Deficiency of GD3 Synthase in Mice Resulting in the Attenuation of Bone Loss with Aging

Gangliosides are widely expressed in almost all tissues and cells and are also considered to be essential in the development and maintenance of various organs and tissues. However, little is known about their roles in bone metabolism. In this study, we investigated the effects of genetic deletion of ganglioside D3 (GD3) synthase, which is responsible for the generation of all b-series gangliosides, on bone metabolism. Although b-series gangliosides were not expressed in osteoblasts, these gangliosides were expressed in pre-osteoclasts. However, the expression of these gangliosides was decreased after induction of osteoclastogenesis by receptor activator of nuclear factor kappa-B ligand (RANKL). Three-dimensional micro-computed tomography (3D-μCT) analysis revealed that femoral cancellous bone mass in GD3 synthase-knockout (GD3S KO) mice was higher than that in wild type (WT) mice at the age of 40 weeks, although there were no differences in that between GD3S KO and WT mice at 15 weeks old. Whereas bone formation parameters (osteoblast numbers/bone surface and osteoblast surface/bone surface) in GD3S KO mice did not differ from WT mice, bone resorption parameters (osteoclast numbers/bone surface and osteoclast surface/bone surface) in GD3S KO mice became significantly lower than those in WT mice at 40 weeks of age. Collectively, this study demonstrates that deletion of GD3 synthase attenuates bone loss that emerges with aging.


Introduction
Gangliosides, which are sialic-acid-containing glycosphingolipids, are highly expressed in the nervous tissues of vertebrates [1] and have been considered as being involved in the regulation of nervous systems [2][3][4][5]. Recent studies of mutant mice lacking various gangliosides have revealed that gangliosides are essential in the maintenance of the integrity of nervous tissues and spermatogenesis, as well as the homeostasis of organisms [6][7][8]. GM3-only mice (double knockout mice lacking GM2/GD2 synthase and GD3 synthase genes) have exhibited severe neurodegeneration caused by excessive complement activation in the brain and spinal cord [9,10]. Mice lacking GM2/GD2 synthase gene have shown that gangliosides also play important roles in the transport of testosterone to the seminiferous tubules and bloodstream from Leydig cells [6].
In particular, b-series gangliosides, including GD3, GD2, GD1b, GT1b, and GQ1b, all of which contain α-2,8 sialic acid, have been considered as playing crucial roles in the maintenance of the sensory nervous system, neuroregeneration of hypoglossal nerves after injury, and secretion of leptin in adipose tissues [7,11,12]. On the other hand, these gangliosides have been identified as cancer-associated antigens, and are known to enhance tumor phenotypes, such as cell proliferation, and invasion in melanomas, osteosarcomas, and small cell lung cancers [13][14][15].
Some studies have reported that gangliosides are expressed in bone marrow mesenchymal stem cells (MSCs) and osteoblasts [16][17][18][19][20]. It has been documented that a-series ganglioside GD1a contributes to the regulation of osteoblast differentiation in MSCs [16][17][18]. However, all these studies were conducted only in cultured cell systems and there have been no reports on the involvement of gangliosides in bone metabolism in animals.
In this study we examine expression levels of b-series gangliosides (GD3, GD2, GD1b, and GT1b) in MC3T3 E1 mouse osteoblast-like cells, RAW264.7 pre-osteoclasts, and primary cultured pre-osteoclasts derived from mouse bone marrow cells using flow cytometry. Furthermore, in order to examine the involvement of b-series gangliosides in bone metabolism in vivo, we analyzed femoral cancellous bone mass in wild type and GD3 synthase-knockout mice using three-dimensional micro-computed tomography (3D-µCT). We also employed bone histomorphometric analyses using tartrate-resistant acid phosphatase (TRAP) and hematoxylin and eosin (HE) staining.
We report here that deletion of GD3 synthase in mice results in the suppression of bone resorption. To our knowledge, this report suggests, for the first time, that gangliosides are involved in bone metabolism in animals.

No Expression of b-Series Gangliosides and GD3 Synthase Gene in Osteoblasts
Expression levels of b-series gangliosides (GD3, GD2, GD1b, and GT1b) in osteoblasts were analyzed by flow cytometry (Figure 1). These gangliosides were not detected in MC3T3 E1 cells before (Day 0) and after (Day 7) induction to mature osteoblasts ( Figure 2A). Consistent with these results of flow cytometry, the GD3 synthase gene (St8sia1) was at the limit of detection in MC3T3 E1 cells before (Day 0) and after induction (Day 7, 14, and 21) to mature osteoblasts ( Figure 2B). Both RAW264.7 cells and primary cultured pre-osteoclasts expressed GD3, GD2, and GD1b, and GT1b was below the limit of detection in both cells ( Figure 3A,B). The expression of these gangliosides except for GD1b in primary cultured pre-osteoclasts was significantly decreased after induction of osteoclastogenesis by RANKL ( Figure 3A,B). Consistent with these results was that expression levels of GD3 synthase gene (St8sia1) were downregulated by RANKL administration ( Figure 3C).

Attenuation of Loss of Femoral Cancellous Bone Mass with Aging in GD3 Synthase-Knockout (GD3S KO) Mice
No significant difference in the parameters of femoral cancellous bone mass was found at 15 weeks old between wild type (WT) and GD3S KO mice ( Figure 4A). However, bone volume/total volume (BV/TV) in GD3S KO mice was significantly higher than that in WT mice at 40 weeks old ( Figure 4B,C). Although trabecular thickness (Tb.Th) was not significantly different between 40-week-old WT mice and GD3S KO mice (p = 0.0503), Tb.Th in GD3S KO mice was higher than that in WT mice. Trabecular number (Tb.N) and trabecular separation (Tb.Sp) were not significantly different at 40 weeks old between WT and GD3S KO mice.

No Differences in Osteoblast Number in Femoral Cancellous Bone Between WT and GD3S KO Mice
Bone formation parameters such as osteoblast numbers/bone surface (Ob.N/BS) and osteoblast surface/bone surface (Ob.S/BS) were almost equivalent between 15-and 40-week-old WT and GD3S KO mice ( Figure 5), indicating that b-series gangliosides are not involved in bone formation.

Discussion
In this study, we present the idea that deficiency of GD3 synthase in mice attenuates bone loss with aging ( Figure 7). Although b-series gangliosides were not expressed in osteoblasts, these gangliosides were detected in pre-osteoclasts such as RAW264.7 and primary cultured pre-osteoclasts derived from mouse bone marrow cells. Furthermore, 3D-µCT and bone histomorphometric analyses using GD3S KO mice revealed that deficiency of GD3 synthase in mice suppressed bone resorption, leading to attenuation of a decrease in bone mass with aging.
It has been reported that leptin secretion is disturbed by deficiency of b-series gangliosides, leading to a reduction in serum leptin levels in GD3S KO mice [7]. Leptin plays essential roles in the regulation of the mass of adipose tissue and body weight, and lack of leptin causes obesity [21][22][23]. Leptin is also known to increase bone resorption by the sympathetic nervous system [24]. In this study, we observed that body weight as well as bone mass in GD3S KO mice was higher than that in WT mice at the age of 40 weeks (supplementary Figure S1). Furthermore, bone resorption parameters in GD3S KO mice were lower than those in WT mice at the age of 40 weeks. These results suggest that lacking GD3 synthase reduces serum leptin levels, resulting in the attenuation of bone resorption. To determine whether secretion of leptin regulated by GD3 synthase is involved in bone metabolism, it is necessary to confirm serum levels of leptin in WT and GD3S KO mice, as we have reported previously [7], and examine whether injection of leptin into GD3S KO mice increases bone resorption with aging up to the same levels as WT mice.
Some studies have reported that gangliosides regulate differentiation of MSCs and tendon stem cells into osteoblasts [16][17][18]20]. For instance, it has been reported that GD1a plays important roles in the regulation of osteoblast differentiation of MSCs through the activation of epidermal growth factor receptor signaling [16,17]. GM1 also promotes osteoblastic differentiation of tendon stem cells by suppression of the phosphorylation of platelet-derived growth factor receptor-β (PDGFR-β) [20]. Furthermore, serum deprivation results in upregulation of the GM3 synthase gene as well as osteoblastic differentiation markers, such as Runx2 and osteocalcin [19]. Collectively, a-series gangliosides may be involved in the induction of osteoblastic differentiation. Levels of a-series gangliosides are known to be increased in GD3S KO mice [25] and we have confirmed that GD1a, among the a-series gangliosides, is expressed in MC3T3 E1 cells. However, elevation of a-series gangliosides in osteoblast cells may not be caused by a lack of GD3 synthase, since expression of b-series gangliosides was not observed in osteoblast cells and bone formation parameters were almost equivalent between WT and GD3S KO mice.
Although action mechanisms of gangliosides in bone metabolism were not fully understood in this study, our observations provide the first evidence of the role of GD3 synthase in bone metabolism in animals and suggest that altered ganglioside metabolism may underlie this phenomenon. In summary, we have demonstrated that a lack of GD3 synthase in mice results in the prevention of bone loss through the suppression of bone resorption based on a decrease in osteoclast number.

Mice
The generation of GD3S KO mice was carried out as described previously [11]. In brief, a targeting vector was used for gene knockout of GD3 synthase. The neo r gene was inserted between the BalI and AccI sites in exon 1 of the GD3 synthase gene and a 9.5-kb gene fragment was used as a targeting vector. The diphtheria toxin A gene was attached to eliminate nonhomologous recombinants [11]. WT and GD3S KO mice were mated and the resultant heterozygotes were mated, and genotypes of the offspring were screened as described [11]. The mice were housed under a 12 h light/dark cycle, and water and food were provided ad libitum. All protocols for animal experiments were approved by the Aichi Gakuin University Animal Research Committee (approval number AGUD312; 29 October 2015) (Nagoya, Japan), and were carried out in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (1966).

Cell Culture
The MC3T3 E1 mouse osteoblast-like cells and RAW264.7 mouse pre-osteoclast (monocyte/macrophage) cells were obtained from RIKEN BRC (Tsukuba, Japan) and American Type Culture Collection (Mnassas, VA, USA), respectively. The MC3T3 E1 cells, RAW264.7 cells, and mouse bone marrow cells isolated from long bones (femur and tibia) [26] were cultured in α-Minimum Essential Media with 10% fetal bovine serum and antibiotics (100 U/mL penicillin and 100 µg/mL streptomycin, Wako Pure Chemical Industries, Ltd., Osaka, Japan). Cells were maintained at 37 • C with 5% CO 2 in a humidified incubator. They were then stained with trypan blue and the numbers of living and dead cells were counted using a hemacytometer.

Induction to Mature Osteoblasts
The MC3T3 E1 cells were plated in 60 mm dishes. When cells were confluent, 50 µg/mL of ascorbic acid (Wako Pure Chemical Industries, Ltd., Osaka, Japan) and 5 mM β-glycerophosphate (Sigma-Aldrich, St. Louis, MO, USA) were added. The medium was changed every other day. After 7, 14, and 21 days of culture, cells were used for flow cytometry or quantitative real-time PCR.

In Vitro Osteoclast Induction
Mouse bone marrow cells 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 as osteoclast precursors [26]. These cells were cultured with 10 ng/mL M-CSF and 50 ng/mL RANKL (PeproTech, Inc.). After 24 h of treatment with RANKL, the cells were used for flow cytometry.

qPCR
Total RNA was extracted using an RNeasy Plus mini kit (Qiagen, Germantown, MD, USA). Reverse transcription was conducted with a high-capacity cDNA reverse transcription kit (Applied Biosystems, Carlsbad, CA, USA) and quantitative real-time PCR was performed using TaKaRa Thermal Cycler Dice Real Time System III with THUNDERBIRD SYBR qPCR mix kits (TOYOBO, Osaka, Japan). One µg total RNA was used for reverse transcription and the cycling conditions were 25 • C for 10 min, 37 • C for 120 min, and 85 • C for 5 min. The PCR cycling conditions were 95 • C for 10 min (pre-denaturation), 40 cycles at 95 • C for 15 sec (denaturation), and 60 • C for 1 min (extension). We evaluated the mRNA levels of St8sia1 using the primers listed in Table 1. Gapdh was used for an internal control. The relative mRNA abundance for the selected genes with respect to the level of Gapdh mRNA was expressed as a ratio of S MC Day 0, 7, 14, 21 / S RAW ( Figure 2B) or S RAW Day 2, 4 / S RAW Day 0 ( Figure 3C). S MC Day 0, 7, 14, 21 was the mRNA level for MC3T3 E1 cells before (Day 0) and after induction (days 7, 14, and 21) to mature osteoblasts. S RAW was the mRNA level for RAW264.7 cells. S RAW Day 0 and S RAW Day 2, 4 were the mRNA level for RAW264.7 cells before (Day 0) and after administration of RANKL (days 2 and 4), respectively.

Target
Forward Primer Backward Primer

Three-Dimensional Micro-Computed Tomography (3D-µCT Analysis)
The distal region of the femur was scanned using 3D-µCT (CosmoScan R-mCT-GX-T1, RIGAKU, Tokyo, Japan) and the parameters of cancellous bone were analyzed by TRI/3D-BON (Ratoc, Tokyo, Japan) software. In brief, the scanning was initiated at 0.5 mm above the distal femoral growth plate, and a total of 75 consecutive 20-µm-thick sections were analyzed.

TRAP Staining
TRAP staining was conducted as described previously [28]. In brief, the slides (femur samples) were incubated in sodium acetate buffer (0.1 M, pH 5.0) containing naphthol AS-MX phosphate, Fast Red Violet LB Salt, and MnCl 2 in the presence of sodium tartrate at 37 • C for 60 min for TRAP staining.

Bone Histomorphometric Analyses
Femur samples were fixed with 4% paraformaldehyde and decalcified in 10% ethylenediaminetetraacetic acid (EDTA) for three weeks, embedded in paraffin, and sectioned at 5 µm thickness. These slides were used for hematoxylin and eosin (HE) and TRAP staining. After staining with TRAP, osteoclast numbers on the bone surface and osteoclast surface on the bone surface were evaluated by scoring the TRAP-positive multi-nucleated cells on the bone surface as described previously [29]. Osteoblasts were stained with HE and osteoblast numbers on the bone surface and osteoblast surface on the bone surface were evaluated. The parameters were measured within an area of 0.8 mm 2 (1.0 mm × 0.8 mm), with the closest and furthest edges being 2.0 and 3.0 mm distal to the growth plate of the proximal ends of the femur, respectively [29].

Statistical Analysis
All data were expressed as mean ± SD. Statistical significance was evaluated using a Student's t-test ( Figure 3) and a two-tailed Mann and Whitney U test (Figures 4-6, and supplemental Figure S1) at p < 0.05; the single and double asterisks indicate p < 0.05 and p < 0.01, respectively.