroblast growth factor 2 (FGF2) expression in chondrocytes [47]. The SLC26A2 gene is responsible for a variety of presently incurable human chondrodysplasias, including sarcomas. In SLC26A2-deficient chondrodysplasias, suppressing ATF6-dependent overactivation of the fibroblast growth factor receptor 3 (FGFR3) signaling ameliorates the condition [48].

The ATF6α route in OA, however, appears not crucial. Although the ER stress induction and the signaling of XBP1 has been demonstrated to be associated to ATF6α [35,42], ATF6α knockout mice may not be as relevant as other UPR branches, having normal articular cartilage and developing the same rates as wild types of controls after DMM [33]. The ADAMTS-5 and ADAMTS-4 expression, nevertheless, was reduced by a silencing of ATF6β [49]. Celastrol ameliorates endoplasmic stress-mediated apoptosis of OA via regulating ATF-6/CHOP signaling pathway [50], thus demonstrating how important a greater knowledge of the OA development is to the UPR.

Similarily, we also sent human chondrocytes overexpressing ATF6 to Kangcheng Biotech for DEPC and CTFM. The DEPC results showed that the expression of 362 genes has changed, 23 of which are related to bone development or bone-related diseases (Figures 1C and 1D). Analysis of these 23 differential genes using GO and KEGG pathways revealed that these genes are mostly engaged in the following aspects (Figure 2B): osteoclast development, growth plate cartilage development, osteoblast proliferation, osteogenesis imperfecta, negative regulation of bone resorption, positive regulation of cartilage development, embryonic cranial skeleton morphogenesis, osteoblast differentiation, skeletal system development, and Osteoclast differentiation. Besides, the ATF6 CTFM results found that ATF6 can directly regulate chondrogenesis and bone development through the pathway of endochondral bone growth, proliferation, differentiation and ossification, including vascular endothelial growth factor (VEGF) and IL17 signaling pathway (Figure 3B). ATF6 can directly bind to frizzled-related protein (FRZB), ubiquitin specific peptidase 1(USP1), FGFR3, CEBPB, and mitogen-activated protein kinase 1 (MAPK1) to directly regulate skeletal system development, endochondral bone growth, chondrocyte proliferation, and regulation of osteoclast differentiation and ossification. In these pathways, ATF6 can directly bind to CEBPB, MAPK1, splicing factor, arginine/serine-rich 1(SFRS1) or mitogen-activated protein kinase 13(MAPK13) in the IL-17 signaling pathway to directly regulate it (Figure 5).

In OA samples the ERK (PERK)1/2-positive phosphorylated chondrocytes were considerably greater than in healthy controls which suggested the pathological role of that pathway in OA [51]. The breakdown in the PERK pathway in OA chondrocytes with the accompanying COL2A1 expression reduction was identified leading to the extracellular matrix disintegration. The reduction in PERK in OA chondrocytes causes collagen II to be reduced and collagen I to be enhanced [29]. Loss of collagen II and increased collagen I expression are known to interfere with articular cartilage function [34]. Besides, Diazoxide prevents chondrocyte apoptosis and cartilage degeneration OA by PERK1/2 and ERK1/2 signaling pathways [52]. Sirtuin-1 (SIRT1), a NAD+-dependent deacetylase, is dysregulated in osteoarthritis. SIRT1 is a highly essential regulator of cartilage production and maintenance of cartilage homeostasis, it deacetylates PERK and weakens the PERK-eIF-2-CHOP-axis to enhance cartilage production and bone growth [53]. Under these results, weakened PERK can be an OA biomarker and increasing PERK can therapeutically develop OA [54]. Studies by Botter., Glasson. et al. have shown that Aggrecanase 2 deficient (ADAMTS5^-/-) mice can be protected against OA-caused cartilage injury [55]. After overexpression of IREla, we found that the expression of ADAMTS5 decreased significantly. This also provides support for UPR signaling molecules as a biological indicator of OA.

As a result of an imbalance between bone resorption by osteoclasts and bone synthesis by osteoblasts, osteoporosis (OP) is manifested. Estrogen has a protective effect on osteoblast apoptosis during the development of osteoporosis. 17-estradiol (17-E₂, a main estrogen) inhibits osteoblast apoptosis in response to ER stress via the Ras-ERK1/2-TFIII signaling pathway [56]. Progranulin, a moderator of estrogen/estrogen receptor α binding, also be proved can regulate bone homeostasis through the PERK/p-eIF2α signaling pathway [57]. Besides, Salubrinal's (eIF2α dephosphorylation inhibitor) improvement of senile osteoporosis is also achieved by activating eIF2α signaling [58].

Autophagy is essential in the development of cartilage. Osteoblast-specific autophagy related 7 (ATG7) conditional knockout (cKO) mice showed chondrocyte apoptosis was enhanced, whereas proliferation and differentiation declined. PERK-ATF4-CHOP axis may responsible for the symptoms caused by knockout ATG7, including decreased expression of Col10a1, MMP-13, and VEGFA and increased death of chondrocytes [59]. By influencing DDIT3/CHOP and MAPK8/JNK1/SMAD1/5/8, ATG7 deficiency can also inhibit osteoblast mineralization and increase osteoblast death [60]. In addition, autophagy-related bone formation is similarly influenced by the PERK/eIF2 arm of the UPR [61].

Comprehensive analysis of the results of two kinds of differential expression profile chips, overexpression of IRE1a and ATF6 can obviously regulate the mRNA levels in MMP-14/ADAM-9/FGFR2 pathway, which is closely associated with bone development. The relevance of MMP-2, -9, -13, -14 and -16 for bone formation was established by the mice knockout and human genetic disorders [62]. Some scholars have proposed a new approach involving skull development: MMP-14/ADAM-9/FGFR2 [62]. Fibroblast growth factor receptor (FGFR) is the main factor for proper skull intramembrane ossification. Crane facial deformities are more likely be induced by the depletion of FGFR2 gene. Metalloproteases can release FGFR2 (such as ADAM-9). In MMP-14 (^-/-), ADAM-9 is up-regulated, which leads to the cleavage form of FGFR2 accumulation [62]. The results of the genome-wide expression profiling chip we made found that IRE1a may be involved in this pathway to regulate skull development. After human chondrocytes overexpressed ATF6 or IRE1a, the expression of MMP-14 decreased, and the expression of ADAM-9 and FGFR2 increased

Figure 6), which is consistent with above-mentioned research results, indicating that UPR may be involve in this pathway partly to regulating bone development.

On basis of the result of XBP1S and ATF6 CTFM, we first selected CEBPB gene regulated by ATF6, and CTGF gene regulated by XBP1S to verify our results through the online software of prediction binding site, then analyzed the location of the gene of CEBPB and CTGF in chromosome (Figures 7A and 7B). Besides, the ATF6 binding sequence and the XBP1S binding sequence were analyzed by Jaspar database (http://jaspar.genereg.net/) on line (Figures 7C and 7D). The prediction results of the website also indicate that CEBPB or CTGF can be directly combined with ATF6 or XBP1S respectively, and the predicted sequence is GCATGGCGTGGCAT and AATGCCGCATCATT (Figures 7E and 7F).

Furthermore, XBP1 and ATF6 both can activate PERK-eIF2α-ATF4 signaling [38,63]. ATF4 up-regulates the expression of IRE1α mRNA in HeLa cells and mouse embryonic fibroblasts (MEF), thereby increasing the splicing rate of XBP1 [64], and a ChIP assay identified a XBP1S binding motif at the promoter region of the ATF4 gene [38]. Similarly, our microarray results also show that ATF4 can be directly combined with ATF6 and XBP1(Figure 8). All of these are proved that there exists a crosstalk between IRE1-, ATF6- and PERK-Dependent Signaling Pathways [65]. Three signal pathways of UPR obviously influenced and crosstalked each other in chondrocyte and bone metabolism.

PERK has garnered wide interest in its function in bone formation and remodeling. The knockout mouse has shown serious osteopenia which indicates that PERK is intimately associated with bone metabolism [66]. Mutations in the gene EIF2AK3 (encoding PERK) may be caused to induce a skeletal dysplasia and a delay in growth of Wolcott-Rallison syndrome. In addition, the EIF2AK3 haplotype gene is correlated with low bone mineral density in humans [67,68]. These all indicate the potential role of PERK in bone homeostasis. PERK inhibition or silencing have a significantly negatively effects on osteoclast differentiation, and accompanied by the down-regulation of osteoclast marker genes [66], such as MMP9 and Cathepsin K. EIF2AK3^-/- Mice bone tissue expression levels are low compared to wild-type mouse with transcripted levels of osteoblast marker (including alkaline phosphatase, col I, Ocn and bone salival protein). In contrast, in the case of the EIF2AK3 and wild type mice, there is no substantial difference in osteopontin transcription (an early osteoblast differential marker) that shows that PERK is necessary for osteoblast maturation or late differentiation [69].

In the formation of skeletons, VEGF plays a vital function. Osteoblast-derived VEGF controls osteoblastogenesis and bone marrow adipogenesis, thereby stimulating osteoclast differentiation. VEGF seems to manage the differentiation of MSCs via activating the RUNX2 transcription factor and suppressing PParα2 [70]. ATF4 also promotes bone angiogenesis through its activity in osteoclasts and bone breakdown, by encouraging VEGF release from the Bone Matrix [71]. Moreover, activating ATF6 and PERK helps via favorably regulating mTORC2-mediated phospha-phosphorylation of AKT on Ser473 to the survival impact of VEGF on endothelial cells (ECs) [72,73].

Additionally, the activation of UPR was associated to cytokine dysregulation, which promotes the synthesis of lL-23, IFN-β and lL-1α and can trigger IL-23/IL-17 axis [74].The IL-17 receptor family consists of IL-17RA to IL-17RE [75], binding the heterodimer IL-17RA/IL-17RC is the most popular IL-17 signaling system. IL-17RA/IL-17RC may mediate the osteogenesis effect and IL-17RB, D and E can mediate the osteo-inhibitory effect [76-78]. Our Chip-chip transcription factor Microarray data reveal furthermore that ATF6 can directly bind to CEBPB, MAPK1, SFRS1 or MAPK13 in the IL-17 signaling pathway to directly regulate it, and also is directly engaged in regulating the VEGF signaling pathway, which may be the link between ATF6 and PERK/ATF4 pathway.

The PERK-eIf2α signal is necessary for normal bone development [79]. Salubrinal inhibits the dephosphorylation of eIf2α and increases the differentiation of osteoblasts. Regulating endoplasmic reticulum stress through eIF2α and ATF4 may be a good anti-osteoporosis system [80,81]. In diabetic patients, decreased insulin and high blood sugar lead to low bone mineral density (BMD), which impairs bone formation. In addition, diabetes can induce the expression of CHOP in osteoblasts, leading to the progression of apoptosis [82,83]. The equilibrium of osteoblasts and osteoclasts is therefore disrupted and causes diabetic osteoporosis to develop. For osteoblast maturation, the IRE1α-XBP1 route is critical to the creation of bone and the resorption of the bone under pathological circumstances [8]. ER molecular chaperones such BiP and protein disulfide isomerase (PDI) have been decreased among osteoblasts in individuals with osteoporosis(OP) [84]. These research shows the relevance for the prevention and treatment of osteoporosis of ER stress.

Furthermore, studies in the fibroblast-like synoviocytes and macrophage of rheumatoid arthritis (RA) patients demonstrated a substantial increase in activation of the IRE1/XBP1 axis [85]. IRE1α activation increases the generation of pro-inflammatory cytokines in macrophages and neutrophils mediated by the Toll-like receptor (TLR) [86]. Recently, several scientists have also hypothesized ER stress to induce inflammation by stabilization cytokine mRNA through the IRE1 RIDD activity [87]. In RA patients, RNase activity is enhanced compared with healthy persons by IRE1 in peripheral mononuclear blood cell (PBMC). Transcriptional modifications to downstream IRE1 objectives (particularly XBP1S) may therefore provide new chances to improve present diagnostic indicators and RA therapy choices [88]. The ATF6 gene is up-regulated in the macrophages of RA patients [89]. Moreover, the cleaved ATF6 form can be boosted by TNF, while proteasome or autophagy suppression can inhibit the process [90]. ATF6 may be related to the acute phase reaction and is responsible for activating a series of inflammatory mediators [89,91,92]. In synovial tissues and macrophages of RA patients the expression levels of EIF2AK3 gene and phosphorylated eIF2α are increased [89,93]. PERK/eIF2α may stimulate the NF-κB pathway to promote RA pathological inflammation [94]. In the PBMC of RA patients is also raised the level of expression of GADD34, the downstream target of the PERK pathway and is associated to the production of pro-inflammatory cytokines [85,95].

Studies have regularly been carried out in patients with RA or OA concerning the involvement of IL-17 in the articular environment. IL-17A operates on cell chondrocytes through the activation of inducible synthase of nitrogen oxide (iNOS), COX2 or cartilage degradation-related secretions IL-6. IL-17A also inhibits the synthesis of proteoglycan, increases NO generation, and works in combination with TNF alpha to kill cartilage matrix [96-98]. Additional studies have shown that IL-17F increases cartilage breakdown by boosting collagenase expression (MMP-1 and MMP-13) and stromelysin-1 (MMP-3) as well as by reducing inhibitor expression (TIMP-2 and TIMP-4) or ECM component (type II collagen, aggrecan) [99,100]. Our work also shows that overexpression IRE1a can increase collagenase expression (MMP-2, MMP-3, MMP-14). Moreover, IL-17A stimulates human chondrocytes to produce a few chemokines (CXCL1, IL-8 and CCL2), which also causes MMPs or iNOS secretions. IL-17A suppresses human MSC's Chondrogenesis, inhibiting the action of protein-kinasis A (PKA) and SOX9 phosphorylation, under inflammatory circumstances [101-103].

In the past decade, research has demonstrated that UPR is a critical regulator for bone growth and homeostasis. The UPR has been used in the skeletal system to normalize ER hemostasis under ER-stress, but also actively regulates cell differentiation and maturation. We do not yet completely understand how UPR promotes the pathogenesis of bone disorders, including OA, OP, and RA, as well as the physiologic function of UPR in bone formation and homeostasis. However, it is connected to some kinds of bone disorders due to overactivated UPR signaling. The three UPR pathways therefore are the main objective of research in such illnesses. Obviously, our research analyzes and predicts those UPR signaling molecules that affect cell fate through regulating ER Stress, autophagy, ROS in skeleton development and associated diseases. These molecules have certain research prospects and value for the diagnosis and treatment of bone related disease. Deeper knowledge of the link between bone homeostasis and UPR therefore provides crucial biological insights on bone metabolism and forms the basis of bone disease therapeutic intervention.

This work was supported by the National Natural Science Foundation of China (No. 82272550, No.81871769).

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Conceptualization, Yiming Pan and Fengjin Guo; Data curation, Yiming Pan, and Fengjin Guo; Funding acquisition, Fengjin Guo; Investigation, Yiming Pan, and Fengjin Guo; Methodology, Yiming Pan, Kaiwen Liu, Yuanlan Ye, Mengtian Fan, Xiaoli Li, and Fengjin Guo; Software, Yiming Pan, Fengjin Guo; Supervision, Fengjin Guo; Writing – original draft, Yiming Pan and Fengjin Guo; Writing – review & editing, Yiming Pan and Fengjin Guo.

All authors approved the final manuscript prior to submission.

The authors declare that they have no conflicts of interest.

The authors declare no conflict of interest.