help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2008-0111
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Cornish, J.
Right arrow Articles by Reid, I. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cornish, J.
Right arrow Articles by Reid, I. R.
Endocrinology Vol. 149, No. 11 5688-5695
Copyright © 2008 by The Endocrine Society

Modulation of Osteoclastogenesis by Fatty Acids

Jillian Cornish, Alastair MacGibbon, Jian-Ming Lin, Maureen Watson, Karen E. Callon, P. C. Tong, James E. Dunford, Yvonne van der Does, Garry A. Williams, Andrew B. Grey, Dorit Naot and Ian R. Reid

Department of Medicine (J.C., J.-M.L., M.W., K.E.C., P.C.T., G.A.W., A.B.G., D.N., I.R.R.), University of Auckland, 1142 Auckland, New Zealand; Fonterra Research Centre (A.M., Y.v.d.D.), Palmerston North 4442, New Zealand; and Botnar Research Centre (J.E.D.), University of Oxford, Oxford OX1 3QX, United Kingdom

Address all correspondence and requests for reprints to: Dr. J. Cornish, Department of Medicine, University of Auckland, Private Bag 92019, Auckland, New Zealand. E-mail: j.cornish{at}auckland.ac.nz.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Clinical studies have shown that total body fat mass is related to both bone density and fracture risk and that fat ingestion reduces bone turnover. These effects are at least partially mediated by endocrine mechanisms, but it is possible that lipids might act directly on bone. We assessed the effects of broad fractions of milk lipids in osteoblasts, bone marrow, and neonatal mouse calvariae. Several milk fractions and their hydrolysates inhibited osteoclastogenesis in bone marrow cultures, so we assessed the effects of free fatty acids in this model. Saturated fatty acids (0.1–10 µg/ml) inhibited osteoclastogenesis in bone marrow cultures and RAW264.7 cells. This effect was maximal for C14:0 to C18:0 fatty acids. The introduction of greater than 1 double bond abrogated this effect; {omega}3 and {omega}6 fatty acids had comparable low activity. Osteoblast proliferation was modestly increased by the antiosteoclastogenic compounds, ruling out a nonspecific toxic effect. Active fatty acids did not consistently change expression of receptor activator of nuclear factor-{kappa}B ligand or osteoprotegerin in osteoblastic cells nor did they affect the activity of key enzymes in the mevalonate pathway. However, receptors known to bind fatty acids were found to be expressed in osteoblastic (GPR120) and osteoclastic (GPR40, 41, 43, 120) cells. A synthetic GPR 40/120 agonist mimicked the inhibitory effects of fatty acids on osteoclastogenesis. These findings provide a novel link between lipid and bone metabolism, which might contribute to the positive relationship between adiposity and bone density as well as provide novel targets for pharmaceutical and nutriceutical development.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
THERE IS A growing literature exploring the links between lipids and bone. Total body fat mass is related to bone density and to fracture risk, and changes in fat mass are positively related to changes in bone density (1). It is likely that these important relationships are mediated by adipocyte-derived hormones that act on bone (leptin, adiponectin) and by the β-cells of the pancreas, which produce increased amounts of bone anabolic hormones (insulin, amylin, preptin) in obesity (2). There is also evidence from feeding experiments that fat ingestion influences bone turnover, an effect that may be mediated by some of the factors listed above but that also involves gut hormones such as glucose-dependent insulinotropic peptide and glucagon-like peptide-2 (3, 4). A further possibility that has received little attention is that lipids might act directly on bone.

Most previous work assessing the effects of lipids on bone has focused on polyunsaturated fatty acids (PUFAs). {omega}3 PUFAs have been suggested to reduce production of inflammatory cytokines in a number of conditions, which would be expected to be associated with reduced bone loss, whereas {omega}6 PUFAs have the opposite effect (5). These actions may be mediated by changes in prostaglandin production [because PUFAs act as substrates in this process (6)], and PUFAs may also modulate nitric oxide production. There is some evidence that manipulation of PUFA intakes results in changes in osteoblast activity in vitro and bone strength in studies in intact animals [recently reviewed by Watkins et al. (6)].

Dairy products are a major source of lipids in the human diet. Milk is a rich biological fluid that provides nutrition at a time of rapid skeletal growth and development in the neonate. Because of this, it contains many growth regulators, in addition to the substrates necessary for infant development. We previously screened the protein fraction of milk for compounds that act directly on bone cells and found significant activity (7). Because of the possibility that lipids might have direct bone effects, we have now extended this approach to milk lipids. Bovine milk fat contains numerous fatty acids, ranging in length from 4 to 18 carbon atoms. The most common are palmitic (C16:0, 28% by weight), oleic (C18:1 cis, 18%), stearic (C16:0, 12%), myristic (C14:0, 11%), and butyric (C4:0, 4%) (8).

In preliminary experiments we assessed the effects of broad fractions of milk lipids on osteoclastogenesis in primary murine bone marrow cultures, osteoblast proliferation in primary rat osteoblasts, and osteoblast and osteoclast indices in mouse calvarial cultures. The latter studies showed no consistent effects, but inhibition of osteoclastogenesis in bone marrow cultures was apparent with several milk fractions, and similar or greater activity was found with their hydrolysates (data not shown). In light of these findings, we hypothesized that the activity detected could result from the action of free fatty acids released during hydrolysis and proceeded to systematically assess a range of these molecules in bone cell and tissue cultures. We also examined the expression in bone cells of four G protein-coupled receptors (GPRs) that have recently been shown to mediate biological responses to fatty acids (9). GPR40 and GPR120 specifically bind medium and long-chain fatty acids (10, 11), whereas GPR41 and GPR43 bind short-chain fatty acids (12).


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Bone marrow cultures
Bone marrow was obtained from the long bones of normal Swiss male mice aged 4–6 wk, as previously described (13). Marrow cells were placed in 90-mm petri dishes for 2 h and nonadherent cells collected and grown in 48-well plates. 1,25 Dihydroxyvitamin D3 (10–8 M) was added (d 0) to all wells except to negative controls. The cultures were fed 0.5 ml fresh medium on d 2, and 0.5 ml was replaced with fresh medium on d 4. 1,25 Dihydroxyvitamin D3 (10–8 M) was added on d 0, 2, and 4, whereas test substances were added on d 2 and 4. After culture for 7 d, multinucleated cells (≥3 nuclei) staining with tartrate-resistant acid phosphatase (TRAP) were counted. There were at least eight wells for each group and each experiment was repeated two or three times.

RAW264.7 cell cultures
RAW264.7 cells were seeded (d 0) into 48-well plates at a density of 1500 cells per 0.5 ml/well, in 10% fetal bovine serum (FBS)/{alpha}-MEM. Four hours later, a further 0.5 ml of media was added containing receptor activator of nuclear factor-{kappa}B ligand (RANKL; final concentration 10 ng/ml) plus fatty acids or their vehicle. On d 2, the media were completely replaced and RANKL, fatty acids/vehicle readded. On d 4, the cells were fixed and stained for TRAP.

To investigate the effects of fatty acid on bone resorption, RAW264.7 cells were seeded on ivory slices in 96-well plates at 650 cells/well in 10% FBS/{alpha}-MEM. Four hours later, the ivory slices were transferred to a 48-well plate, and the medium supplemented with RANKL 10 ng/ml and either stearic acid 10 µg/ml or vehicle. At 2 d, the media and test substances were replaced. At d 6, the cells were fixed and stained for TRAP. Counting of TRAP-positive cells and analysis of resorption were undertaken by an observer who was blinded to treatment allocation. After counting of TRAP-positive cells, the ivory slices were scrubbed and resorption pits visualized by staining with toluidine blue. Quantification of resorption was undertaken using a semiquantitative scoring system. No resorption within a microscopic field scored 0, 1–10% of bone area resorbed scored 1, 11–30% resorbed area scored 2, and greater than 30% scored 3.

Osteoblast-like cell culture
Osteoblasts were isolated from 20-d fetal rat calvariae, as previously described (11). Briefly, calvariae were excised and the frontal and parietal bones, free of suture and periosteal tissue, were collected. The calvariae were sequentially digested using collagenase (Sigma, St. Louis, MO), and the cells from third and fourth digests were collected, pooled, and washed. Cells were grown in T75 flasks in 10% FBS/DMEM (Invitrogen, Mount Waverley, Australia) and 5 µg/ml L-ascorbic acid 2-phosphate (Sigma, St. Louis, MO) for 2 d and then changed to 10% FBS/MEM) (Invitrogen) per 5 µg/ml L-ascorbic acid 2-phosphate and grown to 90% confluency. Cells were then seeded into 24-well plates in 5% FBS/MEM/5 µg/ml L-ascorbic acid 2-phosphate for 24 h. Cells were growth arrested in 0.1% BSA (ICP, Auckland, New Zealand) per 5 µg/ml L-ascorbic acid 2-phosphate for 24 h. Fresh media and experimental compounds were then added for a further 24 h. Cells were pulsed with [3H] thymidine 6 h before the end of the experimental incubation. The experiment was terminated and cell counts or thymidine incorporation assessed. There were 6 wells in each group and each experiment was repeated three or four times.

Bone organ culture
Mice were injected sc with 3 µCi 45Ca at 1–3 d of age, and hemicalvariae were dissected out 3 d later. Hemicalvariae were preincubated for 24 h in medium 199 with 0.1% BSA and then changed to fresh medium containing test substances or vehicle. Incubation was continued for a further 48 h. There were six to eight hemicalvariae in each group and each experiment was repeated twice.

Analysis of gene expression by real-time PCR
Osteoblast expression of RANKL and osteoprotegerin are important mechanisms for the regulation of osteoclastogenesis (14). For the study of the expression of osteoprotegerin and RANKL, primary osteoblasts from rat calvariae were prepared as above, and seeded into 6-well plates. After 24 h growth arrest, stearic acid was added in fresh media, and cells were harvested after a further 48 or 72 h for RNA extraction. The murine bone marrow stromal cell line ST2 was cultured in the same conditions. Murine preosteoblastic MC3T3-E1 cells, ST2 cells, and macrophage-osteoclast RAW264.7 cells were used for analysis of the expression of fatty acid receptors. MC3T3-E1 and RAW264.7 cells were cultured in {alpha}-MEM with 10% fetal calf serum, and ST2 cells were cultured in DMEM plus 5% fetal calf serum.

Total cellular RNA was purified using RNeasy minikit (QIAGEN, Doncastle, Australia), and genomic DNA was removed using ribonuclease-free deoxyribonuclease set (QIAGEN, Valencia, CA). cDNA was synthesized with Superscript III (Invitrogen) and subsequently used for real-time PCR. The primer-probe sets were purchased from Applied Biosystems (Foster City, CA). Multiplex PCR was performed in triplicates with FAM-labeled probes specific for the gene of interest and VIC-labeled 18S rRNA probes according to the company’s instructions, using ABI PRISM 7900HT sequence detection system (Applied Biosystems).

Culture media for all studies described above contained penicillin (100 U/ml) and streptomycin (100 µg/ml). All animal procedures were approved by the Animal Ethics Committee of our institution.

Mevalonate pathway enzymes
Farnesyl pyrophosphate synthase and geranylgeranyl pyrophosphate synthase, key enzymes in the mevalonate pathway, are targets for some factors that influence osteoclastogenesis, such as bisphosphonates (15, 16) and alkylamines (17), the latter having some structural similarity to fatty acids. Clones encoding human farnesyl pyrophosphate synthase (gi 61680822) and human geranylgeranyl pyrophosphate synthase (gi 4758430) were expressed in Escherichia coli BL21(DE3) as N terminally His6-tagged fusion protein with a TEV cleavage site as described (15, 16). Cells were lysed using a high-pressure cell disruptor, and the protein was purified to near homogeneity using Ni-NTA resin (QIAGEN). The His-tag was removed by incubation with TEV protease and gel filtration chromatography was performed using a Superdex 200 column (GE Healthcare, Applied Biosystems, Foster City, CA).

Farnesyl pyrophosphate synthase and geranylgeranyl pyrophosphate synthase were assayed by the method of Reed and Rilling (18) with modifications. For inhibition analysis 40 ng (1 pmol) of pure farnesyl pyrophosphate synthase or 80 ng (~2 pmol) of geranylgeranyl pyrophosphate synthase were assayed in buffer containing 50 mM Tris (pH 7.7), 2 mM MgCl2, 1 mM Tris (2-carboxyethyl) phosphine hydrochloride (TCEP), and 5 µg/ml BSA. Geranylgeranyl pyrophosphate synthase assay buffer additionally contained 0.2% Tween 20. 10 µl of 10 x fatty acid in methanol was added to the assays and an equivalent volume of methanol added to the uninhibited enzyme maximum rate tube. To allow for any slow binding inhibition, enzyme was preincubated with the inhibitor in an 80 µl volume. After 10 min preincubation, 20 µl of substrate containing geranylgeranyl pyrophosphate and isopentenyl pyrophosphate (IPP) (14C-IPP, 400 KBq/mmol) were added to start the reaction, giving a final volume of 100 µl and 10 µM final concentration each substrate. Assays were terminated after 5 min at 37 C by the addition of 0.2 ml of concentrated HCl/methanol (1:4) and incubated for 10 min at 37 C. The reactions were then extracted with 0.4 ml of ligroin, and after thorough mixing, the amount of radioactivity in the upper phase was determined by mixing 0.2 ml of the ligroin with 4 ml of general purpose scintillant. This was counted using a Tricarb 1900CA scintillation counter (Packard, Meriden, CT). Enzyme activity is expressed as velocity of inhibited enzyme/velocity of uninhibited enzyme x 100.

Lipid preparations and other reagents
Fatty acids were purchased from Sigma. GW9508, a synthetic GPR40/120 agonist (19), was purchased from Cayman Chemicals (Ann Arbor, MI).

Statistics
Data were analyzed using ANOVA with post hoc Dunnett’s tests for significant main effects. A 5% significance level (two tailed) was used throughout. Data are presented as means ± SE, unless indicated otherwise. Analyses were performed using Prism version 3 (graphpad.com).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Bone marrow cultures
Each fatty acid was assessed at three concentrations (0.1, 1.0, and 10 µg/ml), and representative results from a series of saturated fatty acids are shown in Fig. 1Go. Some inhibition of osteoclastogenesis was seen with each fatty acid, but this was maximal for chain lengths of 14–18 carbon atoms, particularly palmitic acid (C16:0; 50% inhibition at 10 µg/ml). The poor solubility of the fatty acids containing more than 20 carbon atoms precluded their adequate assessment, but there did not appear to be any effect of behenic acid (C22:0) on osteoclastogenesis at concentrations up to 1 µg/ml.


Figure 1
View larger version (16K):
[in this window]
[in a new window]

  		FIG. 1. Effects of saturated fatty acids on osteoclastogenesis in bone marrow cultures. Each fatty acid has been assessed at the same three concentrations, shown in micrograms per milliliter. The number of carbon atoms in the fatty acid is denoted by the number after the C. Data are mean ± SE. Significant differences from control are shown as follows: *, P < 0.05; **, P < 0.01, by Dunnett’s test; #, P < 0.05, by t test only.

 
The effect of unsaturated fatty acids was then assessed by comparing fatty acids of the same chain length (C18) having different numbers and positions of double bonds. Figure 2Go demonstrates that increasing the number of double bonds in the cis configuration is associated with progressive loss of the antiosteoclastogenic effect, and 10 µg/ml of these polyunsaturated fatty acids actually stimulate osteoclastogenesis. Two trans isomers of C18:1 showed different effects: that with the double bond at C9 showing greater activity than the cis isomer, whereas that with the double bond at C11 was inactive. Conjugated linoleic acid, another trans isomer of C18, was also inactive.


Figure 2
View larger version (22K):
[in this window]
[in a new window]

  		FIG. 2. Effect of 18-carbon fatty acids on osteoclastogenesis in bone marrow cultures. The fatty acids differ in the number of their double bonds (indicated by the number after the colon) and according to whether the double bonds are in a cis (c) or trans (t) configuration. Each fatty acid has been assessed at the same three concentrations. Data are mean ± SE and are ratios of the number of osteoclasts formed in the experimental wells to those in the control wells for that experiment (t/c). Significant differences from control are shown as follows: *, P < 0.05; **, P < 0.01, by Dunnett’s test; #, P < 0.05, by t test only.

 
Polyunsaturated fatty acids have been implicated in a number of biological processes, and some of these activities differ between {omega}3 and {omega}6 fatty acids. We studied a number of fatty acids of similar chain lengths, grouped according to whether they were {omega}3 or {omega}6 fatty acids. These results are shown in Fig. 3Go and show no clear-cut differences between these two classes of fatty acids. None of these fatty acids showed consistent inhibition of osteoclastogenesis, and some even appeared to be stimulatory.


Figure 3
View larger version (37K):
[in this window]
[in a new window]

  		FIG. 3. Effect of {omega}3 and {omega}6 fatty acids on osteoclastogenesis in bone marrow cultures. C18:0 is shown for comparison. Each fatty acid has been assessed at three concentrations as indicated on the figure. The low solubility of C20:4 and C 20:5 reduced the maximum concentrations that could be studied. Data are mean ± SE and are ratios of the number of osteoclasts formed in the experimental wells to those in the control wells for that experiment (t/c). Significant differences from control are shown as follows: *, P < 0.05; **, P < 0.01, by Dunnett’s test; #, P < 0.05, by t test only.

 
RAW264.7 cell cultures
Whereas the bone marrow cultures were carried out in conditions that promote osteoclastogenesis, it is still possible that the effects seen depended on the presence of other cell types. Therefore, we repeated these studies in RAW264.7 cells, a clonal cell line with a preosteoclast phenotype. Inhibition of osteoclast development was again observed with both palmitic and stearic acids, with a comparable efficacy at 10 µg/ml to that seen in the bone marrow cultures (Fig. 4Go). This implies the effects of the fatty acids are directly on the preosteoclast cells and do not depend on the presence of other cell types.


Figure 4
View larger version (7K):
[in this window]
[in a new window]

  		FIG. 4. Effect of palmitic (C16:0) and stearic (C18:0) acids on osteoclast formation in RAW264.7 cell cultures. The x-axis shows fatty acid concentrations in micrograms per milliliter. Significant differences from control are shown as follows: *, P < 0.05; **, P < 0.01, by Dunnett’s test.

 
To determine that the changes we observed in the number of osteoclasts generated in the presence of fatty acids was accompanied by a decrease in bone resorption, we cultured RAW264.7 cells on ivory slices in the presence of RANKL and either vehicle or stearic acid 10 µg/ml. There was a 30% (95% confidence interval 1.2–59, P < 0.05) decrease in resorbed area in the fatty-acid-treated cultures (Fig. 5Go). The number of TRAP-positive cells declined by 19% in the fatty acid-treated cultures.


Figure 5
View larger version (74K):
[in this window]
[in a new window]

  		FIG. 5. Effect of stearic acid on osteoclastic bone resorption in RAW264.7 cells. Cells were cultured on ivory slices in the presence of RANKL 10 ng/ml and either stearic acid 10 µg/ml or vehicle for 6 d. After fixation, TRAP staining, and counting, cells were removed by scrubbing and resorption pits visualized by staining with toluidine blue.

 
Osteoblast cultures
It was noted consistently throughout the osteoclastogenesis experiments that the number of bone marrow cells not staining for TRAP (i.e. the stromal cells) was not diminished by fatty acid treatment, suggesting that the inhibitory effect was specific, rather than a nonspecific toxic effect. This was assessed further in osteoblast cultures, in which thymidine incorporation was measured. Fatty acids with a range of antiosteoclastogenic effects consistently produced moderate stimulation of thymidine incorporation into osteoblasts, at concentrations of 0.1 and 1.0 µg/ml (Fig. 6Go). The highest concentration (10 µg/ml) was less consistent in its effect but did not decrease thymidine incorporation in any experiment.


Figure 6
View larger version (20K):
[in this window]
[in a new window]

  		FIG. 6. Effect of palmitic (16:0), stearic (18:0), oleic (18:1), and linoleic (18:2) acids on thymidine incorporation into primary cultures of fetal rat osteoblasts at 24 h. Significant differences from control are shown as follows: *, P < 0.05; **, P < 0.01, by Dunnett’s test.

 
Bone organ culture
The effects of palmitic (C16:0), stearic (C18:0), oleic (C18:1), and linoleic (C18:2) acids on bone resorption were assessed in cultures of neonatal mouse calvariae. 45Ca release from the calvariae was unaffected by any of these fatty acids when studied at concentrations of 1 and 10 µg/ml (data not shown).

RANKL/osteoprotegerin expression in osteoblastic cells
To determine whether RANKL and/or osteoprotegerin are involved in the effects of fatty acids on osteoclast development, their expression was assessed in the stromal cell line ST2 and primary cultures of rat osteoblast-like cells treated with stearic acid (C18:0). In ST2 cells, there were increases in RANKL at 48 h with higher concentrations of fatty acid, but these were no longer significant at 72 h (Fig. 7Go). There were no consistent changes in osteoprotegerin in these cells. In contrast, there were no changes in RANKL in primary rat osteoblast cultures, whereas osteoprotegerin showed variable responses in terms of both time and dose. In neither cell type was there a sustained change in RANKL to osteoprotegerin ratios that might account for the consistent inhibition of osteoclastogenesis observed.


Figure 7
View larger version (20K):
[in this window]
[in a new window]

  		FIG. 7. Effect of stearic (18:0) acid on relative expression of RANKL and osteoprotegerin (OPG) in the stromal cell line, ST2, and in primary rat osteoblast cultures. *, Significant differences from control, P < 0.05, by Dunnett’s test.

 
Mevalonate pathway enzymes
Because fatty acids have some structural similarity to long chain hydrophobic compounds that are known to inhibit key enzymes in the mevalonate pathway, we assessed the effects of palmitic (16:0), stearic (18:0), oleic (18:1), and linolenic (18:3) acids on the activity of these enzymes. At concentrations of 0.1–10 µg/ml, there was no evidence of any effect on the activity of either enzyme (data not shown).

Fatty acid receptors
The expression of GPR40, GPR41, GPR43, and GPR120 was assessed by real-time PCR in the osteoblastic cell lines ST2 and MC3T3-E1 and in the osteoclast-like cell line RAW264.7. GPR120 was expressed in all cell lines tested, whereas GPR40, GPR41, and GPR43 were present only in RAW264.7 cells. The expression of GPR120 in RAW264.7 cells was greater than 100-fold that of GPR 40 (Fig. 8Go). GW9508, a synthetic GPR40/120 agonist, inhibited osteoclastogenesis in murine bone marrow cultures to a degree comparable with the C16 and C18 fatty acids (Fig. 9Go).


Figure 8
View larger version (7K):
[in this window]
[in a new window]

  		FIG. 8. Relative expression of fatty acid receptors in RAW264.7 cells, assessed using quantitative PCR. Data are mean (SE) gene expression, relative to that of GPR40.

 

Figure 9
View larger version (12K):
[in this window]
[in a new window]

  		FIG. 9. Effect of the synthetic GPR40/120 agonist, GW9508, on osteoclastogenesis in bone marrow cultures. Stearic acid (C18) is shown for comparison. Data are mean ± SE, pooled from two separate experiments and are ratios of the number of osteoclasts formed in the experimental wells to those in the control wells (t/c). Significant differences from control are shown as follows: *, P < 0.05; **, P < 0.01, by Dunnett’s test.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The present data demonstrate inhibition of osteoclastogenesis in bone marrow cultures by saturated fatty acids. This effect shows a clear structure-activity relationship and is influenced by fatty acid chain length, being maximal for palmitic acid (C16). The fatty acids that inhibit osteoclastogenesis tended to stimulate proliferation of osteoblastic cells and were neutral in their effects in calvarial cultures, indicating that they are not inducing a nonspecific toxic effect. The introduction of two or more double bonds abrogates the antiosteoclastogenic effect and may even lead to its reversal. {omega}3 and {omega}6 fatty acids are of comparable, low activity. There is no effect of these fatty acids on bone resorption in neonatal mouse calvariae, suggesting that the effect is on osteoclastogenesis rather than on the activity of mature osteoclasts, which accounts for most of the resorption in the calvarial model. This notion is supported by the parallel decreases we observed in bone resorption and osteoclast number in RAW264.7 cells cultured on bone slices.

The antiosteoclastogenic actions of fatty acids do not appear to be mediated by indirect actions on stromal cells because equivalent effects were observed in RAW264.7 cells to those in murine bone marrow cultures. Furthermore, expression of key stromal cell regulators of osteoclastogenesis, RANKL, and osteoprotegerin was unaffected by stearic acid in two different osteoblastic cell types. A further putative mechanism, blockade of the mevalonate pathway was investigated because this is critically important to osteoclast activity and has been shown to be influenced by alkylamines (17). No effect of fatty acids on this pathway was found, consistent with the observation that fatty acids act predominantly on osteoclast development rather than the activity of mature cells. Our finding that bone cells express receptors known to bind medium and long-chain fatty acids (9, 10, 11) suggests that one of these receptors may mediate the antiosteoclastogenic actions we report. We found that GPR120 is highly expressed in RAW264.7 cells and that a GPR40/120 agonist mimics the antiosteoclastogenic actions of the fatty acids, identifying GPR120 as a likely mediator of the antiosteoclastogenic actions of C16 and C18 fatty acids.

Previous work assessing the effects of lipids on bone has focused on PUFAs. There is some evidence that manipulation of PUFA intakes results in changes in osteoblast activity in vitro and in bone strength in studies in intact animals (6). However, these effects are not predominantly mediated by changes in bone resorption and are most apparent with fatty acids that are inactive in the assays studied in the present report. Therefore, these effects are likely to be quite distinct from the effects of saturated fatty acids on osteoclastogenesis described in the present work.

The present novel findings are potentially of significance in our understanding of human bone physiology. The finding that fatty acids can directly modulate bone cell activity in a way that would be predicted to increase bone mass may in part explain the positive relationship between adipose tissue mass and bone mass. It may also be part of the mechanism for the increased rates of bone resorption observed with reduced energy intakes (20) and for the opposite change after feeding (4, 21, 22). Insulin resistance is associated with increased circulating concentrations of free fatty acids, which could contribute to the higher bone densities reported in this condition (23). Circulating concentrations of free fatty acids are about 0.5 µmol/liter (~0.1 µg/ml), of which palmitic acid is a substantial fraction, so biological fluids contain the relevant fatty acids in concentrations comparable with those studied in the present experiments.

The inhibition of osteoclastogenesis by saturated fatty acids might also be of potential therapeutic importance. Fatty acids are important components of a normal diet, and the manipulation of dietary fatty acid composition may influence bone resorption, bone formation, and bone mass. Such strategies could provide the basis for the development of nutriceuticals that would positively impact on bone health. However, fatty acids are agents of low toxicity that could also be used pharmaceutically. It will be of interest to extend the study of the structure activity relationships beyond naturally occurring fatty acids, to determine whether even more potent compounds can be produced that might be of value in the treatment of osteoporosis.


   Footnotes
 
This work was supported by the Lactopharma Consortium and the Health Research Council of New Zealand.

Disclosure Statement: J.-M.L., M.W., K.E.C., P.C.T., J.E.D., G.A.W., and D.N. have nothing to declare. A.M. and Y.v.d.D. are employed by Fonterra Co-op Group; A.B.G. has consulted for Glaxo Smith Kline; and J.C. and I.R.R. have received grant support from Fonterra and are inventors on a patent application with all rights assigned to Fonterra.

First Published Online July 10, 2008

Abbreviations: FBS, Fetal bovine serum; GPR, G protein-coupled receptor; PUFA, polyunsaturated fatty acid; RANKL, receptor activator of nuclear factor-{kappa}B ligand; TRAP, tartrate-resistant acid phosphatase.

Received January 24, 2008.

Accepted for publication June 26, 2008.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Reid IR 2002 Relationships among body mass, its components, and bone. Bone 31:547–555[CrossRef][Medline]
  2. Reid IR 2008 Relationships between fat and bone. Osteoporosis International 19:595–606[CrossRef][Medline]
  3. Reid IR, Cornish J, Baldock PA 2006 Nutrition-related peptides and bone homeostasis. J Bone Miner Res 21:495–500[CrossRef][Medline]
  4. Henriksen DB, Alexandersen P, Bjarnason NH, Vilsboll T, Hartmann B, Henriksen EEG, Byrjalsen I, Krarup T, Holst JJ, Christiansen C 2003 Role of gastrointestinal hormones in postprandial reduction of bone resorption. J Bone Miner Res 18:2180–2189[CrossRef][Medline]
  5. Priante G, Bordin L, Musacchio E, Clari G, Baggio B 2002 Fatty acids and cytokine mRNA expression in human osteoblastic cells: a specific effect of arachidonic acid. Clin Sci 102:403–409[CrossRef][Medline]
  6. Watkins BA, Li Y, Lippman HE, Feng S 2003 Modulatory effect of omega-3 polyunsaturated fatty acids on osteoblast function and bone metabolism. Prostaglandins Leukot Essent Fatty Acids 68:387–398[CrossRef][Medline]
  7. Cornish J, Callon KE, Naot D, Palmano KP, Banovic T, Bava U, Watson M, Lin JM, Tong PC, Chen Q, Chan VA, Reid HE, Fazzalari N, Baker HM, Baker EN, Haggarty NW, Grey AB, Reid IR 2004 Lactoferrin is a potent regulator of bone cell activity and increases bone formation in vivo. Endocrinology 145:4366–4374[Abstract/Free Full Text]
  8. MacGibbon AKH, Taylor MW 2006 Composition and structure of bovine milk lipids. In: Fox PF, McSweeney PLH, eds. Advanced dairy chemistry. 3rd ed. New York: Springer; 1–43
  9. Milligan G, Stoddart LA, Brown AJ 2006 G protein-coupled receptors for free fatty acids. Cell Signal 18:1360–1365[CrossRef][Medline]
  10. Hirasawa A, Tsumaya K, Awaji T, Katsuma S, Adachi T, Yamada M, Sugimoto Y, Miyazaki S, Tsujimoto G 2005 Free fatty acids regulate gut incretin glucagon-like peptide-1 secretion through GPR120. Nat Med 11:90–94[CrossRef][Medline]
  11. Itoh Y, Kawamata Y, Harada M, Kobayashi M, Fujii R, Fukusumi S, Ogi K, Hosoya M, Tanaka Y, Uejima H, Tanaka H, Maruyama M, Satoh R, Okubo S, Kizawa H, Komatsu H, Matsumura F, Noguchi Y, Shinohara T, Hinuma S, Fujisawa Y, Fujino M 2003 Free fatty acids regulate insulin secretion from pancreatic β cells through GPR40. Nature 422:173–176[CrossRef][Medline]
  12. Brown AJ, Goldsworthy SM, Barnes AA, Eilert MM, Tcheang L, Daniels D, Muir AI, Wigglesworth MJ, Kinghorn I, Fraser NJ, Pike NB, Strum JC, Steplewski KM, Murdock PR, Holder JC, Marshall FH, Szekeres PG, Wilson S, Ignar DM, Foord SM, Wise A, Dowell SJ 2003 The orphan G protein-coupled receptors GPR41 and GPR43 are activated by propionate and other short chain carboxylic acids. J Biol Chem 278:11312–11319[Abstract/Free Full Text]
  13. Cornish J, Callon KE, Lin CQ, Xiao CL, Mulvey TB, Cooper GJ, Reid IR 1999 Trifluoroacetate, a contaminant in purified proteins, inhibits proliferation of osteoblasts and chondrocytes. Am J Physiol 277:E779–E783
  14. Khosla S 2001 Minireview: the OPG/RANKL/RANK system. Endocrinology 142:5050–5055[Abstract/Free Full Text]
  15. Kavanagh KL, Dunford JE, Bunkoczi G, Russell RGG, Oppermann U 2006 The crystal structure of human geranylgeranyl pyrophosphate synthase reveals a novel hexameric arrangement and inhibitory product binding. J Biol Chem 281:22004–22012[Abstract/Free Full Text]
  16. Kavanagh KL, Guo KD, Dunford JE, Wu XQ, Knapp S, Ebetino FH, Rogers MJ, Russell RGG, Oppermann U 2006 The molecular mechanism of nitrogen-containing bisphosphonates as anti osteoporosis drugs. Proc Natl Acad Sci USA 103:7829–7834[Abstract/Free Full Text]
  17. Thompson K, Kikuvi T, Rogers M 2006 Alkylamines prevent osteoclast formation and activity by inhibiting the mevalonate pathway. Bone 38(Suppl):S31
  18. Reed BC, Rilling HC 1976 Substrate binding of avian liver prenyltransferase. Biochemistry 15:3739–3745[CrossRef][Medline]
  19. Briscoe CP, Peat AJ, McKeown SC, Corbett DF, Goetz AS, Littleton TR, McCoy DC, Kenakin TP, Andrews JL, Ammala C, Fornwald JA, Ignar DM, Jenkinson S 2006 Pharmacological regulation of insulin secretion in MIN6 cells through the fatty acid receptor GPR40: identification of agonist and antagonist small molecules. Br J Pharmacol 148:619–628[CrossRef][Medline]
  20. Ihle R, Loucks AB 2004 Dose-response relationships between energy availability and bone turnover in young exercising women. J Bone Miner Res 19:1231–1240[CrossRef][Medline]
  21. Clowes JA, Hannon RA, Yap TS, Hoyle NR, Blumsohn A, Eastell R 2002 Effect of feeding on bone turnover markers and its impact on biological variability of measurements. Bone 30:886–890[CrossRef][Medline]
  22. Bjarnason NH, Henriksen EE, Alexandersen P, Christgau S, Henriksen DB, Christiansen C 2002 Mechanism of circadian variation in bone resorption. Bone 30:307–313[CrossRef][Medline]
  23. Abrahamsen B, Rohold A, Henriksen JE, Beck-Nielsen H 2000 Correlations between insulin sensitivity and bone mineral density in non-diabetic men. Diabet Med 17:124–129[CrossRef][Medline]



This article has been cited by other articles:


Home page
 DiabetesHome page
T. Alquier, M.-L. Peyot, M. G. Latour, M. Kebede, C. M. Sorensen, S. Gesta, C. Ronald Kahn, R. D. Smith, T. L. Jetton, T. O. Metz, et al.
Deletion of GPR40 Impairs Glucose-Induced Insulin Secretion In Vivo in Mice Without Affecting Intracellular Fuel Metabolism in Islets
Diabetes, November 1, 2009; 58(11): 2607 - 2615.
[Abstract] [Full Text] [PDF]

Home page
 IBMS BoneKEyHome page
J. Cornish, J. L. Costa, and D. Naot
The Bone-Fat Mass Relationship: Laboratory Studies
IBMS BoneKEy, September 1, 2009; 6(9): 311 - 322.
[Abstract] [Full Text] [PDF]

Home page
 DiabetesHome page
T. Alquier and V. Poitout
GPR40: Good Cop, Bad Cop?
Diabetes, May 1, 2009; 58(5): 1035 - 1036.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Cornish, J.
Right arrow Articles by Reid, I. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cornish, J.
Right arrow Articles by Reid, I. R.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals