Pax5 Negatively Regulates Osteoclastogenesis through Downregulation of Blimp1

Paired box protein 5 (Pax5) is a crucial transcription factor responsible for B-cell lineage specification and commitment. In this study, we identified a negative role of Pax5 in osteoclastogenesis. The expression of Pax5 was time-dependently downregulated by receptor activator of nuclear factor kappa B (RANK) ligand (RANKL) stimulation in osteoclastogenesis. Osteoclast (OC) differentiation and bone resorption were inhibited (68.9% and 48% reductions, respectively) by forced expression of Pax5 in OC lineage cells. Pax5 led to the induction of antiosteoclastogenic factors through downregulation of B lymphocyte-induced maturation protein 1 (Blimp1). To examine the negative role of Pax5 in vivo, we generated Pax5 transgenic (Pax5Tg) mice expressing the human Pax5 transgene under the control of the tartrate-resistant acid phosphatase (TRAP) promoter, which is expressed mainly in OC lineage cells. OC differentiation and bone resorption were inhibited (54.2–76.9% and 24.0–26.2% reductions, respectively) in Pax5Tg mice, thereby contributing to the osteopetrotic-like bone phenotype characterized by increased bone mineral density (13.0–13.6% higher), trabecular bone volume fraction (32.5–38.1% higher), trabecular thickness (8.4–9.0% higher), and trabecular number (25.5–26.7% higher) and decreased trabecular spacing (9.3–10.4% lower) compared to wild-type control mice. Furthermore, the number of OCs was decreased (48.8–65.3% reduction) in Pax5Tg mice. These findings indicate that Pax5 plays a negative role in OC lineage specification and commitment through Blimp1 downregulation. Thus, our data suggest that the Pax5–Blimp1 axis is crucial for the regulation of RANKL-induced osteoclastogenesis.


Introduction
The strength and rigidity of bone are maintained by a proper homeostatic balance between bone-forming osteoblasts and bone-resorbing osteoclasts (OCs) [1,2]. As the only cells exhibiting bone resorption activity in vivo, OCs are clinically important; OCs lead to excess bone destruction in bone loss-associated disorders, such as rheumatoid arthritis, periodontitis, metastatic cancer, and osteoporosis [3,4].
OCs are derived from hematopoietic precursors in the bone marrow through the process of OC differentiation, also called osteoclastogenesis. Osteoclastogenesis is mediated mainly by the stimulation of receptor activator of nuclear factor kappa B (RANK) ligand (RANKL) and macrophage-colony stimulating factor (M-CSF). In particular, the crucial roles of RANKL, the primary cytokine involved in osteoclastogenesis, have been well elucidated [1,2,5,6]. Upon RANKL stimulation, the cytoplasmic tail of RANK recruits tumor necrosis factor receptor-associated factors, leading to the activation of downstream

Pax5 Expression Is Downregulated during RANKL-Induced Osteoclastogenesis
Pax5 −/− mice exhibit severe osteopenia with an increase in the OC number and acceleration of bone loss [14,15]. Thus, we first examined whether Pax5 expression is downregulated during osteoclastogenesis. To evaluate Pax5 expression, bone marrowderived macrophages (BMMs) were differentiated from bone marrow cells by culture with M-CSF (150 ng/mL) for 2 days and were then further differentiated into OCs by RANKL (100 ng/mL) stimulation in the presence of M-CSF (50 ng/mL) for 4 days. We examined whether Pax5 expression was abundantly maintained in BMMs on day 0 ( Figure 1). During osteoclastogenesis, Pax5 expression was rapidly reduced by RANKL stimulation after day 1 (the early stage of OC differentiation) and almost returned to the basal level by day 3 (the later maturation stage), while the expression of OC markers, such as TRAP, cathepsin K, NFATc1, Atp6V0d2, DC-STAMP, OSCAR, and Blimp1, was enhanced in a RANKL-dependent manner ( Figure 1). These results indicate that downregulation of Pax5 during osteoclastogenesis might be closely linked to the regulation of OC differentiation and OC marker expression.

Pax5 Negatively Regulates RANKL-Induced Osteoclastogenesis
We next determined whether the expression of Pax5 is involved in RANKL-induced osteoclastogenesis. To examine the role of Pax5 in osteoclastogenesis, Pax5 was overex-pressed in BMMs using a retroviral gene transfer system, and Pax5-transduced BMMs were then differentiated into OCs by RANKL stimulation. TRAP activity in OCs was significantly inhibited (37.7% reduction on day 4, p < 0.01) by forced expression of Pax5 in the TRAP solution assay, and the number of TRAP + multinucleated OCs (MNCs) was reduced (68.9% reduction on day 4, p < 0.01) in Pax5-overexpressing OCs (Figure 2A). Furthermore, the resorption area was decreased (48.0% reduction on day 4, p < 0.01) by Pax5 overexpression in OCs ( Figure 2B). Consistent with these results, the messenger RNA (mRNA) and protein levels of OC markers were significantly decreased in Pax5-overexpressing OCs during osteoclastogenesis ( Figure 2C,D). In analysis of deletion mutants, OC formation assays revealed that OC differentiation was partially restored upon loss of the conserved DNA-binding paired domain (PD) in Pax5, indicating that the DNA-binding activity of Pax5 is negatively involved in RANKL-induced osteoclastogenesis ( Figure 3). Taken together, these results indicate that Pax5 expression is negatively involved in the regulation of RANKL-induced osteoclastogenesis. Bone marrowderived macrophages (BMMs) were differentiated with macrophage-colony stimulating factor (M-CSF) (50 ng/mL) and receptor activator of nuclear factor kappa B (RANK) ligand (RANKL) (100 ng/mL) for 4 days. Total RNA was isolated from cultured cells in triplicate and analyzed by realtime PCR using specific primers for Pax5, tartrate-resistant acid phosphatase (TRAP), cathepsin K, nuclear factor of activated T cells c1 (NFATc1), Atp6V0d2, dendritic cell-specific transmembrane protein (DC-STAMP), osteoclast (OC)-associated receptor (OSCAR), and B lymphocyte-induced maturation protein1 (Blimp1). β-Actin was used as the control. All points and error bars represent the mean ± SD of triplicate real-time PCRs. * p < 0.05, ** p < 0.01, *** p < 0.001.

Pax5 Negatively Regulates RANKL-Induced Osteoclastogenesis
We next determined whether the expression of Pax5 is involved in RANKL-induced osteoclastogenesis. To examine the role of Pax5 in osteoclastogenesis, Pax5 was overexpressed in BMMs using a retroviral gene transfer system, and Pax5-transduced BMMs were then differentiated into OCs by RANKL stimulation. TRAP activity in OCs was significantly inhibited (37.7% reduction on day 4, p < 0.01) by forced expression of Pax5 in the TRAP solution assay, and the number of TRAP + multinucleated OCs (MNCs) was reduced (68.9% reduction on day 4, p < 0.01) in Pax5-overexpressing OCs (Figure 2A). Furthermore, the resorption area was decreased (48.0% reduction on day 4, p < 0.01) by Pax5 overexpression in OCs ( Figure 2B). Consistent with these results, the messenger RNA (mRNA) and protein levels of OC markers were significantly decreased in Pax5-overexpressing OCs during osteoclastogenesis ( Figure 2C,D). In analysis of deletion mutants, OC formation assays revealed that OC differentiation was partially restored upon loss of the conserved DNA-binding paired domain (PD) in Pax5, indicating that the DNA-binding activity of Pax5 is negatively involved in RANKL-induced osteoclastogenesis ( Figure  3). Taken together, these results indicate that Pax5 expression is negatively involved in the regulation of RANKL-induced osteoclastogenesis.  Pax5) is downregulated during osteoclastogenesis. Bone marrowderived macrophages (BMMs) were differentiated with macrophage-colony stimulating factor (M-CSF) (50 ng/mL) and receptor activator of nuclear factor kappa B (RANK) ligand (RANKL) (100 ng/mL) for 4 days. Total RNA was isolated from cultured cells in triplicate and analyzed by real-time PCR using specific primers for Pax5, tartrate-resistant acid phosphatase (TRAP), cathepsin K, nuclear factor of activated T cells c1 (NFATc1), Atp6V0d2, dendritic cell-specific transmembrane protein (DC-STAMP), osteoclast (OC)-associated receptor (OSCAR), and B lymphocyte-induced maturation protein1 (Blimp1). β-Actin was used as the control. All points and error bars represent the mean ± SD of triplicate real-time PCRs. * p < 0.05, ** p < 0.01, *** p < 0.001.

Pax5 Enhances the Expression of Antiosteoclastogenic Factors through Downregulation of Blimp1
Pax5 represses Blimp1 expression through direct binding to the Blimp1 promoter in human leukocytes [11]. Thus, we examined whether RANKL-induced Blimp1 promoter activity is negatively regulated by Pax5. To address this possibility, we constructed a luciferase reporter vector containing the murine Blimp1 promoter (Blimp1-Luc) harboring a putative Pax5 binding site in the region from nt −768 to nt +240 ( Figure 4A). In the Blimp1 luciferase reporter assay, Blimp1 promoter activity was inhibited (60.4-71.1% reduction, p < 0.01) in a dose-dependent manner by forced expression of Pax5 in RANKL-induced RAW264.7 cells ( Figure 4B). The expression of antiosteoclastogenic factors, such as Bcl6, MafB, and IRF8, has been reported to be inhibited by Blimp1 expression during osteoclastogenesis [1,5]. Thus, we next examined whether the expression of antiosteoclastogenic factors is induced by Pax5 expression during osteoclastogenesis. Consistent with these results shown in Figure 4B, the expression of antiosteoclastogenic factors was significantly increased (Bcl6: 4.8-fold (p < 0.01), MafB: 1.8-fold (p < 0.01), and IRF8: 2.1-fold (p < 0.05) on day 4) by forced expression of Pax5 during RANKL-induced osteoclastogenesis ( Figure 4C). These results indicate that Pax5 is crucial for the induction of antiosteoclastogenic factor expression through downregulation of Blimp1 in RANKL-induced osteoclastogenesis.

Pax5 Enhances the Expression of Antiosteoclastogenic Factors through Downregulation of Blimp1
Pax5 represses Blimp1 expression through direct binding to the Blimp1 promoter in human leukocytes [11]. Thus, we examined whether RANKL-induced Blimp1 promoter activity is negatively regulated by Pax5. To address this possibility, we constructed a luciferase reporter vector containing the murine Blimp1 promoter (Blimp1-Luc) harboring a putative Pax5 binding site in the region from nt −768 to nt +240 ( Figure 4A). In the Blimp1 luciferase reporter assay, Blimp1 promoter activity was inhibited (60.4-71.1% reduction, p < 0.01) in a dose-dependent manner by forced expression of Pax5 in RANKL-induced RAW264.7 cells ( Figure 4B). The expression of antiosteoclastogenic factors, such as Bcl6, MafB, and IRF8, has been reported to be inhibited by Blimp1 expression during osteoclastogenesis [1,5]. Thus, we next examined whether the expression of antiosteoclastogenic factors is induced by Pax5 expression during osteoclastogenesis. Consistent with these results shown in Figure 4B, the expression of antiosteoclastogenic factors was significantly increased (Bcl6: 4.8-fold (p < 0.01), MafB: 1.8-fold (p < 0.01), and IRF8: 2.1-fold (p < 0.05) on day 4) by forced expression of Pax5 during RANKL-induced osteoclastogenesis ( Figure  4C). These results indicate that Pax5 is crucial for the induction of antiosteoclastogenic factor expression through downregulation of Blimp1 in RANKL-induced osteoclastogenesis.

Osteoclastogenesis Is Reduced by Pax5 Transgene Expression in OC Lineage Cells
To identify the effects of Pax5 transgene expression by OC lineage-specific control on osteoclastogenesis, we generated two Pax5Tg mouse lines (Pax5Tg2 and Pax5Tg3) through microinjection of a transgenic vector containing a human Pax5 transgene expression cassette under the control of the TRAP promoter ( Figure 5A), which is expressed mainly in OC-specific lineage cells [16]. The levels of Pax5 transgene expression in OCs derived from BMMs of Pax5Tg2 and Pax5Tg3 mice were analyzed by semiquantitative RT-PCR ( Figure 5B). We next analyzed the effects of Pax5 transgene expression on RANKL-induced osteoclastogenesis. BMMs derived from Pax5Tg mice were differentiated into OCs by RANKL stimulation. In the TRAP solution assay, TRAP activity was reduced (Pax5 Tg2 : 23.6% reduction and Pax5 Tg3 : 25.8% reduction on day 4, p < 0.001) in OCs derived from Pax5Tg mice compared to those derived from wild-type mice ( Figure 5C, top right panel). We also observed that the size and number of TRAP + MNCs were decreased (Pax5 Tg2 : 54.2% reduction and Pax5 Tg3 : 76.9% reduction on day 4, p < 0.001) in OCs derived from Pax5Tg mice ( Figure 5C, left panel and bottom right panel). Consistent with these results, the resorption area of OCs derived from Pax5Tg mice was reduced (Pax5 Tg2 : 24.0% and Pax5 Tg3 : 26.2% reduction on day 4, p < 0.001) ( Figure 5D). Lastly, we analyzed the levels of Blimp1 and antiosteoclastogenic factors in OCs derived from Pax5Tg mice using real-time PCR ( Figure 5E). Similar to the results shown in Figures 2C and 4C, the mRNA level of Blimp1 was significantly decreased in OCs derived from Pax5Tg mice compared to those derived from wild-type mice, while the levels of antiosteoclastogenic factors, such as Bcl6, MafB, and IRF8, were increased in OCs derived from Pax5Tg mice ( Figure 5E). Taken together, these results indicate that Pax5 transgene expression via OC lineage-specific control is negatively involved in the regulation of RANKL-induced osteoclastogenesis.  1 μg), and pcDNA3.1HisLacZ (0.13 μg)). The transfected cells were stimulated with RANKL (100 ng/mL) for 2 days and subjected to a luciferase reporter assay. The expression of Pax5 was analyzed by immunoblotting with anti-Flag antibodies. β-Actin was used as the loading control. (C) Effect of Pax5 on the expression of antiosteoclastogenic factors. BMMs expressing the Flag-tagged Pax5 transgene were differentiated into OCs in triplicate by treatment with M-CSF (50 ng/mL) and RANKL (100 ng/mL) for 4 days. The mature OCs were analyzed by RT-PCR. The mRNA levels were normalized to those of β-actin. An empty control vector was used as the mock. All points and error bars represent the mean ± SD of triplicate real-time PCRs. * p < 0.05, ** p < 0.01, *** p < 0.001.

Osteoclastogenesis Is Reduced by Pax5 Transgene Expression in OC Lineage Cells
To identify the effects of Pax5 transgene expression by OC lineage-specific control on osteoclastogenesis, we generated two Pax5 Tg mouse lines (Pax5 Tg2 and Pax5 Tg3 ) through microinjection of a transgenic vector containing a human Pax5 transgene expression cassette under the control of the TRAP promoter ( Figure 5A), which is expressed mainly in OC-specific lineage cells [16]. The levels of Pax5 transgene expression in OCs derived from BMMs of Pax5 Tg2 and Pax5 Tg3 mice were analyzed by semiquantitative RT-PCR ( Figure  5B). We next analyzed the effects of Pax5 transgene expression on RANKL-induced osteoclastogenesis. BMMs derived from Pax5 Tg mice were differentiated into OCs by RANKL stimulation. In the TRAP solution assay, TRAP activity was reduced (Pax5 Tg2 : 23.6% reduction and Pax5 Tg3 : 25.8% reduction on day 4, p < 0.001) in OCs derived from Pax5 Tg mice compared to those derived from wild-type mice ( Figure 5C, top right panel). We also observed that the size and number of TRAP + MNCs were decreased (Pax5 Tg2 : 54.2% reduction and Pax5 Tg3 : 76.9% reduction on day 4, p < 0.001) in OCs derived from Pax5 Tg mice ( Figure 5C, left panel and bottom right panel). Consistent with these results, the resorption area of OCs derived from Pax5 Tg mice was reduced (Pax5 Tg2 : 24.0% and Pax5 Tg3 : 26.2% reduction on day 4, p < 0.001) ( Figure 5D). Lastly, we analyzed the levels of Blimp1 and antiosteoclastogenic factors in OCs derived from Pax5 Tg mice using real-time PCR (Figure  1 µg), and pcDNA3.1HisLacZ (0.13 µg)). The transfected cells were stimulated with RANKL (100 ng/mL) for 2 days and subjected to a luciferase reporter assay. The expression of Pax5 was analyzed by immunoblotting with anti-Flag antibodies. β-Actin was used as the loading control. (C) Effect of Pax5 on the expression of antiosteoclastogenic factors. BMMs expressing the Flag-tagged Pax5 transgene were differentiated into OCs in triplicate by treatment with M-CSF (50 ng/mL) and RANKL (100 ng/mL) for 4 days. The mature OCs were analyzed by RT-PCR. The mRNA levels were normalized to those of β-actin. An empty control vector was used as the mock. All points and error bars represent the mean ± SD of triplicate real-time PCRs. * p < 0.05, ** p < 0.01, *** p < 0.001.

Mice Expressing the Pax5 Transgene Show an Osteopetrotic-Like Bone Phenotype Caused by Reduced OC Formation
To examine the physiological role of Pax5 transgene expression in vivo, we next analyzed the bone phenotype of Pax5 Tg mice by microcomputed tomography (micro-CT) and bone histomorphometry. Three-dimensional images and bone parameters of the femoral trabecular and cortical bone in Pax5 Tg mice were measured and quantified using micro-CT ( Figure 6A). The bone mineral density (BMD) of the trabecular bone in the femur of Pax5 Tg mice was significantly increased (Pax5 Tg2 : 13.6% higher (p < 0.001) and Pax5 Tg3 : 13.0% higher (p < 0.001)) compared to that in the femur of wild-type mice ( Figure 6B, first panel). Consistent with these results, the trabecular bone volume fraction (BV/TV) was 32.5-38.1% higher (p < 0.01) in Pax5 Tg mice than in wild-type mice ( Figure 6B, second panel). In addition, the trabecular thickness (Tb.Th) and trabecular number (Tb.N) were 8.4-9.0% higher (p < 0.01 or p < 0.01) and 25.5-26.7% higher (p < 0.01 or p < 0.001), respectively, in Pax5 Tg mice than in wild-type mice. However, the trabecular spacing (Tb.Sp) was 9.3-10.4% lower (p < 0.05 or p < 0.01) in Pax5 Tg mice than in wild-type control mice ( Figure 6B, third through fifth panels). Next, we compared the number of TRAP + OCs in the bone sections of the trabecular region of femurs obtained from Pax5 Tg and wild-type mice using TRAP staining and hematoxylin counterstaining. Consistent with the results shown in Figure 6A,B, the number of TRAP + OCs was significantly decreased (Pax5 Tg2 : 48.8% reduction and Pax5 Tg3 : 65.3% reduction, p < 0.05 or p < 0.01) in Pax5 Tg mice compared to wild-type mice ( Figure 6C,D). Hence, these results indicate that transgenic mice expressing the Pax5 transgene under the control of the TRAP promoter show an osteopetrotic-like bone phenotype. 5E). Similar to the results shown in Figures 2C and 4C, the mRNA level of Blimp1 was significantly decreased in OCs derived from Pax5 Tg mice compared to those derived from wild-type mice, while the levels of antiosteoclastogenic factors, such as Bcl6, MafB, and IRF8, were increased in OCs derived from Pax5 Tg mice ( Figure 5E). Taken together, these results indicate that Pax5 transgene expression via OC lineage-specific control is negatively involved in the regulation of RANKL-induced osteoclastogenesis.

Discussion
Pax5 is a well-known master transcriptional regulator of B-cell lineage specification and commitment [14,17]. Pax5 has also been shown to be a potential regulator of myeloid cell fate [14,17]. Forced expression of Pax5 in hematopoietic stem cells is sufficient to restrict T-cell lineage development in the thymus but not inhibit the development of myeloid lineage cells in bone marrow [18]. Indeed, Pax5 −/− splenocytes exhibit the characteristics of uncommitted multipotent stem cells, which can differentiate into multiple lineages of myeloid cells, such as macrophages, dendritic cells, and granulocytes [14,19]. Interestingly, Pax5 may also control OC cell fate and lineage commitment. The number and activity of OCs are increased in Pax5 −/− mice [14,15]. Furthermore, OCs can be differentiated from highly enriched OC precursors with increased cell surface expression of the M-CSF receptor (c-Fms) derived from Pax5 −/− splenocytes [14,15]. These findings raise the question of how Pax5 regulates cell fate determination and cellular potency related to OC lineage commitment in osteoclastogenesis. Thus, this study focused particularly on determining the mechanism via which OC lineage-specific Pax5 expression controls OC differentiation.

Discussion
Pax5 is a well-known master transcriptional regulator of B-cell lineage specification and commitment [14,17]. Pax5 has also been shown to be a potential regulator of myeloid cell fate [14,17]. Forced expression of Pax5 in hematopoietic stem cells is sufficient to restrict T-cell lineage development in the thymus but not inhibit the development of myeloid lineage cells in bone marrow [18]. Indeed, Pax5 −/− splenocytes exhibit the characteristics of uncommitted multipotent stem cells, which can differentiate into multiple lineages of myeloid cells, such as macrophages, dendritic cells, and granulocytes [14,19]. Interestingly, Pax5 may also control OC cell fate and lineage commitment. The number and activity of OCs are increased in Pax5 −/− mice [14,15]. Furthermore, OCs can be differentiated from highly enriched OC precursors with increased cell surface expression of the M-CSF receptor (c-Fms) derived from Pax5 −/− splenocytes [14,15]. These findings raise the question of how Pax5 regulates cell fate determination and cellular potency related to OC lineage commitment in osteoclastogenesis. Thus, this study focused particularly on determining the mechanism via which OC lineage-specific Pax5 expression controls OC differentiation.
Pax5 contains a conserved DNA-binding PD, an octapeptide domain (OP), a partial homeodomain (HD), a transactivation domain (TA), and an inhibitory domain (ID) [7]. The PD is crucial for DNA binding to regulate the expression of downstream target genes [20,21]. This domain consists of an N-and a C-terminal subdomain, each of which binds independently to a distinct half-site in the recognition sequence [7,21]. Via the bipartite DNA binding of its PD subdomains, Pax5 binds to relatively degenerate DNA consensus sequences [22,23]. Thus, the binding of Pax5 to specific recognition sequences is potentially affected by interactions with other regulators that function as either transcriptional activators or repressors [14,21,24]. In our current study, we found that Pax5 acts as a negative regulator of osteoclastogenesis. Osteoclastogenesis was significantly inhibited by OC lineage-specific Pax5 expression (Figures 2 and 5). Interestingly, we also showed that the inhibitory effect of Pax5 on osteoclastogenesis was partially abolished by deletion of its PD, while the inhibitory effect of Pax5 on osteoclastogenesis was not affected by deletion of the OP, HD, TA, or ID ( Figure 3). Thus, according to our results, we postulate that the DNA-binding ability of Pax5 via the PD domain to its target consensus sequences is not absolutely required for its negative regulation of osteoclastogenesis. As DNA binding ambiguity in the PD subdomains of Pax5 has been revealed by previous structural studies [20][21][22][23], we cannot completely exclude the possibility that one or more other additional OC lineage-specific regulators that can interact with Pax5 are also involved in the negative regulation of osteoclastogenesis by Pax5. Pax5 has been reported to physically interact with several transcriptional regulators, such as groucho-related gene 4, Ets-1, PU.1, and runt-related transcription factor 1 [21,[25][26][27]. Intriguingly, among these regulators, the functional roles of PU.1 as a transcriptional activator and RUNX1 as a transcriptional repressor in osteoclastogenesis have been identified [28,29]. Hence, further exploration of the transcriptional regulatory pathway linked to the negative regulation of osteoclastogenesis by Pax5 would be interesting.
Blimp1, a zinc finger motif-containing transcriptional repressor encoded by the Prdm1 gene, is a key regulator of the terminal differentiation of plasma cells [17]. Blimp1 inhibits the expression of Pax5, which is crucial for B-cell lineage commitment and early B-cell development; in contrast, Pax5 represses Blimp1 expression through direct binding to exon 1 of the Blimp1 gene, thereby inhibiting the terminal differentiation of plasma cells [10][11][12]. Thus, Blimp1 and Pax5 are reciprocal and antagonistic regulators of B-cell development and plasma-cell differentiation [11]. Interestingly, we observed that Pax5 expression was timedependently downregulated by RANKL stimulation during osteoclastogenesis, whereas Blimp1 expression was enhanced, peaking on day 3 ( Figure 1). Moreover, studies have demonstrated that Blimp1 induced by RANK-RANKL signaling represses the expression of antiosteoclastogenic factors, such as Bcl6, IRF8, and MafB, during osteoclastogenesis [1,30]. We, therefore, postulate that Pax5 negatively regulates Blimp1 expression during osteoclastogenesis. Indeed, we observed that both the expression level and the promoter activity of Blimp1 were clearly downregulated by Pax5 during RANKL-induced osteoclastogenesis, thereby repressing the expression of antiosteoclastogenic factors, such as Bcl6, IRF8, and MafB (Figures 2 and 4). Consistent with previous studies on the crucial role of Pax5 in controlling cell fate decisions during B-cell development and plasma-cell differentiation, our findings suggest that the reciprocal and antagonistic regulation between Pax5 and Blimp1 ensures the specificity of the cell fate decision during RANKL-induced osteoclastogenesis. In addition, we interestingly observed increased trabecular thickness in Pax5 Tg mice compared to wild-type control mice ( Figure 6B), indicating that the bone formation by osteoblasts is possibly enhanced in Pax5 Tg mice. Although we did not test the bone formation rate and osteoblast function in Pax5 Tg mice, we postulate that Pax5 transgene expression in OC lineage cells is linked to osteoblast differentiation and bone formation. Thus, further studies will be required to elucidate the role of Pax5 transgene expression in OC lineage cells affecting osteoblast differentiation and function.
In conclusion, we identified a negative role of Pax5 in osteoclastogenesis. Forced expression of Pax5 in OC lineage cells in vitro and in vivo inhibited RANKL-induced osteoclastogenesis. The inhibitory role of Pax5 in RANKL-induced osteoclastogenesis is mediated by downregulation of Blimp1 expression, resulting in the activation of antiosteoclastogenic factors, such as Bcl6, IRF8, and MafB. Thus, our data suggest that the regulation β-Actin expression was used as an internal control. The specific primers used were as follows: Pax5 (sense), 5 -AGA GAA AAA TTA CCC GAC TCC TC-3 ; Pax5 (antisense),  5 -CAT CCC TCT TGC GTT TGT TGG TG-3 ; TRAP (sense), 5 -AAA TCA CTC TTC AAG  ACC AG-3 ; TRAP (antisense), 5 -TTA TTG AAC AGC AGT GAC AG-3

Micro-CT and Histological Analysis
Femurs from Pax5 Tg mice (6 week old males, n = 10 per group) were fixed with 10% formaldehyde, and trabecular morphometry of distal femurs was performed using micro-CT (SkyScan 1076, Bruker micro-CT, Kontich, Belgium) at 6.775 µm image pixel resolution with 50 kV, 200 µA, and a 0.5 mm aluminum filter. Image reconstruction was performed with the software interface (Nrecon 1.6.1.5), which was provided by the manufacturer of the scanner, with a smoothing of 1, ring artefact of 8, and beam hardening of 30%. Reconstructed images were realigned in Dataviewer software 1.5.2 (Bruker micro-CT, Kontich, Belgium). Bone phenotypes were analyzed by measuring the BMD, BV/TV, Tb.Th, Tb.N, and Tb.Sp in the region of interest using 50 slices approximately 0.4 mm away from the growth plate of the distal femur, as previously described [37]. For histological analysis, the fixed femurs were decalcified in 15% ethylenediaminetetraacetic acid solution at 4 • C for 21 days with mild agitation prior to embedding in paraffin, as previously described [32]. Paraffin sections (7 µm) were sliced and subjected to TRAP staining and hematoxylin counterstaining. The TRAP + OCs in the femur were counted by visualization under a microscope.

Statistical Analysis
Data are expressed as the mean ± SD of values from at least three independent experiments. Statistical analyses were performed using a two-tailed Student's t-test to analyze differences between two groups or one-way and two-way ANOVA, if there was one or more than two conditions, respectively, followed by Dunnett's multiple comparisons test to evaluate differences among at least three groups. A p-value < 0.05 (*) was considered statistically significant.