G Model BONSOI-4001; No. of Pages 7
ARTICLE IN PRESS Joint Bone Spine xxx (2014) xxx–xxx
Available online at www.sciencedirect.com
Review
Bone marrow fat Pierre Hardouin a,b,c,∗ , Vittorio Pansini a,d,e,f , Bernard Cortet a,d,e,g a
Université Lille Nord-de-France, 59000 Lille, France Université du Littoral Côte-d’Opale PMOI, boulevard Napoléon, BP 120, 62327 Boulogne-sur-Mer, France c PMOI EA 4490, 62327 Boulogne-sur-Mer, France d PMOI EA 4490, 59000 Lille, France e Université droit et santé Lille 2, 59000 Lille, France f Service d’imagerie musculo-squeletique, CHRU, 59000 Lille, France g Service de rhumatologie, CHRU, 59000 Lille, France b
a r t i c l e
i n f o
Article history: Accepted 6 January 2014 Available online xxx Keywords: Bone marrow Adipocyte Osteoblast Fat Osteoporosis Anorexia nervosa
a b s t r a c t Bone marrow fat (BMF) results from an accumulation of fat cells within the bone marrow. Fat is not a simple filling tissue but is now considered as an actor within bone microenvironment. BMF is not comparable to other fat depots, as in subcutaneous or visceral tissues. Recent studies on bone marrow adipocytes have shown that they do not appear only as storage cells, but also as cells secreting adipokines, like leptin and adiponectin. Moreover bone marrow adipocytes share the same precursor with osteoblasts, the mesenchymal stem cell. It is now well established that high BMF is associated with weak bone mass in osteoporosis, especially during aging and anorexia nervosa. But numerous questions remain discussed: what is the precise phenotype of bone marrow adipocytes? What is the real function of BMF, and how does bone marrow adipocyte act on its environment? Is the increase of BMF during osteoporosis responsible for bone loss? Is BMF involved in other diseases? How to measure BMF in humans? A better understanding of BMF could allow to obtain new diagnostic tools for osteoporosis management, and could open major therapeutic perspectives. © 2014 Société franc¸aise de rhumatologie. Published by Elsevier Masson SAS. All rights reserved.
The organ “bone” is not only composed of bone tissue, but also includes bone marrow, which is located between bone trabeculae of spongious bone, and inside the diaphysis of long bones. Bone marrow is called “red marrow” when it is mainly filled with haematopoietic cells, and “yellow marrow” when it contains a majority of adipocytes. The existence of fat inside bones is known since hundred years, but it is much more recently that bone marrow fat (BMF) aroused the interest of researchers, as illustrated by the increasing number of publications over the last fifteen years in the database “PubMed” (Fig. 1). Indeed BMF does not appear any more as a simple filling tissue, but is considered from now as an actor within bone microenvironment. Due to these recent works, BMF is known better and better and its involvement in bone physiopathology is more and more argued. However numerous questions remain debated.
∗ Corresponding author. E-mail address: [email protected] (P. Hardouin).
1. Bone Marrow Fat (BMF): main characteristics BMF results from an accumulation of fat cells within bone marrow. These adipocytes contain a big lipid vacuole of triglycerides made of fatty acids, which can be saturated, mono or polyunsaturated.
1.1. Variations with age At birth, bone cavities are mainly filled with red hematopoietic marrow. Then occurs during childhood a “conversion” of the red marrow which is gradually replaced by yellow, fat marrow. This conversion of bone marrow begins in terminal phalanges after birth, then goes forward by a centripetal evolution up to the axial skeleton [1]. So, at the age of 25, red marrow is limited to the axial skeleton, ribs and breastbone. A “reconversion” can be observed in conditions of hypoxia, such as in smokers or in patients with obstructive sleep apnoea. Although large individual variations are found, it exists globally a positive correlation between BMF and age [2,3].
doi:10.1016/j.jbspin.2014.02.013 1297-319X/© 2014 Société franc¸aise de rhumatologie. Published by Elsevier Masson SAS. All rights reserved.
Please cite this article in press as: Hardouin P, et al. Bone marrow fat. Joint Bone Spine (2014), doi:10.1016/j.jbspin.2014.02.013
G Model
ARTICLE IN PRESS
BONSOI-4001; No. of Pages 7
P. Hardouin et al. / Joint Bone Spine xxx (2014) xxx–xxx
2 Bone Marrow fat (Medline)
250 200 150 100 50 0 1998
2000
2002
2004
2006
2008
2010
2012
Fig. 1. Annual number of publications referenced in the PubMed database between 1998 and 2012 with keywords “bone marrow fat”.
1.2. Location and quantity In adults, yellow bone marrow is mainly located in the appendical skeleton. Its volume is approximately estimated at 7% of total fat [3] and so 2,6 kg in an adult, which represents a storage of about 23,000 calories [4]. However important variations are found, not only with age as previously seen, but also according to gender (BMF is higher in men than in women) and to anatomical location (BMF is higher in long bones diaphysis than in axial skeleton) [1,5,6]. In a given long bone, yellow marrow is preferentially located in the diaphysis and in the epiphyses, whereas red marrow is mainly localized in the metaphysis [1]. Moreover, quantitative modifications of BMF can be observed in several diseases, especially in osteoporosis [5–7]. 1.3. Microscopic aspect Bone marrow adipocytes are not grouped in lobules as in the other fat depots, but they are scattered within the hematopoietic tissue. Mean diameter of these adipocytes is around 50 , which is lower than the diameter of subcutaneous or visceral adipocytes [8,9], but great variations are found in literature according to the location. In usual optic microscopy of undecalcified bone, the technical preparation of the sample removes the lipid content, then the shapes of adipocytes appear as cellular “ghosts” (Fig. 2). Bone marrow adipocytes can be also studied with confocal microscopy or on frozen slides or by microtomography after labelling with osmium. 1.4. In vivo imaging Measurement of BMF can be made by X-rays quantitative tomography but the geometrical resolution of this method is weak and a preferential loss of photons with low energy leads to underestimation of the quantity of fat in long bones [10]. Magnetic resonance imaging (MRI) is the best method to distinguish yellow fat marrow from haematopoietic red marrow. Yellow marrow appears in hypersignal in T1-weighted sequences having intermediate to elevated signal on spin-echo T2 images and on T2 fast spin-echo images and low signal on sequences with fat saturation. Yellow bone marrow does not enhance significantly after gadolinium injection. Red bone marrow shows intermediate signal on T1 or on T2 images, generally more elevated than muscular or yellow bone marrow signal. On fat-suppressed images red bone marrow shows intermediate to elevated signal, thus constantly higher than yellow bone marrow. In any case the difference of signal between the two types of bone marrow (red and yellow) is less evident on T2 images than on T1 or on STIR ones. Red bone marrow shows generally a mild enhancement after injection of gadolinium.
Fig. 2. Microscopic aspect of bone marrow fat on iliac crest biopsy in osteoporosis. Undecalcified microscopy, May Grünwald Giemsa staining (bar = 100 ): a: osteoporotic woman, 56 years old; weak amount of marrow adiposity; mean adipocyte diameter = 51,5 ; adipocytes number = 108/mm2 : b: osteoporotic woman, 68 years old; important amount of adiposity; mean adipocyte diameter = 61,9 ; adipocytes number = 170/mm2 .
Altough MRI permits to depict precisely yellow and red bone barrow, it provides no quantification of these two components. Proton magnetic resonance spectroscopy (1 H MRS) allows to quantify BMF compared to water quantity with a good reproducibility [11,12]. The obtained spectra show water and fat peaks that allows to estimate the rate of BMF (Fig. 3). The percentage of fat fraction linearly increases with age, less fast in women than in men [13], ranging from 20.5% for the second and third decades of life to 49.4% for the eighth and ninth decades [14]. MRI thus confirms the increase of BMF in elderly patients but also in osteoporotic patients. 2. Recent data on BMF 2.1. BMF is a specific fat depot It is now established that BMF constitutes a specific fat location, which is not comparable to subcutaneous and abdominal fat depots. BMF differs because of its location inside bones, which leads to interactions with the bone microenvironment, because of a more scattered distribution of fat cells, and because of its composition in fatty acids [15]. BMF is not correlated with weight, body mass index or body fat [6,16,17]. BMF is even increased in anorexia nervosa, contrasting with the meagreness of these patients [8,18], which
Please cite this article in press as: Hardouin P, et al. Bone marrow fat. Joint Bone Spine (2014), doi:10.1016/j.jbspin.2014.02.013
G Model BONSOI-4001; No. of Pages 7
ARTICLE IN PRESS P. Hardouin et al. / Joint Bone Spine xxx (2014) xxx–xxx
3
the expense of adipogenesis, and ob/ob mice, which fail to express leptin, have a low trabecular bone volume, and an increase of femoral BMF [28]. 2.4. There is a link between BMF and bone mass Pierre Meunier was the first to describe in 1971 the inverse correlation existing between BMF and age or trabecular bone volume [5]. These correlations were confirmed by numerous other studies, in animals as in humans [29,30]. In humans it is now well established that high BMF is associated with low bone mass in osteoporosis, in particular during anorexia nervosa and with aging [6,7,18]. 3. Discussed questions Bone marrow adipocyte is a specific and active cell which interacts with its environment. A good knowledge of BMF would bring us a lot, in diagnosis as in therapeutics, but several aspects remain not completely elucidated and the following list of open matters is far to be exhaustive. 3.1. What is the phenotype of marrow adipocyte?
Fig. 3. Normal MRI (1H MR spectroscopy) characteristics of bone marrow in a 28year-old woman; a: zone of interest (black square) in the femoral neck on a T1weighted oblique coronal image; b: spectra obtained in the zone of interest during the same examination. Medullar fat quantity was estimated at 73.5%.
indicates that BMF is not submitted to the same regulations as other fat locations. Bone marrow adipocytes result from a differentiation of mesenchymal stem cells, and data obtained from other adipocytes cannot be extrapolated to bone marrow cells. Functionally, numerous genes are differentially expressed in bone marrow adipocytes compared to epididymal adipocytes in mice [19] and compared to subcutaneous adipocytes in human [20]. In mice, gene profiling reveals a unique phenotype for primary bone marrow adipocytes characterized by low adipose-specific gene expression and high expression of inflammatory response genes [19]. Moreover bone marrow adipocytes are less responsive to high fat diet than epididymal adipocytes. 2.2. Bone marrow adipocyte is an active cell As other adipocytes, bone marrow adipocytes are secretory cells and have not only a simple function of storage; they are involved in secreting numerous adipokines in the bone microenvironment, in particular leptin and adiponectin. Leptin can be secreted in greater quantity, at least in vitro, by marrow adipocytes than by subcutaneous adipocytes [21]. Locally produced adiponectin may have a positive effect on bone [22]. Moreover, adipocyte lipolysis produces fatty acids suspected to have a negative effect on osteoblasts [23]. 2.3. Bone marrow adipocytes and osteoblasts have a common origin Osteoblasts and adipocytes share a common progenitor, the mesenchymal stem cell (MSC) [24]. This common origin leads to a competition between the alternative differentiation towards adipogenesis or osteogenesis. This alternative is mainly regulated by the transcription factors PPAR␥ for adipogenesis and runx2 for osteoblastogenesis, with the cooperation of numerous factors, in particular the Wnt pathway [25–27]. Leptin favours osteogenesis at
Bone marrow adipocyte remains incompletely known and at the moment we do not have a specific marker of “bone marrow adipocyte”. A debate regarding the white or brown phenotype of theses adipocytes is still open. Brown adipocytes contain several small lipid vacuoles, numerous mitochondria, and they are dedicated to heat production thanks to the presence of UCP-1 (“uncoupled protein-1”) while white adipocytes have a role in metabolic storage. Some characteristics of brown fat have been attributed to marrow adipocytes [31] but these results have to be confirmed after a rigorous selection of bone marrow adipocytes. Nevertheless, BMF is sometimes presented as an intermediate tissue between brown and white fat, the “yellow adipose tissue” [22]. 3.2. How does marrow adipocyte act on its environment? Similarly to other adipocytes bone marrow adipocytes can store or release lipids. Saturated fatty acids secreted by MSC-derived adipocytes have a negative effect on osteoblasts in coculture (lower mineralization, alkaline phosphatase activity and expression of osteogenic mRNA markers) [23]. However, the concentrations of palmitic and oleic acids necessary to exert these effects are more elevated than those physiologically released by MSC-derived adipocytes [32]. Thus it is not established that fatty acids released by marrow adipocytes play a significant role in physiology. On the other hand, MSC-derived adipocytes are capable of inducing MSC-derived osteoblasts to develop some adipocyte characteristics [33]. The whole secretome of marrow adipocytes is not currently known, and factors responsible for this possible transdifferentiation have not yet been identified. Furthermore, marrow adipocytes also secrete microvesicles which can be involved in this conversion. A paracrine secretion exists, and an increase of pro-inflammatory cytokines has been found in medullary fluids from osteoporotic patients [34], but it would be necessary to make sure that this increase results from a local production by marrow adipocytes. Finally, marrow adipocytes express RANKL which allows to promote osteoclastic differentiation [35]. 3.3. What is the precise function of BMF? The precise function of BMF is not known, what makes difficult its integration in physiopathological regulations. Some authors consider that BMF is involved in thermogenesis because marrow
Please cite this article in press as: Hardouin P, et al. Bone marrow fat. Joint Bone Spine (2014), doi:10.1016/j.jbspin.2014.02.013
G Model BONSOI-4001; No. of Pages 7
ARTICLE IN PRESS P. Hardouin et al. / Joint Bone Spine xxx (2014) xxx–xxx
4
adipocytes could possess some characteristics of brown fat [31]. An involvement in energy metabolism can also be situated at the local level, as marrow adipocytes could supply energy for bone formation or to optimize local temperature. On the secretory plan it has been recently suggested that circulating adiponectin could mainly result from BMF secretion, what could explains the “adiponectin paradox” resulting from the inverse correlation between fat mass and circulating adiponectin level [4]. However the differences observed in gene expression profile between bone marrow adipocytes and adipocytes from other origins [4,19,20] do not argue for a strong metabolic implication of marrow adipocytes, or in any case not for the typical adipocyte functions (low expression of the adipocyte specific genes PPAR␥, FABP4, perilipin). Bone marrow adipocytes express some stem cell markers, and a plasticity of theses cells is suggested. Marrow adipocytes modify the osteoblasts phenotype [33] and favour osteoclast differentiation [35]. Thus several data suggest an active role of bone marrow adipocytes on their local microenvironment. 3.4. Is the increase of BMF responsible for bone loss during osteoporosis? Osteoporosis is associated with an increase of BMF [5–7,9]. But this association does not imply that BMF is responsible for bone loss. A lot of in vitro and in vivo works have shown a link between aging, or oestrogen deficiency, and an increase of BMF. PPAR ␥2 increases with age because of a decreased level of inhibitors such as estrogens, TGF, IGF-1 or members of wnt pathway, that favours adipogenesis at the expense of osteoblastic differentiation [26]. In animals, ovariectomy increases BMF [36], which is reversible with estrogens administration. In humans, the increase of BMF begins before the third decade, which seems premature for a simple compensatory filling of bone loss, and bone acquisition in young subjects is conversely correlated to modifications of BMF [3,17]. BMF is strongly negatively correlated with bone mineral density (BMD) [16,37]; correlations with BMD show that BMF is higher in osteoporotic than in osteopenic subjects, and is higher in patients with vertebral fractures than in osteoporotic patients without fracture [2]. Premenopausal osteoporotic women have substantially higher marrow adipocyte number, size and volume than controls, and these results remain significant after adjusting for age, body mass index, and bone volume [9]. Finally the increase of BMF associated with menopause is corrected by the administration of estrogens [38]. Thus, numerous arguments suggest a responsibility of BMF in bone loss, even if this involvement is not yet proved. 3.5. Is BMF involved in other bone diseases? 3.5.1. Anorexia nervosa Osteoporosis is a frequent complication of anorexia nervosa. The increase of BMF is strongly suspected to participate in bone loss in anorexia nervosa [8,18]. The increase of BMF measured with MRI is conversely correlated to BMD and is associated with a decrease of bone formation. The increase of the circulating level of Pref1, a preadipocyte factor which regulates adipogenesis, could explain this increase of BMF [39]. In animals, caloric restriction leads to high marrow adiposity [40], which is blunted by leptin administration [41]. 3.5.2. Disuse osteoporosis Patients with disuse osteoporosis present MRI abnormalities of bone marrow with presence of subchondral lobules of fat [42]. On the opposite mechanical strains inhibit PPAR␥, and athletes involved in weight-bearing impact sports have lower BMF when
compared with athletes involved in non impact loading sports, and with non athletic controls [43]. 3.5.3. Femoral head osteonecrosis In femoral head osteonecrosis there is a hypertrophy of bone marrow adipocytes which could favour secondary vascular blood perturbations. Moreover, the main risk factors for femoral head osteonecrosis, alcohol and glucocorticoids, increase adipocyte differentiation. In the same way, femoral heads with the highest amount of BMF before treatment seem to have the highest risk of necrosis during glucocorticoids administration [44]. 3.6. Is BMF involved in non-osseous diseases? 3.6.1. Obesity and metabolic diseases During obesity and metabolic diseases an increase of BMF is likely [29] but few studies are available on this subject, and this notion has to be confirmed, especially as the increase of BMF noticed in animal models of type 1 diabetes was not found in humans [45]. The amount of BMF is not correlated to metabolic risk in young healthy patients [17]. 3.6.2. Intervertebral disc pathologies The vertebral bone marrow is involved in intervertebral disc nutrition, and a fat conversion is suspect to participate in disc degeneration [46]. 3.6.3. Haematological diseases Marrow adipocytes, and the balance between adipocytes and osteoblasts, control the haematopoietic activity of haematopoietic stem cells. Marrow adipocyte is a negative regulator of haematopoiesis [47], and may influence the progression of chronic myeloid leukaemia [48]. The balance between adipocytes and osteoblasts also plays an important role in haematopoietic recovery after radiotherapy. 3.6.4. Fat embolism We only mention the particular case of fat embolism that results from vascular obstruction by fat droplets stemming from bone marrow, especially after long bone fracture, fracture nailing or vertebroplasty. Fig. 4 summarizes the potential roles of marrow adipocytes, in normal situation and in pathology. 3.7. How to evaluate BMF in patients? MRI appears to be the best method for measuring BMF in humans. However various equipments and methods are described in literature, which may lead to discrepancies, favoured by the heterogeneity of BMF [49]. 1 H MRS (12) is the method of choice for measuring the water/lipid ratio, and so to evaluate BMF [13]. The chemical fat fraction is poorly correlated with histological measurements but the latter measures the adipocyte volume while spectroscopy separates fat and water to measure the chemical fraction of fat [50]. MRI spectroscopy also allows studying lipid saturation (presence and type of hydrogen bindings) within the bone marrow. The obtained spectrum contains peaks corresponding to water, saturated lipids, unsaturated lipids and residual lipids [51], and the prevalence of fragility fractures is associated with a lower unsaturation levels. The application of magnetic resonance spectroscopy to quantify marrow fat is limited because of the non homogeneous distribution of BMF in bones. “Whole body” BMF acquisition (10 mm thick axial images at 40 mm intervals from fingers to toes) has been proposed to overcome this limitation and to
Please cite this article in press as: Hardouin P, et al. Bone marrow fat. Joint Bone Spine (2014), doi:10.1016/j.jbspin.2014.02.013
G Model BONSOI-4001; No. of Pages 7
ARTICLE IN PRESS P. Hardouin et al. / Joint Bone Spine xxx (2014) xxx–xxx
5
Fig. 4. Summary of the potential roles of bone marrow adipocytes, in normal and pathological conditions. OB: osteoblasts; OC: osteoclast; MSC: mesenchymal stem cell; HSC: haematopoietic stem cell.
allow evaluation of interaction between fat and bone independent of subcutaneous or visceral adipose tissue by adjustment for body composition [3]. The study of bone marrow perfusion is another interesting new approach [52]. However there is currently no standardized protocol for BMF imaging.
3.8. Is the evaluation of BMF relevant to estimate bone fracture risk? A negative correlation is found between BMF and BMD, even after adjustment for age and body fat. This finding does not mean that BMF is an independent variable. Also vertebral 1 H MRS percent fat fraction is higher in subjects with MR findings of weakened bone compared to control group [14], and higher BMF is associated with prevalent fracture in men, even after adjustment for BMD [53]. In postmenopausal women, patients with fragility fractures or type 2 diabetes have lower unsaturation levels and higher saturation levels, that suggests that 1 HMRS of spinal bone marrow may therefore serve as a novel tool for BMD-independent fracture risk assessment [51]. Besides, an increase of BMF, as observed in osteoporotic patients, modifies the measurement of bone mineral content with DEXA (dual-energy X-ray absorptiometry), but this modification seems minor for fracture risk assessment.
3.9. What are the therapeutic implications? Considering the links between the two parameters, factors modifying BMF should have an impact on bone mass. Adipogenesis is mainly under the control of PPAR␥2 [26] and long-term use of thiazolidinediones, which are PPAR␥ agonists proposed in the treatment of type 2 diabetes, reduces lumbar spine BMD and doubles the risk fracture among women with type 2 diabetes [54]. An increase of BMF is also observed with long-term use of
glucocorticoids which favours in vitro adipogenesis, as well as methotrexate, antiproteases, or after radiotherapy [55,56]. On the opposite, a decrease of BMF could be a strategy in the aim of an increase of bone mass. Indeed, a decrease of marrow adipogenesis is observed in postmenopausal women treated with risedronate [57] or with estrogens [38]. In rodents, a decrease of BMF is also observed after administration of strontium ranelate, or after PTH administration. In the future, new drugs against bone loss could be specifically designed to target BMF. In this aim, the inhibition of PPAR␥ in mice indeed increases bone mass [58]. 4. Conclusions and perspectives More and more findings argue for an involvement of BMF in bone physiopathology, and BMF is no more considered as a simple filling tissue. Bone marrow adipocyte is a specific adipocyte which interacts with bone and haematopoietic cells, and there is an inverse correlation between BMF and bone mass. But numerous unsolved questions remain, and BMF is probably not a simple antagonist of bone formation. In addition to a physiological role in the bone microenvironment, which remains to be clarified it is possible that the pathological increase of BMF results from a defence process against stress, and especially against oxidative stress [59,60]. This mechanism, not specific to bone, would lead in this particular tissue to a vicious circle by interaction with bone cells, ending in bone loss. Anyway, a better understanding of BMF could give us new diagnosis or prognosis criteria, and could open major therapeutic perspectives. Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
Please cite this article in press as: Hardouin P, et al. Bone marrow fat. Joint Bone Spine (2014), doi:10.1016/j.jbspin.2014.02.013
G Model BONSOI-4001; No. of Pages 7
ARTICLE IN PRESS P. Hardouin et al. / Joint Bone Spine xxx (2014) xxx–xxx
6
Acknowledgements Authors thank Anne Cotten, Séverine Delplace and Stéphanie Lucas for their help. References [1] Vande Berg BC, Malghem J, Lecouvet FE, et al. Magnetic resonance imaging of normal bone marrow. Eur Radiol 1998;8:1327–34. [2] Wehrli FW, Hopkins JA, Hwang SN, et al. Cross-sectional study of osteopenia with quantitative MR imaging and bone densitometry. Radiology 2000;217:527–38. [3] Shen W, Chen J, Gantz M, et al. MRI-measured pelvic bone marrow adipose tissue is inversely related to DXA-measured bone mineral in younger and older adults. Eur J Clin Nutr 2012;66:983–8. [4] Scheller E, Cawthorn W, Learman B, et al. The metabolic nature of marrow fat: insulin signaling. CREB phosphorylation and the « adiponectin paradox ». J Bone Miner Res 2013:S241 [ASBMR 2013 Annual Meeting]. [5] Meunier P, Aaron J, Edouard C, et al. Osteoporosis and the replacement of cell populations of the marrow by adipose tissue. A quantitative study of 84 iliac bone biopsies. Clin Orthop Rel Res 1971;80:147–54. [6] Justesen J, Stenderup K, Ebbesen EN, et al. Adipocyte tissue volume in bone marrow is increased with aging and in patients with osteoporosis. Biogerontology 2001;2:165–71. [7] Verma S, Rajaratnam JH, Denton J, et al. Adipocytic proportion of bone marrow is inversely related to bone formation in osteoporosis. J Clin Pathol 2002;55:693–8. [8] Abella E, Feliu E, Granada I, et al. Bone marrow changes in anorexia nervosa are correlated with the amount of weight loss and not with other clinical findings. Am J Clin Pathol 2002;118:582–8. [9] Cohen A, Dempster DW, Stein EM, et al. Increased marrow adiposity in premenopausal women with idiopathic osteoporosis. J Clin Endocrinol Metab 2012;97:2782–91. [10] Di Iorgi N, Rosol M, Mittelman SD, et al. Reciprocal relation between marrow adiposity and the amount of bone in the axial and appendicular skeleton of young adults. J Clin Endocrinol Metab 2008;93:2281–6. [11] Schellinger D, Lin CS, Lim J, et al. Bone marrow fat and bone mineral density on proton MR spectroscopy and dual-energy X-ray absorptiometry: their ratio as a new indicator of bone weakening. AJR Am J Roentgenol 2004;183:1761– 5. [12] Li X, Kuo D, Schafer AL, et al. Quantification of vertebral bone marrow fat content using 3 Tesla MR spectroscopy: reproducibility, vertebral variation, and applications in osteoporosis. J Magn Reson Imaging 2011;33:974– 9. [13] Pansini V, Monnet A, Salleron J, et al. 3 Tesla (1) H MR spectroscopy of hip bone marrow in a healthy population, assessment of normal fat value and influence of age and sex. J Magn Reson Imaging 2014;39:369–76. [14] Schellinger D, Lin CS, Hatipoglu HG, et al. Potential value of vertebral proton MR spectroscopy in determining bone weakness. AJNR Am J Neuroradiol 2001;22:1620–7. [15] Griffith JF, Yeung DKW, Ahuja AT, et al. A study of bone marrow and subcutaneous fatty acid composition in subjects of varying bone mineral density. Bone 2009;44:1092–6. [16] Shen W, Scherzer R, Gantz M, et al. Relationship between MRI-measured bone marrow adipose tissue and hip and spine bone mineral density in AfricanAmerican and Caucasian participants: the CARDIA study. J Clin Endocrinol Metab 2012;97:1337–46. [17] Di Iorgi N, Mo AO, Grimm K, et al. Bone acquisition in healthy young females is reciprocally related to marrow adiposity. J Clin Endocrinol Metab 2010;95:2977–82. [18] Ecklund K, Vajapeyam S, Feldman HA, et al. Bone marrow changes in adolescent girls with anorexia nervosa. J Bone Miner Res 2010;25:298–304. [19] Liu L-F, Shen W-J, Ueno M, et al. Characterization of age-related gene expression profiling in bone marrow and epididymal adipocytes. BMC Genomics 2011;12:212. [20] Poloni A, Maurizi G, Serrani F, et al. Molecular and functional characterization of human bone marrow adipocytes. Exp Hematol 2013;41:558e2–66e2. [21] Laharrague P, Fontanilles AM, Tkaczuk J, et al. Inflammatory/haematopoietic cytokine production by human bone marrow adipocytes. Eur Cytokine Netw 2000;11:634–9. [22] Lecka-Czernik B. Marrow fat metabolism is linked to the systemic energy metabolism. Bone 2012;50:534–9. [23] Wang D, Haile A, Jones LC. Dexamethasone-induced lipolysis increases the adverse effect of adipocytes on osteoblasts using cells derived from human mesenchymal stem cells. Bone 2013;53:520–30. [24] Pittenger MF, Mackay AM, Beck SC, et al. Multilineage potential of adult human mesenchymal stem cells. Science 1999;284:143–7. [25] Sadie-Van Gijsen H, Hough FS, Ferris WF. Determinants of bone marrow adiposity: The modulation of peroxisome proliferator-activated receptor-␥2 activity as a central mechanism. Bone 2013;56:255–65. [26] Marie PJ, Saidak S. Bone marrow adipocytes and bone aging. Osteoporos Int 2013;24:S453–5. [27] Nuttall ME, Shah F, Singh V, et al. Adipocytes and the Regulation of Bone Remodeling: A Balancing Act. Calcif Tissue Int 2014;94:78–87.
[28] Hamrick MW, Pennington C, Newton D, et al. Leptin deficiency produces contrasting phenotypes in bones of the limb and spine. Bone 2004;34:376– 83. [29] Rosen CJ, Bouxsein ML. Mechanisms of disease: is osteoporosis the obesity of bone? Nat Clin Pract Rheumatol 2006;2:35–43. [30] Fazeli PK, Horowitz MC, MacDougald OA, et al. Marrow fat and bone – new perspectives. J Clin Endocrinol Metab 2013;98:935–45. [31] Krings A, Rahman S, Huang S, et al. Bone marrow fat has brown adipose tissue characteristics, which are attenuated with aging and diabetes. Bone 2012;50:546–52. [32] Lucas S, Clabaut A, Ghali O, et al. Implication of fatty acids in the inhibitory effect of human adipocytes on osteoblastic differentiation. Bone 2013;55:429– 30. [33] Clabaut A, Delplace S, Chauveau C, et al. Human osteoblasts derived from mesenchymal stem cells express adipogenic markers upon coculture with bone marrow adipocytes. Differentiation 2010;80:40–5. [34] Pino AM, Ríos S, Astudillo P, et al. Concentration of adipogenic and proinflammatory cytokines in the bone marrow supernatant fluid of osteoporotic women. J Bone Miner Res 2010;25:492–8. [35] Goto H, Hozumi A, Osaki M, et al. Primary human bone marrow adipocytes support TNF-␣-induced osteoclast differentiation and function through RANKL expression. Cytokine 2011;56:662–8. [36] Elbaz A, Rivas D, Duque G. Effect of estrogens on bone marrow adipogenesis and Sirt1 in aging C57BL/6J mice. Biogerontology 2009;10:747–55. [37] Griffith JF, Yeung DKW, Antonio GE, et al. Vertebral marrow fat content and diffusion and perfusion indexes in women with varying bone density: MR evaluation. Radiology 2006;241:831–8. [38] Syed FA, Oursler MJ, Hefferanm TE, et al. Effects of estrogen therapy on bone marrow adipocytes in postmenopausal osteoporotic women. Osteoporos Int J Establ Result Coop Eur Found Osteoporos Natl Osteoporos Found USA 2008;19:1323–30. [39] Fazeli PK, Bredella MA, Misra M, et al. Preadipocyte factor-1 is associated with marrow adiposity and bone mineral density in women with anorexia nervosa. J Clin Endocrinol Metab 2010;95:407–13. [40] Devlin MJ, Cloutier AM, Thomas NA, et al. Caloric restriction leads to high marrow adiposity and low bone mass in growing mice. J Bone Miner Res 2010;25:2078–88. [41] Devlin M, Conlon C, Van Vliet M, et al. Leptin blunts the starvation-induced increase in bone marow adiposity. J Bone Miner Res 2013:S 342 [ASBMR Annual Meeting]. [42] De Abreu MR, Wesselly M, Chung CB, et al. Bone marrow MR imaging findings in disuse osteoporosis. Skeletal Radiol 2011;40:571–5. [43] Rantalainen T, Nikander R, Heinonen A, et al. Differential effects of exercise on tibial shaft marrow density in young female athletes. J Clin Endocrinol Metab 2013;98:2037–44. [44] Vande Berg BC, Gilon R, Malghem J, et al. Correlation between baseline femoral neck marrow status and the development of femoral head osteonecrosis in corticosteroid-treated patients: a longitudinal study by MR imaging. Eur J Radiol 2006;58:444–9. [45] Slade JM, Coe LM, Meyer RA, et al. Human bone marrow adiposity is linked with serum lipid levels not T1-diabetes. J Diabetes Complications 2012;26:1– 9. [46] Liu Q, Jin L, Shen FH, et al. Fullerol nanoparticles suppress inflammatory response and adipogenesis of vertebral bone marrow stromal cells-a potential novel treatment for intervertebral disc degeneration. Spine 2013;13:1571– 80. [47] Naveiras O, Nardi V, Wenzel PL, et al. Bone marrow adipocytes as negative regulators of the haematopoietic microenvironment. Nature 2009;460:259– 63. [48] Belaid-Choucair Z, Lepelletier Y, Poncin G, et al. Human bone marrow adipocytes block granulopoiesis through neuropilin-1-induced granulocyte colony-stimulating factor inhibition. Stem Cells Dayt Ohio 2008;26:1556– 64. [49] Shen W, Gong X, Weiss J, et al. Comparison among T1-weighted magnetic resonance imaging, modified dixon method, and magnetic resonance spectroscopy in measuring bone marrow fat. J Obes 2013;2013:75–2986. [50] Pichardo JC, Milner RJ, Bolch WE. MRI measurement of bone marrow cellularity for radiation dosimetry. J Nucl Med 2011;52:1482– 9. [51] Patsch JM, Li X, Baum T, et al. Bone marrow fat composition as a novel imaging biomarker in postmenopausal women with prevalent fragility fractures. J Bone Miner Res 2013;28:1721–8. [52] Griffith JF, Yeung DKW, Leung JCS, et al. Prediction of bone loss in elderly female subjects by MR perfusion imaging and spectroscopy. Eur Radiol 2011;21:1160–9. [53] Schwartz AV, Sigurdsson S, Hue TF, et al. Vertebral bone marrow fat associated with lower trabecular BMD and prevalent vertebral fracture in older adults. J Clin Endocrinol Metab 2013;98:2294–300. [54] Loke YK, Singh S, Furberg CD. Long-term use of thiazolidinediones and fractures in type 2 diabetes: a meta-analysis. CMAJ Can Med Assoc J 2009;180:32–9. [55] García AI, Milinkovic A, Tomás X, et al. MRI signal changes of the bone marrow in HIV-infected patients with lipodystrophy: correlation with clinical parameters. Skeletal Radiol 2011;40:1295–301. [56] Georgiou KR, Hui SK, Xian CJ. Regulatory pathways associated with bone loss and bone marrow adiposity caused by aging, chemotherapy, glucocorticoid therapy and radiotherapy. Am J Stem Cells 2012;1:205–24.
Please cite this article in press as: Hardouin P, et al. Bone marrow fat. Joint Bone Spine (2014), doi:10.1016/j.jbspin.2014.02.013
G Model BONSOI-4001; No. of Pages 7
ARTICLE IN PRESS P. Hardouin et al. / Joint Bone Spine xxx (2014) xxx–xxx
[57] Duque G, Li W, Adams M, et al. Effects of risedronate on bone marrow adipocytes in postmenopausal women. Osteoporos Int J Establ Result Coop Eur Found Osteoporos Natl Osteoporos Found USA 2011;22:1547–53. [58] Duque G, Li W, Vidal C, et al. Pharmacological inhibition of PPAR␥ increases osteoblastogenesis and bone mass in male C57BL/6 mice. J Bone Miner Res 2013;28:639–48.
7
[59] Iyer S, Ambrogini E, Bartell SM, et al. FOXOs attenuate bone formation by suppressing Wnt signaling. J Clin Invest 2013;123:3409–19. [60] Tseng P-C, Hou S-M, Chen R-J, et al. Resveratrol promotes osteogenesis of human mesenchymal stem cells by upregulating RUNX2 gene expression via the SIRT1/FOXO3A axis. J Bone Miner Res 2011;26:2552–63.
Please cite this article in press as: Hardouin P, et al. Bone marrow fat. Joint Bone Spine (2014), doi:10.1016/j.jbspin.2014.02.013
ARTICLE IN PRESS Joint Bone Spine xxx (2014) xxx–xxx
Available online at www.sciencedirect.com
Review
Bone marrow fat Pierre Hardouin a,b,c,∗ , Vittorio Pansini a,d,e,f , Bernard Cortet a,d,e,g a
Université Lille Nord-de-France, 59000 Lille, France Université du Littoral Côte-d’Opale PMOI, boulevard Napoléon, BP 120, 62327 Boulogne-sur-Mer, France c PMOI EA 4490, 62327 Boulogne-sur-Mer, France d PMOI EA 4490, 59000 Lille, France e Université droit et santé Lille 2, 59000 Lille, France f Service d’imagerie musculo-squeletique, CHRU, 59000 Lille, France g Service de rhumatologie, CHRU, 59000 Lille, France b
a r t i c l e
i n f o
Article history: Accepted 6 January 2014 Available online xxx Keywords: Bone marrow Adipocyte Osteoblast Fat Osteoporosis Anorexia nervosa
a b s t r a c t Bone marrow fat (BMF) results from an accumulation of fat cells within the bone marrow. Fat is not a simple filling tissue but is now considered as an actor within bone microenvironment. BMF is not comparable to other fat depots, as in subcutaneous or visceral tissues. Recent studies on bone marrow adipocytes have shown that they do not appear only as storage cells, but also as cells secreting adipokines, like leptin and adiponectin. Moreover bone marrow adipocytes share the same precursor with osteoblasts, the mesenchymal stem cell. It is now well established that high BMF is associated with weak bone mass in osteoporosis, especially during aging and anorexia nervosa. But numerous questions remain discussed: what is the precise phenotype of bone marrow adipocytes? What is the real function of BMF, and how does bone marrow adipocyte act on its environment? Is the increase of BMF during osteoporosis responsible for bone loss? Is BMF involved in other diseases? How to measure BMF in humans? A better understanding of BMF could allow to obtain new diagnostic tools for osteoporosis management, and could open major therapeutic perspectives. © 2014 Société franc¸aise de rhumatologie. Published by Elsevier Masson SAS. All rights reserved.
The organ “bone” is not only composed of bone tissue, but also includes bone marrow, which is located between bone trabeculae of spongious bone, and inside the diaphysis of long bones. Bone marrow is called “red marrow” when it is mainly filled with haematopoietic cells, and “yellow marrow” when it contains a majority of adipocytes. The existence of fat inside bones is known since hundred years, but it is much more recently that bone marrow fat (BMF) aroused the interest of researchers, as illustrated by the increasing number of publications over the last fifteen years in the database “PubMed” (Fig. 1). Indeed BMF does not appear any more as a simple filling tissue, but is considered from now as an actor within bone microenvironment. Due to these recent works, BMF is known better and better and its involvement in bone physiopathology is more and more argued. However numerous questions remain debated.
∗ Corresponding author. E-mail address: [email protected] (P. Hardouin).
1. Bone Marrow Fat (BMF): main characteristics BMF results from an accumulation of fat cells within bone marrow. These adipocytes contain a big lipid vacuole of triglycerides made of fatty acids, which can be saturated, mono or polyunsaturated.
1.1. Variations with age At birth, bone cavities are mainly filled with red hematopoietic marrow. Then occurs during childhood a “conversion” of the red marrow which is gradually replaced by yellow, fat marrow. This conversion of bone marrow begins in terminal phalanges after birth, then goes forward by a centripetal evolution up to the axial skeleton [1]. So, at the age of 25, red marrow is limited to the axial skeleton, ribs and breastbone. A “reconversion” can be observed in conditions of hypoxia, such as in smokers or in patients with obstructive sleep apnoea. Although large individual variations are found, it exists globally a positive correlation between BMF and age [2,3].
doi:10.1016/j.jbspin.2014.02.013 1297-319X/© 2014 Société franc¸aise de rhumatologie. Published by Elsevier Masson SAS. All rights reserved.
Please cite this article in press as: Hardouin P, et al. Bone marrow fat. Joint Bone Spine (2014), doi:10.1016/j.jbspin.2014.02.013
G Model
ARTICLE IN PRESS
BONSOI-4001; No. of Pages 7
P. Hardouin et al. / Joint Bone Spine xxx (2014) xxx–xxx
2 Bone Marrow fat (Medline)
250 200 150 100 50 0 1998
2000
2002
2004
2006
2008
2010
2012
Fig. 1. Annual number of publications referenced in the PubMed database between 1998 and 2012 with keywords “bone marrow fat”.
1.2. Location and quantity In adults, yellow bone marrow is mainly located in the appendical skeleton. Its volume is approximately estimated at 7% of total fat [3] and so 2,6 kg in an adult, which represents a storage of about 23,000 calories [4]. However important variations are found, not only with age as previously seen, but also according to gender (BMF is higher in men than in women) and to anatomical location (BMF is higher in long bones diaphysis than in axial skeleton) [1,5,6]. In a given long bone, yellow marrow is preferentially located in the diaphysis and in the epiphyses, whereas red marrow is mainly localized in the metaphysis [1]. Moreover, quantitative modifications of BMF can be observed in several diseases, especially in osteoporosis [5–7]. 1.3. Microscopic aspect Bone marrow adipocytes are not grouped in lobules as in the other fat depots, but they are scattered within the hematopoietic tissue. Mean diameter of these adipocytes is around 50 , which is lower than the diameter of subcutaneous or visceral adipocytes [8,9], but great variations are found in literature according to the location. In usual optic microscopy of undecalcified bone, the technical preparation of the sample removes the lipid content, then the shapes of adipocytes appear as cellular “ghosts” (Fig. 2). Bone marrow adipocytes can be also studied with confocal microscopy or on frozen slides or by microtomography after labelling with osmium. 1.4. In vivo imaging Measurement of BMF can be made by X-rays quantitative tomography but the geometrical resolution of this method is weak and a preferential loss of photons with low energy leads to underestimation of the quantity of fat in long bones [10]. Magnetic resonance imaging (MRI) is the best method to distinguish yellow fat marrow from haematopoietic red marrow. Yellow marrow appears in hypersignal in T1-weighted sequences having intermediate to elevated signal on spin-echo T2 images and on T2 fast spin-echo images and low signal on sequences with fat saturation. Yellow bone marrow does not enhance significantly after gadolinium injection. Red bone marrow shows intermediate signal on T1 or on T2 images, generally more elevated than muscular or yellow bone marrow signal. On fat-suppressed images red bone marrow shows intermediate to elevated signal, thus constantly higher than yellow bone marrow. In any case the difference of signal between the two types of bone marrow (red and yellow) is less evident on T2 images than on T1 or on STIR ones. Red bone marrow shows generally a mild enhancement after injection of gadolinium.
Fig. 2. Microscopic aspect of bone marrow fat on iliac crest biopsy in osteoporosis. Undecalcified microscopy, May Grünwald Giemsa staining (bar = 100 ): a: osteoporotic woman, 56 years old; weak amount of marrow adiposity; mean adipocyte diameter = 51,5 ; adipocytes number = 108/mm2 : b: osteoporotic woman, 68 years old; important amount of adiposity; mean adipocyte diameter = 61,9 ; adipocytes number = 170/mm2 .
Altough MRI permits to depict precisely yellow and red bone barrow, it provides no quantification of these two components. Proton magnetic resonance spectroscopy (1 H MRS) allows to quantify BMF compared to water quantity with a good reproducibility [11,12]. The obtained spectra show water and fat peaks that allows to estimate the rate of BMF (Fig. 3). The percentage of fat fraction linearly increases with age, less fast in women than in men [13], ranging from 20.5% for the second and third decades of life to 49.4% for the eighth and ninth decades [14]. MRI thus confirms the increase of BMF in elderly patients but also in osteoporotic patients. 2. Recent data on BMF 2.1. BMF is a specific fat depot It is now established that BMF constitutes a specific fat location, which is not comparable to subcutaneous and abdominal fat depots. BMF differs because of its location inside bones, which leads to interactions with the bone microenvironment, because of a more scattered distribution of fat cells, and because of its composition in fatty acids [15]. BMF is not correlated with weight, body mass index or body fat [6,16,17]. BMF is even increased in anorexia nervosa, contrasting with the meagreness of these patients [8,18], which
Please cite this article in press as: Hardouin P, et al. Bone marrow fat. Joint Bone Spine (2014), doi:10.1016/j.jbspin.2014.02.013
G Model BONSOI-4001; No. of Pages 7
ARTICLE IN PRESS P. Hardouin et al. / Joint Bone Spine xxx (2014) xxx–xxx
3
the expense of adipogenesis, and ob/ob mice, which fail to express leptin, have a low trabecular bone volume, and an increase of femoral BMF [28]. 2.4. There is a link between BMF and bone mass Pierre Meunier was the first to describe in 1971 the inverse correlation existing between BMF and age or trabecular bone volume [5]. These correlations were confirmed by numerous other studies, in animals as in humans [29,30]. In humans it is now well established that high BMF is associated with low bone mass in osteoporosis, in particular during anorexia nervosa and with aging [6,7,18]. 3. Discussed questions Bone marrow adipocyte is a specific and active cell which interacts with its environment. A good knowledge of BMF would bring us a lot, in diagnosis as in therapeutics, but several aspects remain not completely elucidated and the following list of open matters is far to be exhaustive. 3.1. What is the phenotype of marrow adipocyte?
Fig. 3. Normal MRI (1H MR spectroscopy) characteristics of bone marrow in a 28year-old woman; a: zone of interest (black square) in the femoral neck on a T1weighted oblique coronal image; b: spectra obtained in the zone of interest during the same examination. Medullar fat quantity was estimated at 73.5%.
indicates that BMF is not submitted to the same regulations as other fat locations. Bone marrow adipocytes result from a differentiation of mesenchymal stem cells, and data obtained from other adipocytes cannot be extrapolated to bone marrow cells. Functionally, numerous genes are differentially expressed in bone marrow adipocytes compared to epididymal adipocytes in mice [19] and compared to subcutaneous adipocytes in human [20]. In mice, gene profiling reveals a unique phenotype for primary bone marrow adipocytes characterized by low adipose-specific gene expression and high expression of inflammatory response genes [19]. Moreover bone marrow adipocytes are less responsive to high fat diet than epididymal adipocytes. 2.2. Bone marrow adipocyte is an active cell As other adipocytes, bone marrow adipocytes are secretory cells and have not only a simple function of storage; they are involved in secreting numerous adipokines in the bone microenvironment, in particular leptin and adiponectin. Leptin can be secreted in greater quantity, at least in vitro, by marrow adipocytes than by subcutaneous adipocytes [21]. Locally produced adiponectin may have a positive effect on bone [22]. Moreover, adipocyte lipolysis produces fatty acids suspected to have a negative effect on osteoblasts [23]. 2.3. Bone marrow adipocytes and osteoblasts have a common origin Osteoblasts and adipocytes share a common progenitor, the mesenchymal stem cell (MSC) [24]. This common origin leads to a competition between the alternative differentiation towards adipogenesis or osteogenesis. This alternative is mainly regulated by the transcription factors PPAR␥ for adipogenesis and runx2 for osteoblastogenesis, with the cooperation of numerous factors, in particular the Wnt pathway [25–27]. Leptin favours osteogenesis at
Bone marrow adipocyte remains incompletely known and at the moment we do not have a specific marker of “bone marrow adipocyte”. A debate regarding the white or brown phenotype of theses adipocytes is still open. Brown adipocytes contain several small lipid vacuoles, numerous mitochondria, and they are dedicated to heat production thanks to the presence of UCP-1 (“uncoupled protein-1”) while white adipocytes have a role in metabolic storage. Some characteristics of brown fat have been attributed to marrow adipocytes [31] but these results have to be confirmed after a rigorous selection of bone marrow adipocytes. Nevertheless, BMF is sometimes presented as an intermediate tissue between brown and white fat, the “yellow adipose tissue” [22]. 3.2. How does marrow adipocyte act on its environment? Similarly to other adipocytes bone marrow adipocytes can store or release lipids. Saturated fatty acids secreted by MSC-derived adipocytes have a negative effect on osteoblasts in coculture (lower mineralization, alkaline phosphatase activity and expression of osteogenic mRNA markers) [23]. However, the concentrations of palmitic and oleic acids necessary to exert these effects are more elevated than those physiologically released by MSC-derived adipocytes [32]. Thus it is not established that fatty acids released by marrow adipocytes play a significant role in physiology. On the other hand, MSC-derived adipocytes are capable of inducing MSC-derived osteoblasts to develop some adipocyte characteristics [33]. The whole secretome of marrow adipocytes is not currently known, and factors responsible for this possible transdifferentiation have not yet been identified. Furthermore, marrow adipocytes also secrete microvesicles which can be involved in this conversion. A paracrine secretion exists, and an increase of pro-inflammatory cytokines has been found in medullary fluids from osteoporotic patients [34], but it would be necessary to make sure that this increase results from a local production by marrow adipocytes. Finally, marrow adipocytes express RANKL which allows to promote osteoclastic differentiation [35]. 3.3. What is the precise function of BMF? The precise function of BMF is not known, what makes difficult its integration in physiopathological regulations. Some authors consider that BMF is involved in thermogenesis because marrow
Please cite this article in press as: Hardouin P, et al. Bone marrow fat. Joint Bone Spine (2014), doi:10.1016/j.jbspin.2014.02.013
G Model BONSOI-4001; No. of Pages 7
ARTICLE IN PRESS P. Hardouin et al. / Joint Bone Spine xxx (2014) xxx–xxx
4
adipocytes could possess some characteristics of brown fat [31]. An involvement in energy metabolism can also be situated at the local level, as marrow adipocytes could supply energy for bone formation or to optimize local temperature. On the secretory plan it has been recently suggested that circulating adiponectin could mainly result from BMF secretion, what could explains the “adiponectin paradox” resulting from the inverse correlation between fat mass and circulating adiponectin level [4]. However the differences observed in gene expression profile between bone marrow adipocytes and adipocytes from other origins [4,19,20] do not argue for a strong metabolic implication of marrow adipocytes, or in any case not for the typical adipocyte functions (low expression of the adipocyte specific genes PPAR␥, FABP4, perilipin). Bone marrow adipocytes express some stem cell markers, and a plasticity of theses cells is suggested. Marrow adipocytes modify the osteoblasts phenotype [33] and favour osteoclast differentiation [35]. Thus several data suggest an active role of bone marrow adipocytes on their local microenvironment. 3.4. Is the increase of BMF responsible for bone loss during osteoporosis? Osteoporosis is associated with an increase of BMF [5–7,9]. But this association does not imply that BMF is responsible for bone loss. A lot of in vitro and in vivo works have shown a link between aging, or oestrogen deficiency, and an increase of BMF. PPAR ␥2 increases with age because of a decreased level of inhibitors such as estrogens, TGF, IGF-1 or members of wnt pathway, that favours adipogenesis at the expense of osteoblastic differentiation [26]. In animals, ovariectomy increases BMF [36], which is reversible with estrogens administration. In humans, the increase of BMF begins before the third decade, which seems premature for a simple compensatory filling of bone loss, and bone acquisition in young subjects is conversely correlated to modifications of BMF [3,17]. BMF is strongly negatively correlated with bone mineral density (BMD) [16,37]; correlations with BMD show that BMF is higher in osteoporotic than in osteopenic subjects, and is higher in patients with vertebral fractures than in osteoporotic patients without fracture [2]. Premenopausal osteoporotic women have substantially higher marrow adipocyte number, size and volume than controls, and these results remain significant after adjusting for age, body mass index, and bone volume [9]. Finally the increase of BMF associated with menopause is corrected by the administration of estrogens [38]. Thus, numerous arguments suggest a responsibility of BMF in bone loss, even if this involvement is not yet proved. 3.5. Is BMF involved in other bone diseases? 3.5.1. Anorexia nervosa Osteoporosis is a frequent complication of anorexia nervosa. The increase of BMF is strongly suspected to participate in bone loss in anorexia nervosa [8,18]. The increase of BMF measured with MRI is conversely correlated to BMD and is associated with a decrease of bone formation. The increase of the circulating level of Pref1, a preadipocyte factor which regulates adipogenesis, could explain this increase of BMF [39]. In animals, caloric restriction leads to high marrow adiposity [40], which is blunted by leptin administration [41]. 3.5.2. Disuse osteoporosis Patients with disuse osteoporosis present MRI abnormalities of bone marrow with presence of subchondral lobules of fat [42]. On the opposite mechanical strains inhibit PPAR␥, and athletes involved in weight-bearing impact sports have lower BMF when
compared with athletes involved in non impact loading sports, and with non athletic controls [43]. 3.5.3. Femoral head osteonecrosis In femoral head osteonecrosis there is a hypertrophy of bone marrow adipocytes which could favour secondary vascular blood perturbations. Moreover, the main risk factors for femoral head osteonecrosis, alcohol and glucocorticoids, increase adipocyte differentiation. In the same way, femoral heads with the highest amount of BMF before treatment seem to have the highest risk of necrosis during glucocorticoids administration [44]. 3.6. Is BMF involved in non-osseous diseases? 3.6.1. Obesity and metabolic diseases During obesity and metabolic diseases an increase of BMF is likely [29] but few studies are available on this subject, and this notion has to be confirmed, especially as the increase of BMF noticed in animal models of type 1 diabetes was not found in humans [45]. The amount of BMF is not correlated to metabolic risk in young healthy patients [17]. 3.6.2. Intervertebral disc pathologies The vertebral bone marrow is involved in intervertebral disc nutrition, and a fat conversion is suspect to participate in disc degeneration [46]. 3.6.3. Haematological diseases Marrow adipocytes, and the balance between adipocytes and osteoblasts, control the haematopoietic activity of haematopoietic stem cells. Marrow adipocyte is a negative regulator of haematopoiesis [47], and may influence the progression of chronic myeloid leukaemia [48]. The balance between adipocytes and osteoblasts also plays an important role in haematopoietic recovery after radiotherapy. 3.6.4. Fat embolism We only mention the particular case of fat embolism that results from vascular obstruction by fat droplets stemming from bone marrow, especially after long bone fracture, fracture nailing or vertebroplasty. Fig. 4 summarizes the potential roles of marrow adipocytes, in normal situation and in pathology. 3.7. How to evaluate BMF in patients? MRI appears to be the best method for measuring BMF in humans. However various equipments and methods are described in literature, which may lead to discrepancies, favoured by the heterogeneity of BMF [49]. 1 H MRS (12) is the method of choice for measuring the water/lipid ratio, and so to evaluate BMF [13]. The chemical fat fraction is poorly correlated with histological measurements but the latter measures the adipocyte volume while spectroscopy separates fat and water to measure the chemical fraction of fat [50]. MRI spectroscopy also allows studying lipid saturation (presence and type of hydrogen bindings) within the bone marrow. The obtained spectrum contains peaks corresponding to water, saturated lipids, unsaturated lipids and residual lipids [51], and the prevalence of fragility fractures is associated with a lower unsaturation levels. The application of magnetic resonance spectroscopy to quantify marrow fat is limited because of the non homogeneous distribution of BMF in bones. “Whole body” BMF acquisition (10 mm thick axial images at 40 mm intervals from fingers to toes) has been proposed to overcome this limitation and to
Please cite this article in press as: Hardouin P, et al. Bone marrow fat. Joint Bone Spine (2014), doi:10.1016/j.jbspin.2014.02.013
G Model BONSOI-4001; No. of Pages 7
ARTICLE IN PRESS P. Hardouin et al. / Joint Bone Spine xxx (2014) xxx–xxx
5
Fig. 4. Summary of the potential roles of bone marrow adipocytes, in normal and pathological conditions. OB: osteoblasts; OC: osteoclast; MSC: mesenchymal stem cell; HSC: haematopoietic stem cell.
allow evaluation of interaction between fat and bone independent of subcutaneous or visceral adipose tissue by adjustment for body composition [3]. The study of bone marrow perfusion is another interesting new approach [52]. However there is currently no standardized protocol for BMF imaging.
3.8. Is the evaluation of BMF relevant to estimate bone fracture risk? A negative correlation is found between BMF and BMD, even after adjustment for age and body fat. This finding does not mean that BMF is an independent variable. Also vertebral 1 H MRS percent fat fraction is higher in subjects with MR findings of weakened bone compared to control group [14], and higher BMF is associated with prevalent fracture in men, even after adjustment for BMD [53]. In postmenopausal women, patients with fragility fractures or type 2 diabetes have lower unsaturation levels and higher saturation levels, that suggests that 1 HMRS of spinal bone marrow may therefore serve as a novel tool for BMD-independent fracture risk assessment [51]. Besides, an increase of BMF, as observed in osteoporotic patients, modifies the measurement of bone mineral content with DEXA (dual-energy X-ray absorptiometry), but this modification seems minor for fracture risk assessment.
3.9. What are the therapeutic implications? Considering the links between the two parameters, factors modifying BMF should have an impact on bone mass. Adipogenesis is mainly under the control of PPAR␥2 [26] and long-term use of thiazolidinediones, which are PPAR␥ agonists proposed in the treatment of type 2 diabetes, reduces lumbar spine BMD and doubles the risk fracture among women with type 2 diabetes [54]. An increase of BMF is also observed with long-term use of
glucocorticoids which favours in vitro adipogenesis, as well as methotrexate, antiproteases, or after radiotherapy [55,56]. On the opposite, a decrease of BMF could be a strategy in the aim of an increase of bone mass. Indeed, a decrease of marrow adipogenesis is observed in postmenopausal women treated with risedronate [57] or with estrogens [38]. In rodents, a decrease of BMF is also observed after administration of strontium ranelate, or after PTH administration. In the future, new drugs against bone loss could be specifically designed to target BMF. In this aim, the inhibition of PPAR␥ in mice indeed increases bone mass [58]. 4. Conclusions and perspectives More and more findings argue for an involvement of BMF in bone physiopathology, and BMF is no more considered as a simple filling tissue. Bone marrow adipocyte is a specific adipocyte which interacts with bone and haematopoietic cells, and there is an inverse correlation between BMF and bone mass. But numerous unsolved questions remain, and BMF is probably not a simple antagonist of bone formation. In addition to a physiological role in the bone microenvironment, which remains to be clarified it is possible that the pathological increase of BMF results from a defence process against stress, and especially against oxidative stress [59,60]. This mechanism, not specific to bone, would lead in this particular tissue to a vicious circle by interaction with bone cells, ending in bone loss. Anyway, a better understanding of BMF could give us new diagnosis or prognosis criteria, and could open major therapeutic perspectives. Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
Please cite this article in press as: Hardouin P, et al. Bone marrow fat. Joint Bone Spine (2014), doi:10.1016/j.jbspin.2014.02.013
G Model BONSOI-4001; No. of Pages 7
ARTICLE IN PRESS P. Hardouin et al. / Joint Bone Spine xxx (2014) xxx–xxx
6
Acknowledgements Authors thank Anne Cotten, Séverine Delplace and Stéphanie Lucas for their help. References [1] Vande Berg BC, Malghem J, Lecouvet FE, et al. Magnetic resonance imaging of normal bone marrow. Eur Radiol 1998;8:1327–34. [2] Wehrli FW, Hopkins JA, Hwang SN, et al. Cross-sectional study of osteopenia with quantitative MR imaging and bone densitometry. Radiology 2000;217:527–38. [3] Shen W, Chen J, Gantz M, et al. MRI-measured pelvic bone marrow adipose tissue is inversely related to DXA-measured bone mineral in younger and older adults. Eur J Clin Nutr 2012;66:983–8. [4] Scheller E, Cawthorn W, Learman B, et al. The metabolic nature of marrow fat: insulin signaling. CREB phosphorylation and the « adiponectin paradox ». J Bone Miner Res 2013:S241 [ASBMR 2013 Annual Meeting]. [5] Meunier P, Aaron J, Edouard C, et al. Osteoporosis and the replacement of cell populations of the marrow by adipose tissue. A quantitative study of 84 iliac bone biopsies. Clin Orthop Rel Res 1971;80:147–54. [6] Justesen J, Stenderup K, Ebbesen EN, et al. Adipocyte tissue volume in bone marrow is increased with aging and in patients with osteoporosis. Biogerontology 2001;2:165–71. [7] Verma S, Rajaratnam JH, Denton J, et al. Adipocytic proportion of bone marrow is inversely related to bone formation in osteoporosis. J Clin Pathol 2002;55:693–8. [8] Abella E, Feliu E, Granada I, et al. Bone marrow changes in anorexia nervosa are correlated with the amount of weight loss and not with other clinical findings. Am J Clin Pathol 2002;118:582–8. [9] Cohen A, Dempster DW, Stein EM, et al. Increased marrow adiposity in premenopausal women with idiopathic osteoporosis. J Clin Endocrinol Metab 2012;97:2782–91. [10] Di Iorgi N, Rosol M, Mittelman SD, et al. Reciprocal relation between marrow adiposity and the amount of bone in the axial and appendicular skeleton of young adults. J Clin Endocrinol Metab 2008;93:2281–6. [11] Schellinger D, Lin CS, Lim J, et al. Bone marrow fat and bone mineral density on proton MR spectroscopy and dual-energy X-ray absorptiometry: their ratio as a new indicator of bone weakening. AJR Am J Roentgenol 2004;183:1761– 5. [12] Li X, Kuo D, Schafer AL, et al. Quantification of vertebral bone marrow fat content using 3 Tesla MR spectroscopy: reproducibility, vertebral variation, and applications in osteoporosis. J Magn Reson Imaging 2011;33:974– 9. [13] Pansini V, Monnet A, Salleron J, et al. 3 Tesla (1) H MR spectroscopy of hip bone marrow in a healthy population, assessment of normal fat value and influence of age and sex. J Magn Reson Imaging 2014;39:369–76. [14] Schellinger D, Lin CS, Hatipoglu HG, et al. Potential value of vertebral proton MR spectroscopy in determining bone weakness. AJNR Am J Neuroradiol 2001;22:1620–7. [15] Griffith JF, Yeung DKW, Ahuja AT, et al. A study of bone marrow and subcutaneous fatty acid composition in subjects of varying bone mineral density. Bone 2009;44:1092–6. [16] Shen W, Scherzer R, Gantz M, et al. Relationship between MRI-measured bone marrow adipose tissue and hip and spine bone mineral density in AfricanAmerican and Caucasian participants: the CARDIA study. J Clin Endocrinol Metab 2012;97:1337–46. [17] Di Iorgi N, Mo AO, Grimm K, et al. Bone acquisition in healthy young females is reciprocally related to marrow adiposity. J Clin Endocrinol Metab 2010;95:2977–82. [18] Ecklund K, Vajapeyam S, Feldman HA, et al. Bone marrow changes in adolescent girls with anorexia nervosa. J Bone Miner Res 2010;25:298–304. [19] Liu L-F, Shen W-J, Ueno M, et al. Characterization of age-related gene expression profiling in bone marrow and epididymal adipocytes. BMC Genomics 2011;12:212. [20] Poloni A, Maurizi G, Serrani F, et al. Molecular and functional characterization of human bone marrow adipocytes. Exp Hematol 2013;41:558e2–66e2. [21] Laharrague P, Fontanilles AM, Tkaczuk J, et al. Inflammatory/haematopoietic cytokine production by human bone marrow adipocytes. Eur Cytokine Netw 2000;11:634–9. [22] Lecka-Czernik B. Marrow fat metabolism is linked to the systemic energy metabolism. Bone 2012;50:534–9. [23] Wang D, Haile A, Jones LC. Dexamethasone-induced lipolysis increases the adverse effect of adipocytes on osteoblasts using cells derived from human mesenchymal stem cells. Bone 2013;53:520–30. [24] Pittenger MF, Mackay AM, Beck SC, et al. Multilineage potential of adult human mesenchymal stem cells. Science 1999;284:143–7. [25] Sadie-Van Gijsen H, Hough FS, Ferris WF. Determinants of bone marrow adiposity: The modulation of peroxisome proliferator-activated receptor-␥2 activity as a central mechanism. Bone 2013;56:255–65. [26] Marie PJ, Saidak S. Bone marrow adipocytes and bone aging. Osteoporos Int 2013;24:S453–5. [27] Nuttall ME, Shah F, Singh V, et al. Adipocytes and the Regulation of Bone Remodeling: A Balancing Act. Calcif Tissue Int 2014;94:78–87.
[28] Hamrick MW, Pennington C, Newton D, et al. Leptin deficiency produces contrasting phenotypes in bones of the limb and spine. Bone 2004;34:376– 83. [29] Rosen CJ, Bouxsein ML. Mechanisms of disease: is osteoporosis the obesity of bone? Nat Clin Pract Rheumatol 2006;2:35–43. [30] Fazeli PK, Horowitz MC, MacDougald OA, et al. Marrow fat and bone – new perspectives. J Clin Endocrinol Metab 2013;98:935–45. [31] Krings A, Rahman S, Huang S, et al. Bone marrow fat has brown adipose tissue characteristics, which are attenuated with aging and diabetes. Bone 2012;50:546–52. [32] Lucas S, Clabaut A, Ghali O, et al. Implication of fatty acids in the inhibitory effect of human adipocytes on osteoblastic differentiation. Bone 2013;55:429– 30. [33] Clabaut A, Delplace S, Chauveau C, et al. Human osteoblasts derived from mesenchymal stem cells express adipogenic markers upon coculture with bone marrow adipocytes. Differentiation 2010;80:40–5. [34] Pino AM, Ríos S, Astudillo P, et al. Concentration of adipogenic and proinflammatory cytokines in the bone marrow supernatant fluid of osteoporotic women. J Bone Miner Res 2010;25:492–8. [35] Goto H, Hozumi A, Osaki M, et al. Primary human bone marrow adipocytes support TNF-␣-induced osteoclast differentiation and function through RANKL expression. Cytokine 2011;56:662–8. [36] Elbaz A, Rivas D, Duque G. Effect of estrogens on bone marrow adipogenesis and Sirt1 in aging C57BL/6J mice. Biogerontology 2009;10:747–55. [37] Griffith JF, Yeung DKW, Antonio GE, et al. Vertebral marrow fat content and diffusion and perfusion indexes in women with varying bone density: MR evaluation. Radiology 2006;241:831–8. [38] Syed FA, Oursler MJ, Hefferanm TE, et al. Effects of estrogen therapy on bone marrow adipocytes in postmenopausal osteoporotic women. Osteoporos Int J Establ Result Coop Eur Found Osteoporos Natl Osteoporos Found USA 2008;19:1323–30. [39] Fazeli PK, Bredella MA, Misra M, et al. Preadipocyte factor-1 is associated with marrow adiposity and bone mineral density in women with anorexia nervosa. J Clin Endocrinol Metab 2010;95:407–13. [40] Devlin MJ, Cloutier AM, Thomas NA, et al. Caloric restriction leads to high marrow adiposity and low bone mass in growing mice. J Bone Miner Res 2010;25:2078–88. [41] Devlin M, Conlon C, Van Vliet M, et al. Leptin blunts the starvation-induced increase in bone marow adiposity. J Bone Miner Res 2013:S 342 [ASBMR Annual Meeting]. [42] De Abreu MR, Wesselly M, Chung CB, et al. Bone marrow MR imaging findings in disuse osteoporosis. Skeletal Radiol 2011;40:571–5. [43] Rantalainen T, Nikander R, Heinonen A, et al. Differential effects of exercise on tibial shaft marrow density in young female athletes. J Clin Endocrinol Metab 2013;98:2037–44. [44] Vande Berg BC, Gilon R, Malghem J, et al. Correlation between baseline femoral neck marrow status and the development of femoral head osteonecrosis in corticosteroid-treated patients: a longitudinal study by MR imaging. Eur J Radiol 2006;58:444–9. [45] Slade JM, Coe LM, Meyer RA, et al. Human bone marrow adiposity is linked with serum lipid levels not T1-diabetes. J Diabetes Complications 2012;26:1– 9. [46] Liu Q, Jin L, Shen FH, et al. Fullerol nanoparticles suppress inflammatory response and adipogenesis of vertebral bone marrow stromal cells-a potential novel treatment for intervertebral disc degeneration. Spine 2013;13:1571– 80. [47] Naveiras O, Nardi V, Wenzel PL, et al. Bone marrow adipocytes as negative regulators of the haematopoietic microenvironment. Nature 2009;460:259– 63. [48] Belaid-Choucair Z, Lepelletier Y, Poncin G, et al. Human bone marrow adipocytes block granulopoiesis through neuropilin-1-induced granulocyte colony-stimulating factor inhibition. Stem Cells Dayt Ohio 2008;26:1556– 64. [49] Shen W, Gong X, Weiss J, et al. Comparison among T1-weighted magnetic resonance imaging, modified dixon method, and magnetic resonance spectroscopy in measuring bone marrow fat. J Obes 2013;2013:75–2986. [50] Pichardo JC, Milner RJ, Bolch WE. MRI measurement of bone marrow cellularity for radiation dosimetry. J Nucl Med 2011;52:1482– 9. [51] Patsch JM, Li X, Baum T, et al. Bone marrow fat composition as a novel imaging biomarker in postmenopausal women with prevalent fragility fractures. J Bone Miner Res 2013;28:1721–8. [52] Griffith JF, Yeung DKW, Leung JCS, et al. Prediction of bone loss in elderly female subjects by MR perfusion imaging and spectroscopy. Eur Radiol 2011;21:1160–9. [53] Schwartz AV, Sigurdsson S, Hue TF, et al. Vertebral bone marrow fat associated with lower trabecular BMD and prevalent vertebral fracture in older adults. J Clin Endocrinol Metab 2013;98:2294–300. [54] Loke YK, Singh S, Furberg CD. Long-term use of thiazolidinediones and fractures in type 2 diabetes: a meta-analysis. CMAJ Can Med Assoc J 2009;180:32–9. [55] García AI, Milinkovic A, Tomás X, et al. MRI signal changes of the bone marrow in HIV-infected patients with lipodystrophy: correlation with clinical parameters. Skeletal Radiol 2011;40:1295–301. [56] Georgiou KR, Hui SK, Xian CJ. Regulatory pathways associated with bone loss and bone marrow adiposity caused by aging, chemotherapy, glucocorticoid therapy and radiotherapy. Am J Stem Cells 2012;1:205–24.
Please cite this article in press as: Hardouin P, et al. Bone marrow fat. Joint Bone Spine (2014), doi:10.1016/j.jbspin.2014.02.013
G Model BONSOI-4001; No. of Pages 7
ARTICLE IN PRESS P. Hardouin et al. / Joint Bone Spine xxx (2014) xxx–xxx
[57] Duque G, Li W, Adams M, et al. Effects of risedronate on bone marrow adipocytes in postmenopausal women. Osteoporos Int J Establ Result Coop Eur Found Osteoporos Natl Osteoporos Found USA 2011;22:1547–53. [58] Duque G, Li W, Vidal C, et al. Pharmacological inhibition of PPAR␥ increases osteoblastogenesis and bone mass in male C57BL/6 mice. J Bone Miner Res 2013;28:639–48.
7
[59] Iyer S, Ambrogini E, Bartell SM, et al. FOXOs attenuate bone formation by suppressing Wnt signaling. J Clin Invest 2013;123:3409–19. [60] Tseng P-C, Hou S-M, Chen R-J, et al. Resveratrol promotes osteogenesis of human mesenchymal stem cells by upregulating RUNX2 gene expression via the SIRT1/FOXO3A axis. J Bone Miner Res 2011;26:2552–63.
Please cite this article in press as: Hardouin P, et al. Bone marrow fat. Joint Bone Spine (2014), doi:10.1016/j.jbspin.2014.02.013
