Impact of sphingolipids on osteoblast and osteoclast activity in Gaucher disease

Abstract

Gaucher disease (GD) is an inherited disorder in which mutations in the GBA1 gene lead to deficient β-glucocerebrosidase activity and accumulation of its substrate glucosylceramide. Bone disease is present in around 84% of GD patients, ranging from bone loss including osteopenia and osteonecrosis to abnormal bone remodelling in the form of Erlenmeyer flask formation. The range of severity and variety of types of bone disease found in GD patients indicate the involvement of several mechanisms. Here we investigate the effects of exogenous sphingolipids on osteoclasts, osteoblasts, plasma cells and mesenchymal stem cells (MSC) and the interactions between these cell types. Osteoclasts were differentiated from the peripheral blood of Gaucher patients and control subjects. Osteoblasts were differentiated from mesenchymal stem cells isolated from bone marrow aspirates of Gaucher patients and control subjects. The human osteoblast cell line SaOS-2 was also investigated. Osteoclasts, osteoblasts and a human myeloma plasma cell line NCI-H929 were cultured with relevant exogenous sphingolipids to assess effects on cellular viability and function. Calcium deposition by osteoblasts differentiated from Gaucher patient MSC's was on average only 11.4% of that deposited by control subject osteoblasts. Culture with glucosylsphingosine reduced control subject MSC viability by 10.4%, SaOS-2 viability by 17.4% and plasma cell number by 40%. Culture with glucosylceramide decreased calcium deposition by control MSC-derived osteoblasts while increasing control subject osteoclast generation by 55.6%, Gaucher patient osteoclast generation by 37.6% and plasma cell numbers by up to 29.7%. Excessive osteoclast number and activity and reduced osteoblast activity may have the overall effect of an uncoupling between osteoclasts and osteoblasts in the GD bone microenvironment.

Introduction

Gaucher disease (GD) is an inherited disorder in which mutations in the GBA1 gene lead to deficient β-glucocerebrosidase (GC) activity and the consequent accumulation of its substrate glucosylceramide [1]. GD is typically divided into three types distinguished by the presence of neurological features in types 2 and 3 and an absence of such features in type 1 [2], the most common form of GD comprising ~94% of the GD patient population [3]. Common features of type 1 GD include hepatosplenomegaly, cytopenias and bone disease [4]. Present in around 84% of GD patients [5] bone disease can manifest in a number of forms ranging from bone loss including osteopenia, osteonecrosis osteosclerosis, osteolytic lesions, pathological fracture and abnormal bone remodelling in the form of Erlenmeyer flask formation of the femoral head [6].

GD bone disease can be partially explained by bone marrow infiltration of Gaucher cells, hypothesised to cause displacement of marrow cells to the periphery [7] which may lead to marrow infarcts including osteonecrosis of joints and to elicit an inflammatory response which may affect bone metabolism [8]. However, the range of severity and variety of types of bone disease found in GD patients indicate the involvement of several mechanisms which cannot be explained by Gaucher cell infiltration alone. Previous research by our group [9] and also Mucci et al. [10] have shown increased in vitro osteoclast generation and activity when differentiated from PBMC's isolated from GD patients. In addition, murine and zebrafish models of GD demonstrated impaired osteoblast differentiation and bone mineralisation [11, 12] and altered osteocyte function and viability in a CBE murine cell line model [13].

Studies have shown an increased risk of multiple myeloma, a bone marrow malignancy, for type 1 GD patients [14, 15]. In de Fost et al., multiple myeloma risk was estimated to be 51.1 fold elevated and prevalence of monoclonal gammopathy of uncertain significance also increased. This may be relevant as the onset of MGUS can coincide with skeletal fragility, and bone destruction might occur at a very early stage [16]. >80% of MM patients develop bony lesions leading to pain and fractures with lytic lesions most common in the spine, skull and long bones with widespread osteopenia also a common feature [17]. Patients with MM exhibit increased numbers of osteoclast progenitor cells in the peripheral blood [18], with osteoclast numbers increased in the bone marrow of patients with monoclonal gammopathy, increasing further in patients with malignant disease [19]. Evidence suggests myeloma cells stimulate osteoclast generation via direct interaction [20]. In turn, osteoclasts have been shown to support myeloma cell survival and enhance proliferation [21, 22]. A flow cytometry based study of bone marrow from MM patients noted an increased number of colony forming mesenchymal stem cells (MSC's) which correlated with disease burden at time of diagnosis, suggesting MSC's and osteoblasts may play an important role in the proliferation and survival of myeloma cells [23].

Sphingolipids are a major category of lipids and are present in all mammalian cells [24]. Analysis of the lipid composition of GD patient plasma and urine found elevated levels of 20 plasma and 10 urinary lipids including species of phosphatidylcholine, sphingomyelin and ceramides [25]. With several publications also showing substantially higher levels of glucosylsphingosine [26, 27] in GD plasma and serum these findings demonstrate the deficiency of GC affects the synthesis and degradation pathways of many sphingolipids.

Recent research in unrelated diseases have shown that sphingolipids are highly bioactive and have been theorised to play a role in a number of conditions including cancer [28]. Ceramide may be a mediator of pro-apoptotic pathways and an anti-inflammatory agent [[29], [30], [31]]. Glucosylsphingosine has been suggested to mediate cellular dysfunction [32] whereas glucosylceramide has also been linked with immunosurveillance [33] and cell proliferation [34].

Sphingolipids have been found to modulate osteoclast formation and function. Ceramide has been shown to reduce osteoclast activity by inhibiting actin ring formation [35] whilst lactosylceramide increases RANK expression in osteoclasts [36]. Furthermore sphingosine-1-phosphate (S1P) has been found to be a chemotactic factor, regulating precursor osteoclast migration between the bone marrow and the blood [37]. Research is also revealing potential roles of sphingolipids in osteoblast recruitment and activity. S1P produced by osteoclasts is a chemoattractant for MSC's expressing its receptors, S1PR1 and S1PR2, resulting in their migration to the bone marrow and thus also acting as a coupling factor between osteoclasts and osteoblasts [38].

Here we investigate the effects of exogenous sphingolipids on osteoclasts, osteoblasts, plasma cells and mesenchymal stem cells and the interactions between these cell types.