The balance between osteoclast and osteoblast activity can dictate pathogenesis of bone diseases. Osteoclasts are the exclusive bone resorbing cells involved in bone remodeling and resorption, and perform opposing roles to osteoblasts to affect skeletal mass and structure (Teitelbaum and Ross, 2003). Augmented osteoclast activity can lead to bone loss in osteoporosis, inflammatory arthritis and tumor invasion in bone, while osteopetrosis is characterized by increased bone mass and results from attenuation in osteoclast function/recruitment or arrested osteoclastogenesis (Teitelbaum, 2007). Abnormalities in osteoclasts are considered the primary cause of many bone diseases including osteoporosis, the most common bone disorder, and Paget disease of the bone (PDB) in which accelerated osteoclastic bone resorption leads to osteolytic or osteosclerotic bone lesions (Shaker, 2009).

Osteoclasts are multi-nucleated cells belonging to myeloid lineage, and generated via the fusion of monocytes/macrophage precursor cells (Roodman, 1999). Importantly, signaling via tumor necrosis factor (TNF)-family cytokine, receptor activator of nuclear factor (NF)-kappaB ligand (RANKL) and macrophage colony-stimulating factor (M-CSF) are identified as the primary pathways associated with osteoclast differentiation (Asagiri and Takayanagi, 2007). Osteoclast precursors express receptor activator of nuclear factor kappa B (RANK), while its ligand (RANKL) is expressed on osteoblasts/stromal cell precursors, and inhibited by the decoy receptor osteoprotegerin (OPG) (Yasuda et al., 1998). Moreover, various genes including TNF receptor-associated factor (TRAF) 6, NFKB1, FOS, nuclear factor of activated T cells 1 (NFATC1), and dendritic cell-specific transmembrane protein (DC-STAMP) have been associated with osteoclastogenesis (Asagiri and Takayanagi, 2007), while the tartrate-resistant acid phosphatase (TRAcP) and cathepsin K (CTSK), released by osteoclasts during resorption are identified as specific osteoclast markers (Boyle et al., 2003; Kirstein et al., 2006). Osteoclast functionality in bone resorption is dependent on αvβ3 integrin-mediated induction, binding and polarization, while deficiencies in acidified bone resorptive components disrupt regulated bone remodeling, leading to osteopetrosis. Amplified stimulation of osteoclastogenesis primarily via NFKB signaling can lead to increased osteolysis in osteoporosis (Novack and Teitelbaum, 2008).

Genome-wide screening and the in vitro induction of osteoclastogenesis using osteoclast precursor cells and soluble mediators has enabled the identification of molecular mechanisms governing osteoclast differentiation (Asagiri and Takayanagi, 2007). Sequencing techniques and microarray analyses have also contributed to disclosing important genes related to osteoclast differentiation (Cappellen et al., 2002; Day et al., 2004; Nomiyama et al., 2005; Garlet et al., 2008; Coudert et al., 2014; Kim and Lee, 2014; Purdue et al., 2014; Madel et al., 2020). In addition, recent reports have highlighted the roles of long noncoding RNAs (LncRNAs), which have emerged as vital regulators of gene expression, epigenetics and protein translation (Fatica and Bozzoni, 2014), in osteoclast differentiation (Fei et al., 2020; Liu et al., 2020; Yang et al., 2021). However, available data have primarily focused on differences between fully differentiated/generated osteoclasts compared to precursors or focused predominantly on targeted sets of genes, previously linked with osteoclast differentiation/activity. Therefore, a comprehensive insight into the differentially regulated genes and their associated pathways during osteoclast differentiation is warranted.

In this study, we performed sequential and comprehensive transcriptomic profiling of murine osteoclasts generated in vitro by comparing the total RNA transcriptomes of bone marrow derived precursor cells with differentiated osteoclasts over different time points. Importantly, we focused on novel genes and LncRNAs, which have not been previously associated with osteoclast differentiation. These progressive analyses provide important insights into the differentially regulated genes and their associated pathways during osteoclast differentiation. Identification of novel gene signatures and LncRNAs associated with osteoclastogenesis can serve as molecular biomarkers for osteoclast differentiation or explored for therapeutic benefits, while the comprehensive transcriptome of osteoclast differentiation provided herein can serve as an invaluable resource for deciphering gene dysregulation in diseases related to osteoclast differentiation/activity.

The molecular mechanisms behind osteoclast differentiation have been extensively explored. For instance, Capellen et al., used microarray analyses to decipher changes in gene expression during osteoclast differentiation and showed synergy between MCSF and RANKL-induced gene expression (Cappellen et al., 2002). More recently, several potential genetic regulators of osteoclast differentiation have been identified. Nishikawa et al., reported the epigenetic control of osteoclast differentiation via DNA (cytosine-5)-methyltransferase 3A (Dnmt3a) (Nishikawa et al., 2015) and Laha et al., reported regulation of osteoclastogenesis by KLF2 (kruppel-like factor 2 [lung]) via reduction in autophagic cells (Laha et al., 2019). In addition, low-density lipoprotein receptor-related protein1 (LRP1), and COMMD1 were reported as critical regulators of osteoclastogenesis, osteoclast activity and bone mass (Murata et al., 2017; Bartelt et al., 2018), while leucine-rich repeat-containing G-protein-coupled receptor 4 (LGR4) has been identified as another putative receptor for RANKL (Luo et al., 2016).

While these studies revealed important osteoclast-related targets, a comprehensive list of genes provided in our study is of paramount significance in understanding the dynamic changes in gene expression during osteoclast differentiation. Irrefutably, the preexistent available literature related to osteoclast identification and functionality enabled us to confirm the generation of osteoclasts in vitro. However, our data revealed the dynamic changes in osteoclast transcriptome during osteoclastogenesis, showing diminishing of genes related to immune processes with concurrent upregulation of genes associated with anatomical morphogenesis and development. Moreover, these genes encoded for important mediators of cell differentiation, structural morphogenesis and bone and tissue remodeling.

The potentially novel genes related to osteoclast differentiation disclosed herein include PH Domain Containing Endocytic Trafficking Adaptor 1 (Pheta1), Hydroxyacylglutathione hydrolase (Hagh), Solute Carrier Family 18 Member A1 (Slc18a1), Glutamine-Fructose-6-Phosphate transaminase 1 (Gfpt1), Coenzyme Q8A (Coq8a), Solute Carrier Family 14 Member 2 (Slc14a2), Whirlin (Whrn) and C-Type Lectin Domain Family four Member A (Clec4a4), among others. The proteins encoded by these genes are primarily associated with biochemical or signaling pathways at the cellular level and defects in expression may lead to certain disorders. For instance, anomalies in PHETA1/2 have been associated with abnormal bone formation resulting in craniofacial defects (Ates et al., 2020), HAGH has been associated with skin, bone and joint infections in Yaws disease (Cheng et al., 2018) and mutations in GFPT1 have been associated with muscle weakness in congenital myasthenic syndrome (Helman et al., 2019). In addition, histone-related genes H2B Clustered Histone 21 (H2bc21) and H2bc4 were also significantly upregulated during osteoclast differentiation. Epigenetic regulation of bone development and remodeling has been extensively reported (Yi et al., 2019). The upregulation of histone-related genes in osteoclast differentiation indicates chromatin remodeling as a critical step in osteoclast generation.

One of the pivotal novel findings of this study is the upregulation and confirmatory evidence of the potential role of Nol4 in mature osteoclasts, which has not been reported previously. Investigating bone-related phenotypes of some of the potentially novel genes we recorded showed that selective genes including Pheta1, Hagh, Plasminogen Activator urokinase (Plau), Sell, Retnlg and Cp are related to bone phenotypes in the IMPC database (Dickinson et al., 2016; International Mouse Phenotyping Consortium, 2021). However, only Nol4 and Zfp831 showed statistically significant associations with BMD or tibia length, respectively. Importantly, the knockout of Nol4 led to increase in BMD compared to wild type mice, thereby providing evidence for its role in bone biology. Our differential gene analyses showed upregulation of Nol4 expression in differentiated osteoclasts, which supports their contribution in affecting BMD in bone resorption and presents Nol4 as a key gene related to osteoclast functionality. Of note, downregulation of Zfp831 in mature osteoclasts in our data and association of Zfp831 knockout with reduced bone (tibia) length in IMPC datasets suggest its potential roles in osteoclast-mediated effects on bone growth and remodeling. However, further investigations are necessitated to fully explore the effects of these genes on bone biology. The protein encoded by Nol4 is associated with RNA binding and has been identified as a cancer/testis antigen in humans, recently presented as a candidate target in small cell lung cancer (Kim et al., 2021). However, the role of Nol4 in bone-related pathologies remains to be fully explored. Similarly, the protein encoded by Zfb831 is associated with nucleic acid binding and was downregulated in tumor-infiltrating cytotoxic T lymphocytes (CTLs) compared to CTLs from spleen in a murine transplantable tumor model (Waugh et al., 2016), but its effects in bone-related diseases are not explored.

Accumulating evidences have highlighted the roles of LncRNAs in controlling gene expression to affect cell differentiation and development (Fatica and Bozzoni, 2014). Importantly, in relation with osteoclasts. Liu et al., reported dysregulation in 1,117 LncRNAs in human osteoclasts differentiated from CD14⁺ monocytes in vitro (Liu et al., 2020). Fei et al., also reported 204 differentially expressed LncRNAs in male osteoporosis (Fei et al., 2020), while Yang et al., identified 46 differentially expressed LncRNAs between osteoarthritis and osteolysis following total hip arthroplasty, and reported potential roles of specific LncRNA-mRNA pairs in regulating CD8A, CD8B and osteoclastogenesis in these patients (Yang et al., 2021). Moreover, Bu et al., reported the role of LncRNA TSIX in promoting osteoblast apoptosis in particle induced osteolysis (PIO), evident from decreased BMD following implantation, via modulation of the microRNA miR-30a-5p (Bu et al., 2018). These reports rationalize the significance of LncRNAs in osteoclast generation and activity. We performed comprehensive total RNA-based analyses to disclose the differentially regulated LncRNAs during osteoclast differentiation and exclusively reported LncRNA not previously reported or associated with osteoclasts. Importantly, we have highlighted the top 20 novel LncRNAs upregulated during murine osteoclast generation.

Osteoporosis is characterized by attenuated bone strength due to deterioration in bone microarchitecture and reduction in bone mass (Sozen et al., 2017). While various risk factors have been identified for predisposition to osteoporosis, endocrine diseases have also been liked with osteoporosis (Sozen et al., 2017). For instance, in diabetes-associated osteoporosis altered bone metabolism also leads to changes in BMD and has been linked with high osteoclast activity (Kemink et al., 2000; Reni et al., 2016). Importantly, insulin is identified as an essential mediator in osteoclast energy metabolism. Kim et al., investigated the effects on gene expression in insulin-induced osteoclast differentiation and reported that insulin conducts similar roles as RANKL in osteoclast activity (Kim and Lee, 2014). These reports reflect the significance of energy and metabolism in osteoclast activity, which leads to osteoporosis. We found that numerous genes encoding important metabolism/energy-related mediators were significantly increased during early osteoclast generation.

Osteoclast progenitors are essentially immune cells, due to their origins from monocytes/macrophage precursors and may present as innate immune cells within the bone (Wu et al., 2008). Jacquin et al., showed that osteoclast progenitor populations in murine BM comprise CD45R⁻ CD11b^−/low populations (Jacquin et al., 2006), while Jacome-Galarza et al., reported that B220⁻CD3⁻CD11b^−/lowCD115⁺CD117^high mouse BM cells possess high osteoclastic potential (Jacome-Galarza et al., 2013). Our data showed that osteoclast differentiation deviates their associations with other immunomodulatory cells, evident from downregulation of genes related to immune/inflammatory response and response to stress/stimulus in differentiated osteoclasts. Moreover, we found that the differences between the transcriptomes of fully-committed/differentiated and early-differentiated osteoclasts were predominantly limited to sustained downregulation of genes related to immune cell characteristics. These differentially expressed genes primarily encoded for cytokine production, leukocyte activation and immune response-related functions. Importantly, upregulation of genes related to cell cycle and DNA replication in fully differentiated osteoclasts exhibited the cellular expansion of committed osteoclasts. Of note, induction of bone resorption in inflammatory disorders and immune-related disorders results from the plethora of secreted cytokines, which induce osteoclastogenesis (Adamopoulos, 2018).

Overall, this study disclosed genes and their associated pathways during osteoclast differentiation which can be explored further in successive studies. Importantly, we highlighted the potentially novel genes and LncRNAs in relation to osteoclast differentiation and activity. However, functional studies are necessitated to disclose the roles of specific targets. The comprehensive list of differentially regulated genes provided herein can serve as an expedient tool in osteoclast-related research.