Volume 163, Issue 7 p. 1441-1446
Free Access

Skeletal lipidomics: regulation of bone metabolism by fatty acid amide family

Itai Bab

Corresponding Author

Itai Bab

Bone Laboratory

These authors contributed equally to writing this review.

Itai Bab, Bone Laboratory, The Hebrew University of Jerusalem, PO Box 12272, Jerusalem 91120, Israel. E-mail: [email protected]Search for more papers by this author
Reem Smoum

Reem Smoum

Bone Laboratory

Institute of Drug Research, the Hebrew University of Jerusalem, Jerusalem, Israel

These authors contributed equally to writing this review.

Search for more papers by this author
Heather Bradshaw

Heather Bradshaw

Indiana University Program in Neuroscience, Indiana University, Bloomington, IN, USA

Search for more papers by this author
Raphael Mechoulam

Raphael Mechoulam

Institute of Drug Research, the Hebrew University of Jerusalem, Jerusalem, Israel

Search for more papers by this author
First published: 10 May 2011
Citations: 24


There is increasing evidence demonstrating that fatty acid derivatives play a key regulatory role in a variety of tissues. However, the study of skeletal lipidomics is just emerging and global strategies, such as targeted lipidomics, have not been applied to bone tissue. Such strategies hold great promises as in the case of genomics and proteomics. A partial profile of endocannabinoids and endocannabinoid-like compounds has demonstrated the presence of several long-chain fatty acid amides (FAAs), some of which displaying potent effects on osteoblasts, the bone forming cells and osteoclasts, the bone resorbing cells. In the skeleton, the FAAs activate the CB1 cannabinoid receptor present in sympathetic nerve terminals as well as CB2 cannabinoid receptor, the Gi-protein coupled receptor GPR55, and the transient receptor potential vanilloid type ion channel expressed by osteoblasts and/or osteoclasts. This review on the skeletal FAA system focuses on the production of FAAs in the skeleton and their net bone anabolic and anti-catabolic activity resulting from the stimulation of bone formation and inhibition of bone resorption. As the FAA family holds great promise as a basis for the treatment of osteoporosis and other diseases involving bone, further studies should aim towards the complete profiling of these lipids and their receptors in bone tissue, followed by elucidation of their function and mechanism of action.

LINKED ARTICLES This article is part of a themed issue on Cannabinoids in Biology and Medicine. To view the other articles in this issue visit http://dx.doi.org/10.1111/bph.2011.163.issue-7


  • BMD
  • bone mineral density
  • CB1
  • cannabinoid receptor type 1
  • CB2
  • cannabinoid receptor type 2
  • CFU-Ob
  • colony forming unit osteoblastic
  • EC
  • endocannabinoid
  • FAA
  • fatty acid amide
  • FAAH
  • fatty acid amide hydrolase
  • GPR55
  • Gi-protein coupled receptor 55
  • long-chain polyunsaturated fatty acid
  • NE
  • norepinephrine
  • OPG
  • osteoprotegerin
  • OS
  • oleoyl serine
  • OVX
  • ovariectomy
  • PTH
  • parathyroid hormone
  • receptor activator of NFκB ligand
  • TRPV1
  • transient receptor potential vanilloid type 1
  • Introduction

    In vertebrates, skeletal metabolism is reflected as a continuous process of bone renewal, known better as bone remodelling, which consists of the concerted and balanced action of osteoclasts, the bone resorbing cells, and osteoblasts, the bone forming cells. Osteoporosis, the most prevalent degenerative disease in developed countries, results from the impairment of this balance, leading to bone loss and increased fracture risk. Bone remodelling is regulated by a complex convergence of circulating hormones including sex steroids, parathyroid hormone, neurotransmitters, neuropeptides and pituitary-derived thyroid and follicle-stimulating hormones, on one hand, and local regulators of bone cell activity such as bone morphogenetic proteins, receptor activators of nuclear factor κB ligand (RANKL) and a number of cytokines, on the other hand (Rosen, 1997; Takeda et al., 2002; Abe et al., 2003; Robling et al., 2006; Sun et al., 2006; Lee and Herzog, 2009).

    The super-family of lipids includes several families of naturally occurring compounds such as fats; cholesterol; fat-soluble vitamins; phospholipids; mono-, di- and triacylglycerols; as well as fatty acids and their derivatives. The biological function of lipids is not restricted to energy storage or as structural determinants of the cell membrane. It is well established now that lipids also serve as signalling molecules important in several physiological and pathological processes (Connor et al., 2010). Yet the structural diversity of lipid compounds, their complex purification procedures and the chemical instability of many lipid entities decelerated their reliable identification and quantification. Recently, however, considerable progress in lipid analytical techniques has been made using advanced liquid chromatography combined with various MS systems, which made possible lipid analysis even in extremely complex mixtures. Based on these new technologies, the term ‘lipidomics’ was coined to define ‘the full characterization of lipid molecular species and of their biological roles with respect to the expression of proteins involved in lipid metabolism and function, including gene regulation’ (Spener et al., 2003). As part of lipidomics, a remarkably growing attention has been channelled to the endocannabinoid system, as a novel paradigm to enhance the understanding of a particular family of lipids as signalling molecules (Astarita et al., 2009). This family of fatty acid derivatives includes compounds consisting of saturated and unsaturated (ω-3, ω-6, ω-7, ω-9) long-chain fatty acids amides (FAAs). The identification of members of this subclass has been carried out concurrently using advances in MS techniques. Lipidomic approaches have been successfully applied to the endocannabinoid system (Piomelli et al., 2007) and related FAAs, mainly in the central nervous system.

    In spite of a growing interest in the circumstantial and functional inter-relationship between bone and adiposity, so far the study of lipid signalling in bone has focused primarily on prostaglandin E2 (Uppal et al., 2008). In addition, there is increasing evidence that various long-chain polyunsaturated fatty acids (LCPUFAs), as well as non-prostanoid LCPUFA metabolites, also regulate bone metabolism and may have therapeutic potential in the management of osteoporosis. Modification of dietary LCPUFA content, particularly increasing the intake of n-3 LCPUFAs, has been shown to minimize the decline in bone mass caused by menopause in women and by ovariectomy in animal models (Jee and Ma, 1997).

    Since the initial publications of the regulation of skeletal metabolism by the endocannabinoid (EC) system half a decade ago (Bab, 2005; Idris et al., 2005; Karsak et al., 2005; Ofek et al., 2006), a substantial body of information has been accumulated on the role of the FAA family in the control of bone remodelling and bone mass. The present article reviews this information, including recent publications on the presence of FAAs in the skeleton and role of a few of them in bone renewal.

    Presence of FAAs in bone tissue

    Several members of the FAA family have already been identified in bone tissue (Table 1). N-oleoyl serine (OS), N-oleoyl ethanolamide, N-stearoyl ethanolamide and the EC arachidonoyl ethanolamide (anandamide) are present in bone in pmol/g concentrations (Smoum et al., 2010, Table 1). A structurally and functionally related compound, the EC 2-arachidonoylglycerol (2-AG), is present in bone at nmol/g levels. Notably, the EC bone levels are similar to those found in the brain (Bab et al., 2008), but the blood EC levels are several orders of magnitude lower. Hence, it is very likely that these compounds, and other FAAs alike, are synthesized locally in the skeleton (Bab et al., 2008). Indeed, at least the ECs and OS are produced by bone marrow stromal cells (that include osteoblast progenitors), osteoblasts and osteoclasts in culture (Jiang et al., 2011; and our unpubl. results). A multitude of fatty acids is present in bone marrow (Griffith et al., 2009) and it has been shown that brain N-acyl ethanolamide levels are correlated with dietary 20:4n-6 and 22:6n-3 fatty acids (Berger et al., 2001). Hence, long-chain fatty acids may serve as precursors of skeletal FAAs. In addition, diacylglycerol lipases α and β, enzymes critically involved in 2-AG biosynthesis, are expressed in osteoblasts, osteocytes, bone-lining cells and osteoclasts (Tam et al., 2008). The FAA biosynthetic and degrading enzymes, N-acyl phosphatidylethanolamine phospholipase D and fatty acid amide hydrolase (FAAH) (Ueda et al., 2010), are also expressed in bone cells (Rossi et al., 2009; our unpubl. results). A recent targeted lipidomics profiling of FAAs in brain revealed the presence of many more such compounds (Tan et al., 2009). Therefore, given the similarity in FAA content so far found in bone and brain, the occurrence of multiple FAAs is anticipated also in bone.

    Table 1. Skeletal FAAs and their functional features

    FAA receptors and bone metabolism

    The known FAA skeletal targets are shown in Table 1. The initial findings pointing to a role for FAAs in bone metabolism include the expression of the CB1 and CB2 cannabinoid receptors in the skeleton and their activity in regulating bone formation by osteoblasts and bone resorption by osteoclasts (Bab, 2005; Idris et al., 2005; Ofek et al., 2006). More recently, another G-protein coupled receptor, GPR55, has also been demonstrated in these cells (Whyte et al., 2009). In addition, the membrane ion channel TRPV1 has been reported in osteoclasts (Rossi et al., 2009).

    Work carried out so far has focused mainly on CB1 and CB2. In the osteogenic cell lineage, undifferentiated osteoprogenitor cells exhibit very low levels, if any, of CB1. CB2 mRNA expression in these cells is also very low (Bab, 2005; Ofek et al., 2006). However, when these cells are allowed to differentiate for 2–4 weeks in ‘osteogenic medium’ (Bellows et al., 1986), CB2 mRNA expression increases progressively together with the expression of osteoblastic marker genes, e.g. tissue non-specific alkaline phosphatase (Zhou et al., 1994), parathyroid hormone receptor 1 (Zhang et al., 1995), and the osteoblastic master regulatory gene, RUNX2 (Araujo et al., 2004). CB1 is expressed at low levels in monocytic cells undergoing osteoclastogenesis induced by RANKL and macrophage colony-stimulating factor (Zou et al., 2002). By contrast, CB2 mRNA transcripts in these cells are present in high abundance (Bab, 2005; Idris et al., 2005; Ofek et al., 2006; Scutt and Williamson, 2007). CB2 is also expressed by periodontal ligament cells, which may undergo osteoblastic differentiation (Qian et al., 2010). In vivo, CB2 was identified in trabecular osteoblasts and their descendants, the osteocytes, as well as in osteoclasts (Ofek et al., 2006). CB1 protein is abundant in skeletal sympathetic nerve terminals in close proximity to osteoblasts (Tam et al., 2006).

    Several mouse lines of FAA receptor null mice have been used to assess the physiologic role of CB1, CB2 and GPR55 in the control of bone metabolism. The skeletal phenotype of cnr1 (the gene encoding mouse CB1) mutated mice depends on the animal strain and/or the construct used for gene mutation. In one CB1-deficient line, backcrossed to CD1 mice (CD1cnr1–/–), the N-terminal 233 codons of cnr1 were ablated (Ledent et al., 1999). The skeletal phenotype of young, sexually mature CD1cnr1–/– mice shows a sex disparity. Females have normal trabecular bone with a slight cortical expansion, whereas males exhibit a high bone mass phenotype (Idris et al., 2005; Tam et al., 2006). Sexually mature young mice of either sex display normal bone formation and resorption parameters, suggesting that the male phenotype is acquired early in life. In the second line, backcrossed to C57Bl/6J mice (C57cnr1–/–), almost the entire protein-encoding sequence was removed (Zimmer et al., 1999). Both male and female C57cnr1–/– have a low bone mass phenotype accompanied by increased osteoclast counts and decreased bone formation rate (Tam et al., 2006). More recently, an aging-related low bone mass phenotype has also been reported in a CD1cnr1–/–-derived mouse line (Idris et al., 2009).

    CB2-deficient animals have a skeletal phenotype that is gender independent. Both male and female cnr2–/– mice accrue a normal peak trabecular bone mass, but later display a markedly enhanced age-related bone loss (Ofek et al., 2006). Reminiscent of human postmenopausal osteoporosis (Brown et al., 1984), the cnr2–/– mice have a high bone turnover characterized by increases in both bone resorption and formation, which are at a net negative balance (Ofek et al., 2006). Because healthy cnr2 null mice are otherwise normal, it appears that the main physiologic role of CB2 is in maintaining bone remodelling at balance.

    Studies in humans have confirmed that CB2 is an important determinant of bone metabolism. The locus of CNR2 (the gene encoding human CB2) is located on chromosome 1p36. This genomic region and its mouse ortholog on chromosome 4 have been linked to bone mineral density (BMD) and osteoporosis in several association analyses (Devoto et al., 1998; 2001; 2005). Several genetic association studies have consistently shown that a common variant of CNR2 contributes to the aetiology of low BMD and osteoporosis in humans (Karsak et al., 2005; 2009; Yamada et al., 2007; Huang et al., 2009). So far, similar analyses of CNR1 (the gene encoding human CB1) have failed to demonstrate such an association with osteoporosis.

    GPR55 expression has been reported in osteoblasts and osteoclasts. While the skeleton of female GPR55 null nice is normal, male GPR55–/– animals have a high bone mass phenotype, resulting from reduced bone resorption, secondary to an osteoclast malfunction. On the other hand, GPR55 agonists also restrain bone resorption, consequent to the inhibition of osteoclast formation (Whyte et al., 2009).

    Signalling pathways targeted by FAAs in bone have been so far reported for CB1, CB2 and the putative OS receptor. It appears that CB1 controls osteoblast function by negatively regulating norepinephrine (NE) release from sympathetic nerve terminals in the immediate proximity of these cells. NE suppresses osteoblast function by binding to osteoblastic β2-adrenergic receptors (Takeda et al., 2002), which is alleviated by activation of sympathetic CB1 (Tam et al., 2008). Although expressed at a low level in osteoblasts and osteoclasts, the relative contribution of bone cell versus sympathetic CB1 to bone metabolism is still an open issue. Given that cnr1flox/flox mice are already available (Quarta et al., 2010), this question should be addressed by phenotypic characterization of mice with conditional cnr1 deletion in osteoblasts, osteocytes, osteoclasts and sympathetic nerves.

    Activation of CB2 in osteoblasts targets a mitogenic Gi protein – Erk1/2 – Mapkapk2 – CREB – cyclin D1 pathway (Ofek et al., 2011). CB2 agonists also stimulate mineralized matrix formation (Ofek et al., 2006). In addition, CB2 activation leads to decreased osteoclastogenesis resulting from decreases in osteoclast progenitor cell proliferation and in the RANKL/OPG expression ratio in bone marrow stromal cells (Ofek et al., 2006).

    Although the putative OS receptor has not been identified yet, it is likely coupled to a Gi protein, as its actions are inhibitable by pertussis toxin. It also activates Erk1/2, but unlike the case of CB2, its signalling downstream of Erk1/2 does not involve Mapkapk2 and CREB (Smoum et al., 2010).

    The presence of FAA receptors in bone does not derogate from the potential role of free fatty acids (FFAs) in skeletal metabolism. Indeed, receptors known to bind FFAs were found in osteoblastic (GPR120) and osteoclastic (GPR40, 41, 43, 120) cells (Cornish et al., 2008). However, the relationship between the FFA and FAA activity in bone remains an open, very fascinating issue. Additional possible FAA targets, such as ion-channels and peroxisome proliferator-activated receptors, have been identified and should be looked for in the skeleton (Hansen, 2010; Pertwee et al., 2007).

    Therapeutic potential of FAAs and related agonists

    Of the FAAs so far identified, only the skeletal effects of OS administration have been studied in vivo. OS increases bone mass in normal and osteoporotic mice by a dual action consisting of stimulating osteoblast proliferation/bone formation and inhibiting bone resorption by enhancing osteoclast apoptosis (Smoum et al., 2010). These features suggest that OS can be used as a bone antiresorptive, as well as anabolic agent.

    Because of its instability, anandamide administration has not been tested in vivo. However, like OS, synthetic CB2 agonists have been shown to be both anabolic and antiosteoporotic by stimulating bone formation and restraining bone resorption (Ofek et al., 2006; Bab et al., 2008).

    TRPV1 is targeted by many FAAs (Movahed et al., 2005). A recent study in osteoclasts derived from osteoporotic patients suggests that its desensitization by FAAs, or its enhanced trafficking, together with TRPV1 agonist-induced CB2 receptor overexpression, might be critical to minimize calcium entry into osteoclasts, which could be responsible in turn for cell over-activation and increased bone resorption and bone loss, advocating the use of TRPV1 agonists together with CB2 agonists in osteoporosis (Rossi et al., 2011).

    If indeed the dual potent bone anabolic-antiresorptive action is shared by many skeletal FAAs, inhibition of FAAH, the FAA degrading enzyme, may prove as a useful therapeutic strategy to combat osteoporosis and perhaps other skeletal deficits. Notably, CB1 or CB2 agonists reduce bone cancer pain in animal models (Curto-Reyes et al., 2010; Kawamata et al., 2010). Therefore, this approach may be applied also for analgetic treatment in patients with bone metastases.


    A handful of studies suggest that FFAs are involved in the regulation of bone remodelling, either by themselves or through their conversion to other molecules such as prostaglandins and FAAs. Recent studies demonstrate several FAAs and FAA receptors in bone tissue. Most of the FAAs identified in the skeleton stimulate osteoblast proliferation. Findings in mice and humans suggest that they also inhibit osteoclastogenessis. Of these FAAs, the occurrence and activity of OS have been characterized in more details. These data, together with skeletal phenotyping of FAA receptor (CB1, CB2, GPR55)-deficient mice, suggest an important role for the FAAs in the regulation of skeletal remodelling and the consequent implications on bone mass and biomechanical function as well as the alleviation of bone pain.


    Work on N-oleyl serine included in this article has been supported by grants from the US-Israel Binational Science Foundation (No. 2007013 to I.B and H.B.) and from the US National Institute on Drug Abuse (No. 9789 to RM).

      Conflict of interest

      We do not have any conflict of interest.