Attenuation of Bone Formation through a Decrease in Osteoblasts in Mutant Mice Lacking the GM2/GD2 Synthase Gene

The ganglioside GD1a has been reported to promote the differentiation of mesenchymal stem cells to osteoblasts in cell culture systems. However, the involvement of gangliosides, including GD1a, in bone formation in vivo remains unknown; therefore, we herein investigated their roles in GM2/GD2 synthase-knockout (GM2/GD2S KO) mice without GD1a. The femoral cancellous bone mass was analyzed using three-dimensional micro-computed tomography. A histomorphometric analysis of bone using hematoxylin and eosin (HE) and tartrate-resistant acid phosphatase was performed to examine bone formation and resorption, respectively. Calcein double labeling was also conducted to evaluate bone formation. Although no significant differences were observed in bone mass or resorption between GM2/GD2S KO mice and wild-type (WT) mice, analyses of the parameters of bone formation using HE staining and calcein double labeling revealed less bone formation in GM2/GD2S KO mice than in WT mice. These results suggest that gangliosides play roles in bone formation.


Introduction
Gangliosides, sialic-acid-containing glycosphingolipids, are expressed in various tissues and are considered to play important roles in the development and maintenance of organs and tissues as well as in the pathogenesis of disease [1][2][3][4][5][6][7]. Gangliosides have been shown to protect against inflammation and neurodegeneration through the suppression of complement systems [8]. They have also been implicated in the pathogenesis of Alzheimer's, Parkinson's, and Huntington diseases [5,9]. Furthermore, gangliosides have been reported to regulate the secretion of various hormones, such as insulin in pancreatic beta cells and leptin in adipose cells [10,11]. Gangliosides, including GD3 and GD2, are not only cancer-associated antigens but also promote malignant properties in melanoma, osteosarcoma, and breast cancer [12][13][14][15].
Previous studies demonstrated the involvement of gangliosides in bone metabolism [16][17][18][19]. The suppression of age-related bone resorption was detected in mutant mice lacking the GD3 synthase gene, which is essential for the production of b-series gangliosides [19]. Furthermore, the a-series ganglioside GD1a has been shown to play a role in the differentiation of mesenchymal stem cells (MSCs) to osteoblasts [16][17][18]. Although findings obtained using a cell culture system suggest that GD1a promotes bone formation, its contribution to bone formation in vivo remains unknown.
GM2/GD2 synthase is essential for the production of a-series (GM2, GM1, GD1a, and GT1a) and b-series (GD2, GD1b, GT1b, and GQ1b) gangliosides ( Figure 1). The GM2/GD2  Therefore, the present study examined the involvement of gangliosides, including GD1a, in bone formation by analyzing the femoral cancellous bone mass in wild-type (WT) and GM2/GD2S KO mice lacking GD1a, using three-dimensional micro-computed tomography (3D-µCT). A histomorphometric analysis of bone was performed using hematoxylin and eosin (HE) and tartrate-resistant acid phosphatase (TRAP) to assess bone formation and resorption, respectively. Calcein double labeling was also conducted to evaluate bone formation.
The results obtained herein demonstrate that bone formation was suppressed in GM2/GD2S KO mice through a decrease in the number of osteoblasts. The present study is the first to indicate the contribution of gangliosides to bone formation in vivo.      (Figure 4). The expression of GM3 in the osteoblasts from GM2/GD2S KO mice was higher than that in the osteoblasts from WT mice. Furthermore, we analyzed the change in the ganglioside composition in MC3T3-E1 cells by knockdown of the GM2/GD2 synthase gene (B4galnt1) using the method of siRNA. B4galnt1 was approximately reduced by 90% through its knockdown using siRNA ( Figure 5A). GD1a was rarely expressed by knockdown of GM2/GD2 synthase. On the other hand, GM3 expression was elevated by knockdown of the GM2/GD2 synthase gene ( Figure 5B). Collectively, knockout or knockdown of GM2/GD2 synthase revealed an elevation in GM3 instead of a reduction in GD1a.

The Change in the Ganglioside Composition in Osteoblasts by Knockout or GM2/GD2 Synthase
Osteoblasts from WT mice expressed GD1a. On the other hand, the GM2/GD2S KO mice did not express GD1a at all (Figure 4). The expressi osteoblasts from GM2/GD2S KO mice was higher than that in the oste mice. Furthermore, we analyzed the change in the ganglioside composit cells by knockdown of the GM2/GD2 synthase gene (B4galnt1) using siRNA. B4galnt1 was approximately reduced by 90% through its knockdo ( Figure 5A). GD1a was rarely expressed by knockdown of GM2/GD2 other hand, GM3 expression was elevated by knockdown of the GM2/GD ( Figure 5B). Collectively, knockout or knockdown of GM2/GD2 synthase vation in GM3 instead of a reduction in GD1a.

The Change in the Ganglioside Composition in Pre-Osteoclasts by Knockout of GM2/GD2 Synthase
We analyzed the change in the ganglioside composition between pre-osteoclasts from WT and GM2/GD2S KO mice ( Figure 6). Pre-osteoclasts from WT mice mainly expressed GM3, GD1a, and GD3 ( Figure 6A). GD1a was not expressed by knockdown of GM2/GD2 synthase at all ( Figure 6B). Similar to osteoblasts from GM2/GD2S KO mice, pre-osteoclasts from the KO mice also showed upregulation of GM3.

The Change in the Ganglioside Composition in Pre-Osteoclasts by Knockout of GM2/GD2 Synthase
We analyzed the change in the ganglioside composition between pre-osteoclasts from WT and GM2/GD2S KO mice ( Figure 6). Pre-osteoclasts from WT mice mainly expressed GM3, GD1a, and GD3 ( Figure 6A). GD1a was not expressed by knockdown of GM2/GD2 synthase at all ( Figure 6B). Similar to osteoblasts from GM2/GD2S KO mice, pre-osteoclasts from the KO mice also showed upregulation of GM3.

The Femoral Cancellous Bone Mass Did Not Differ between WT and GM2/GD2S KO Mice
To evaluate the involvement of GM2/GD2 synthase in bone volume, we analyzed bone volume/total volume (BV/TV) using µCT. No significant differences were observed in BV/TV, trabecular thickness (Tb.Th), trabecular number (Tb.N), or trabecular separation (Tb.Sp) between WT and GM2/GD2S KO mice (Figure 7).

The Femoral Cancellous Bone Mass Did Not Differ between WT and GM2/GD2S K
To evaluate the involvement of GM2/GD2 synthase in bone volume, we bone volume/total volume (BV/TV) using μCT. No significant differences were in BV/TV, trabecular thickness (Tb.Th), trabecular number (Tb.N), or trabecula tion (Tb.Sp) between WT and GM2/GD2S KO mice ( Figure 7).

Decreases in Parameters of Bone Formation by a Deficiency in GM2/GD2S Syntha
Calcein double labeling was performed to assess the bone formation rate. eral surface/bone surface (MS/BS), mineral apposition rate (MAR), and bone f rate (BFR) were significantly lower in GM2/GD2S KO mice than in WT mice ( MS/BS, MAR, and BFR were approximately reduced by 29%, 23%, and 45%, res These results show that GM2/GD2 synthase was involved in bone formation.

Decreases in Parameters of Bone Formation by a Deficiency in GM2/GD2S Synthase
Calcein double labeling was performed to assess the bone formation rate. The mineral surface/bone surface (MS/BS), mineral apposition rate (MAR), and bone formation rate (BFR) were significantly lower in GM2/GD2S KO mice than in WT mice ( Figure 8). MS/BS, MAR, and BFR were approximately reduced by 29%, 23%, and 45%, respectively. These results show that GM2/GD2 synthase was involved in bone formation.

Decreased Number of Osteoblasts in Femoral Cancellous Bone by a Deficiency in GM2/GD2S Synthase
The number of osteoblasts/bone surface (Ob.N/BS) and osteoblast surface/bone surface (Ob.S/BS) were significantly lower in GM2/GD2S KO mice than in WT mice ( Figure  9). Ob.N/BS and Ob.S/BS were approximately reduced by 28% and 23%, respectively. These results may show that GM2/GD2 synthase affected the proliferation of osteoblasts.

The Number of Osteoclasts Did Not Significantly Differ between WT and GM2/GD2S K Mice
The parameters of bone resorption, including the number of osteoclasts/bone su (Oc.N/BS) and osteoclast surface/bone surface (Oc.S/BS), were slightly lowe GM2/GD2S KO mice than in WT mice, but there were no significant differences (Fi 10). These results show that GM2/GD2 synthase did not affect bone resorption.

The Number of Osteoclasts Did Not Significantly Differ between WT and GM2/GD2S KO Mice
The parameters of bone resorption, including the number of osteoclasts/bone surface (Oc.N/BS) and osteoclast surface/bone surface (Oc.S/BS), were slightly lower in GM2/GD2S KO mice than in WT mice, but there were no significant differences ( Figure 10). These results show that GM2/GD2 synthase did not affect bone resorption.

Discussion
The present results demonstrate the negative impact of a GM2/GD2 synthase def ciency in mice on bone formation via decreases in the number of osteoblasts. MC3T3-E osteoblast cells express GM3 and GD1a, while RAW264.7 macrophage cells express GM3 GM1, GD1a, GD3, GD2, and GD1b. Although bone mass did not significantly differ be tween WT and GM2/GD2S KO mice, a histomorphometric analysis of bone in addition t calcein double labeling showed that bone formation was significantly lower in GM2/GD2 KO mice than in WT mice. On the other hand, no significant differences were observed i bone resorption parameters between WT and GM2/GD2S KO mice.
Glycosphingolipids have been reported to play important roles in the proliferatio and differentiation of osteoblasts [16][17][18]26,27]. Globo-series Gb4 is expressed in osteo blasts, and its deletion in mice negatively affected bone formation through a decrease i the number of osteoblasts [26]. Furthermore, the a-series ganglioside GD1a has bee shown to promote the differentiation of MSCs into osteoblasts by promoting the phos phorylation of epidermal growth factor receptors [16,17]. Furthermore, the inhibition o glucosylceramide synthase, which is important for the production of a

Discussion
The present results demonstrate the negative impact of a GM2/GD2 synthase deficiency in mice on bone formation via decreases in the number of osteoblasts. MC3T3-E1 osteoblast cells express GM3 and GD1a, while RAW264.7 macrophage cells express GM3, GM1, GD1a, GD3, GD2, and GD1b. Although bone mass did not significantly differ between WT and GM2/GD2S KO mice, a histomorphometric analysis of bone in addition to calcein double labeling showed that bone formation was significantly lower in GM2/GD2S KO mice than in WT mice. On the other hand, no significant differences were observed in bone resorption parameters between WT and GM2/GD2S KO mice.
Glycosphingolipids have been reported to play important roles in the proliferation and differentiation of osteoblasts [16][17][18]26,27]. Globo-series Gb4 is expressed in osteoblasts, and its deletion in mice negatively affected bone formation through a decrease in the number of osteoblasts [26]. Furthermore, the a-series ganglioside GD1a has been shown to promote the differentiation of MSCs into osteoblasts by promoting the phosphorylation of epidermal growth factor receptors [16,17]. Furthermore, the inhibition of glucosylceramide synthase, which is important for the production of all glycosphingolipids, reduced the expression levels of Gb4 and GD1a, thereby suppressing the proliferation of mouse osteoblasts [27].
As shown in Figure 2A, GM3 and GD1a were mainly expressed in osteoblasts. Furthermore, as shown in Figures 4 and 5, knockout or knockdown of GM2/GD2 synthase resulted in the deletion or suppression of the expression level of GD1a, leading to an increase in the expression level of GM3. A previous study reported that the treatment of human MSCs with GD1a promoted their differentiation to osteoblasts [16]. On the other hand, the treatment of MSCs with GM3 suppressed their differentiation to osteoblasts [16]. Collectively, reductions in the number of osteoblasts in GM2/GD2S KO mice may be caused by the change in the ganglioside composition such as the deletion of GD1a and the increase in GM3. In order to clarify that, it will be necessary to examine whether the treatment of GD1a or GM3 in osteoblasts affects proliferation. Since GD1a was deleted in GM2/GD2S KO mice in the present study, reductions in the number of osteoblasts by the change in the ganglioside composition may have had a negative impact on bone formation, as shown in Figures 8 and 9.
GD1a was expressed in immature osteoblast cells (MC3T3-E1 cells), and its expression level decreased in mature osteoblasts after the induction of osteoblastogenesis ( Figure 2). This result indicates the importance of GD1a in regulating the proliferation of immature osteoblasts. In addition to previous findings showing that GD1a regulates the differentiation of MSCs to osteoblasts [16][17][18], it has been suggested to play a role in the proliferation of immature osteoblasts.
Gangliosides are known to be enriched in the glycolipid-enriched microdomain (GEM)/rafts on the plasma membrane [28] and promote or suppress the cell signaling through the GEM/rafts [7,[29][30][31][32]. GD1a and GM3 may interact with the growth factor receptors that regulate the cell proliferation in GEM/rafts, resulting in the enhancement or attenuation of the proliferation in osteoblasts. Here, we propose the mechanism by which GD1a and GM3 regulate osteoblast proliferation in GEM/rafts ( Figure 11). Int. J. Mol. Sci. 2022, 23, 9044 14 glycosphingolipids, reduced the expression levels of Gb4 and GD1a, thereby suppres the proliferation of mouse osteoblasts [27]. As shown in Figure 2A, GM3 and GD1a were mainly expressed in osteoblasts. thermore, as shown in Figures 4 and 5, knockout or knockdown of GM2/GD2 synt resulted in the deletion or suppression of the expression level of GD1a, leading to an crease in the expression level of GM3. A previous study reported that the treatmen human MSCs with GD1a promoted their differentiation to osteoblasts [16]. On the o hand, the treatment of MSCs with GM3 suppressed their differentiation to osteoblasts Collectively, reductions in the number of osteoblasts in GM2/GD2S KO mice ma caused by the change in the ganglioside composition such as the deletion of GD1a and increase in GM3. In order to clarify that, it will be necessary to examine whether the tr ment of GD1a or GM3 in osteoblasts affects proliferation. Since GD1a was delete GM2/GD2S KO mice in the present study, reductions in the number of osteoblasts by change in the ganglioside composition may have had a negative impact on bone mation, as shown in Figures 8 and 9.
GD1a was expressed in immature osteoblast cells (MC3T3-E1 cells), and its exp sion level decreased in mature osteoblasts after the induction of osteoblastogenesis ( ure 2). This result indicates the importance of GD1a in regulating the proliferation of mature osteoblasts. In addition to previous findings showing that GD1a regulates the ferentiation of MSCs to osteoblasts [16][17][18], it has been suggested to play a role in the liferation of immature osteoblasts.
Gangliosides are known to be enriched in the glycolipid-enriched microdom (GEM)/rafts on the plasma membrane [28] and promote or suppress the cell signa through the GEM/rafts [7,[29][30][31][32]. GD1a and GM3 may interact with the growth facto ceptors that regulate the cell proliferation in GEM/rafts, resulting in the enhancemen attenuation of the proliferation in osteoblasts. Here, we propose the mechanism by w GD1a and GM3 regulate osteoblast proliferation in GEM/rafts ( Figure 11).  The mechanisms by which gangliosides contribute to bone formation were not elucidated in detail in the present study. Nevertheless, the results obtained clearly demonstrate, for the first time, that gangliosides regulate bone formation in vivo.

Mice
GM2/GD2S KO mice were generated as previously reported [20]. WT and GM2/GD2S KO mice were mated to yield heterozygotes, which were then mated, and the genotypes of the offspring were screened according to a previously reported method [20]. Mice were housed in cages with free access to water and food under a 12 h light/dark cycle. Protocols in the present study received approval from the Aichi Gakuin University Animal Research Committee (approval number AGUD312; 29 October 2015, AGUD463; 28 May 2020) (Nagoya, Japan) and were performed in accordance with the Guidelines for Animal Experiments of Aichi Gakuin University. Seventeen WT mice and sixteen GM2/GD2S KO mice were used for 3D-µCT, HE, and TRAP. Nine WT mice and six GM2/GD2S KO mice were used for calcein double labeling. Four WT mice and five GM2/GD2S KO mice were used for in vitro experiments.

Culture of Primary Pre-Osteoclasts
Mouse bone marrow cells from WT and GM2/GD2S KO mice were plated in 150 mm dishes and cultured with 10 ng/mL macrophage colony-stimulating factor (M-CSF, PeproTech, Inc., Rocky Hill, NJ, USA) for 3 days. The surface-attached cells were used for a flow cytometric analysis [36].

Differentiation to Mature Osteoblasts
MC3T3-E1 cells were grown to confluence in 50 µg/mL ascorbic acid and 5 mM β-glycerophosphate. The culture medium was replaced every second day. Cells were subjected to a flow cytometric analysis after 21 days in culture.

Induction of Osteoclastogenesis
RAW264.7 cells were plated in 150 mm culture dishes and treated with 50 ng/mL RANKL (PeproTech, Inc.). After 2 days, cells were subjected to a flow cytometric analysis.

Quantitative Real-Time PCR (qPCR)
An RNeasy Plus mini kit (Qiagen, Germantown, MD, USA) was used to extract total RNA. A high-capacity cDNA reverse transcription kit (Applied Biosystems, Carlsbad, CA, USA) was employed for reverse transcription, and TakaRa Thermal Cycler Dice Real Time System III with THUNDERBIRD SYBR qPCR mix kits (TOYOBO, Osaka, Japan) for qPCR. Total RNA was used as the template for reverse transcription under the following cycling conditions: at 25 • C for 10 min, 37 • C for 120 min, and 85 • C for 5 min. PCR cycling parameters were an initial hold at 95 • C for 10 min, followed by 40 cycles at 95 • C for 15 s, and 60 • C for 1 min. The B4galnt1 (GM2/GD2 synthase gene) (forward: 5 -GCTGGGTCTCCTGTACTCCA-3 , and backward: 5 -TCCTCCCTTGGATTCACAAC-3 ) mRNA level was then measured with Gapdh (forward: 5 -TGCACCACCAACTGCTTAG-3 , and backward: 5 -GGATGCAGGGATGATGTTC-3 ) as the internal control.
4.14. Calcein Double Labeling to Measure the Bone Formation Rate MS/BS, MAR, and BFR were assessed using calcein double labeling. Mice were intraperitoneally administered calcein (8 µg/g; Sigma, St. Louis, MO, USA) 3 days and 1 day before being euthanized. Following their fixation in 4% paraformaldehyde (PFA), femurs were cut to obtain undecalcified 5 µm-thick sections. A bone fraction with a rectangular area of 0.34 mm 2 (0.5 × 0.67 mm), with the closest and furthest edges being 0.5 and 1.0 mm medial to the growth plate, respectively [26], was obtained from metaphyseal cancellous bone in the femur and used to measure MS/BS, MAR, and BFR.

TRAP Staining
TRAP staining was conducted according to a previously reported method [42]. Briefly, slides of femur samples were subjected to TRAP staining at 37 • C for 60 min using sodium acetate buffer (0.1 M, pH 5.0) with naphthol AS-MX phosphate, Fast Red Violet LB Salt, and MnCl 2 in the presence of sodium tartrate.

Histomorphometric Analysis of Bone
Femur samples fixed in 4% PFA were decalcified in 10% ethylenediaminetetraacetic acid (EDTA) for three weeks and embedded in paraffin. Samples were then sectioned at a thickness of 5 µm and subjected to HE and TRAP staining. HE-stained slides were used to assess Ob.N/BS and Ob.S/BS according to a previously reported method [19]. TRAPstained slides were used to assess Oc.N/BS and Oc.S/BS [19]. Parameters were measured within an area of 0.8 mm 2 (1.0 × 0.8 mm), with the closest and furthest edges being 2.0 and 3.0 mm medial to the growth plate in the proximal ends of the femur, respectively [43].

Statistical Analysis
Results are shown as the mean ± S.D. The significance of differences was examined using Student's t-test at p < 0.05, with single and double asterisks indicating p < 0.05 and p < 0.01, respectively.

Conclusions
The present study showed that a GM2/GD2 synthase deficiency in mice had a negative impact on bone formation by decreasing the number of osteoblasts.

Institutional Review Board Statement:
The animal study protocol (AGUD312 and 463) was approved by the Institutional Animal Care and Use Committee of Aichi Gakuin University.

Informed Consent Statement: Not applicable.
Data Availability Statement: The data presented in this study will be available on reasonable request to the corresponding author.