Dietary emu oil supplementation suppresses 5-fluorouracil chemotherapy-induced inflammation, osteoclast formation, and bone loss

Rethi Raghu Nadhanan, Suzanne M. Abimosleh, Yu-Wen Su, Michaela A. Scherer, Gordon S Howarth, Cory J Xiang

01 JUN 2012 

Abstract

Cancer chemotherapy can cause osteopenia or osteoporosis, and yet the underlying mechanisms remain unclear, and currently, no preventative treatments are available. This study investigated damaging effects of 5-fluorouracil (5-FU) on histological, cellular, and molecular changes in the tibial metaphysis and potential protective benefits of emu oil (EO), which is known to possess a potent anti-inflammatory property. Female dark agouti rats were gavaged orally with EO or water (1 ml·day−1·rat−1) for 1 wk before a single ip injection of 5-FU (150 mg/kg) or saline (Sal) was given. The treatment groups were H2O + Sal, H2O + 5-FU, EO + 5-FU, and EO + Sal. Oral gavage was given throughout the whole period up to 1 day before euthanasia (days 3, 4, and 5 post-5-FU). Histological analysis showed that H2O + 5-FU significantly reduced heights of primary spongiosa on days 3 and 5 and trabecular bone volume of secondary spongiosa on days 3 and 4. It reduced density of osteoblasts slightly and caused an increase in the density of osteoclasts on trabecular bone surface on day 4. EO supplementation prevented reduction of osteoblasts and induction of osteoclasts and bone loss caused by 5-FU. Gene expression studies confirmed an inhibitory effect of EO on osteoclasts since it suppressed 5-FU-induced expression of proinflammatory and osteoclastogenic cytokine TNFα, osteoclast marker receptor activator of nuclear factor-κB, and osteoclast-associated receptor. Therefore, this study demonstrated that EO can counter 5-FU chemotherapy-induced inflammation in bone, preserve osteoblasts, suppress osteoclast formation, and potentially be useful in preventing 5-FU chemotherapy-induced bone loss.

Anti-cancer chemotherapy can cause significant adverse effects on tissues, including the bone, in both pediatric and adult cancer patients (25). Short stature, low bone mass or osteoporosis, and/or fractures are some skeletal side effects that are due to chemotherapy among pediatric patients and adult survivors (32, 38, 46). Experimental studies in rats have also shown that drugs such as methotrexate (MTX), cisplatin, doxorubicin, etoposide, cyclophosphamide, and 5-fluorouracil (5-FU) can cause a reduction in bone growth and bone mass (50–52). 5-FU is an antimetabolite drug commonly used to treat adult patients suffering from colorectal and breast cancer, whereas in children 5-FU is used to treat childhood solid tumours (27, 36). 5-FU inhibits thymidylate synthase, an enzyme required to synthesize thymine nucleotide, which is important for synthesis of DNA and RNA (27). In an acute 5-FU chemotherapy model in rats, reduced primary spongiosa height and decreased bone volume in the secondary spongiosa of the metaphysis were observed, suggesting that 5-FU can significantly affect bone growth (48).

Chemotherapeutic agents such as 5-FU, MTX, cyclophosphamide, and etoposide have been shown to cause severe osteopenia by suppressing bone-forming cells (osteoblasts) and promoting recruitment of bone-degrading cells (osteoclasts) (49–51). However, the underlying molecular mechanisms of chemotherapy-induced bone defects still remain largely unknown. Osteoclasts are differentiated from monocyte/macrophage precursor cells under the influence of TNFα-related cytokine, namely receptor activator of nuclear factor-κB ligand (RANKL) and macrophage colony-stimulating factor (M-CSF) (6). Osteoprotegerin (OPG), being the decoy receptor produced by osteoblasts, competitively inhibits the binding of RANKL to RANK and thus suppresses the production of osteoclasts (30). However, a noncanonical pathway of osteoclastogenesis also exists, in which proinflammatory cytokines such as TNFα and interleukin-6 (IL-6) play a role in osteoclast formation and bone loss (15, 21). These proinflammatory cytokines have been shown to be upregulated and responsible for causing tissue inflammation and to stimulate formation and activity of osteoclasts through the RANKL/RANK-dependent or -independent pathway (20, 37, 40).

Currently, there are no safe and cost-effective treatments against chemotherapy-induced bone loss. The available antiresorptive therapies using bisphosphonates are known to reduce resorption and increase bone mass and thus have some efficacy in preventing/reducing osteoporosis (17). A review (34) has indicated that there have been attempts at using bisphosphonates to ameliorate the possible side effects of cancer-induced or treatment-induced bone loss. However, one study has shown that there were no significant changes in terms of bone loss between the treatment groups receiving oral risedronate or placebo, which has called into question the correct dosing of bisphosphonates (16). The review (34) has also mentioned that bisphosphonates inhibit the formation of new bone as well as bone resorption and that the increase in bone density seen in some studies involving the use of bisphosphonates might be due to the increased mineralization of the old bone and inhibition of the new soft bone. However, high costs involved in their administration and occasional toxicities and side effects, including brittle bones due to suppressed bone turnover, have limited their widespread usage, and the cost-effectiveness of their usage or long-term use has been questioned (9, 33). Considering cost-effectiveness, safety, and ease of administration, one attractive approach would be to use a nutraceutical as a dietary supplement to aid skeletal health. Emu (Dromaius novaehollandiae), traditionally native to Australia, are now farmed around the world for their meat, leather, and, most recently, oil (47). Emu oil (EO) is extracted from both the subcutaneous and retroperitoneal fat of the bird by first rendering the macerated tissue and then passing the liquefied fat through a series of filters to obtain purified oil (4). EO is composed predominately of fatty acids, with a lipid content of 98.8% for the subcutaneous adipose tissue and 98% for the retroperitoneal adipose tissue. EO is a natural source of 18:1 (n-9) fatty acid (oleic acid), which constitutes ∼43–46% of the fatty acid composition, with 18:3 (n-3) linolenic acid (9.6%), palmitic acid (23.5%), and 18:2 (n-6) linoleic acid (0.6%) (53). The remaining 1–2% is yet to be identified; however, natural antioxidants such as carotenoids and flavones are present in the EO (24). EO has been used as an anti-inflammatory agent by Australian aboriginals to treat arthritis and joint pains for hundreds of years (43). Recent studies have focused on the effects of EO on arthritis and dermal inflammation in animal models (47, 53). Topical application of EO to animals has been shown to reduce the levels of TNFα and other proinflammatory cytokines in a model of adjuvant-induced inflammation (53). Previous studies comparing different oils showed that EO was most potent in reducing adjuvant-induced dermal inflammation (53). More recently, dietary EO supplementation has been shown to be efficacious at reducing intestinal tissue damage and inflammation caused by 5-FU chemotherapy (1, 24). However, whether EO can protect bone from inflammatory bone loss during chemotherapy remains unknown. In the current study, an acute 5-FU chemotherapy model was used to examine whether a local inflammatory condition in bone is associated with bone loss and to investigate whether the potent anti-inflammatory nutraceutical EO can potentially prevent bone loss caused by chemotherapy.

DISCUSSION

Chemotherapy with 5-FU, a commonly used antimetabolite drug for breast and colon cancers, has been shown to cause bone pain, osteoporosis, and fracture (7, 28). Currently, mechanisms for 5-FU-induced bone defects are unclear, and there has been a lack of effective supplementary treatments to protect bone during cancer chemotherapy. The purpose of our current investigation was to elucidate whether supplementation with EO can prevent or reduce the damage caused by 5-FU chemotherapy and/or improve or hasten the recovery in the bone. Using a rat acute 5-FU chemotherapy model, this current study examined effects of 5-FU chemotherapy on primary spongiosa height, bone volume, bone cells, and the local bone inflammatory condition and addressed the potential protective effects of supplementary treatment with emu oil, which is known to possess an anti-inflammatory property.

Growth plate and the primary spongiosa of the metaphysis are part of the “active growth unit.” In this study, consistent with a previous finding (48), height of primary spongiosa was affected only on days 3 and 5 and not on day 4. However, the obvious reduction in height was also noted between days 5 and 10 in that previous study. Height retardation as a consequence of 5-FU has been reflected on the primary spongiosa since this is where the resorption of growth plate cartilage template and formation of trabecular bone occur during endochondral ossification (52). The present study also observed a decreased bone volume at the secondary spongiosa and a slight suppression of osteoblasts but increased osteoclast formation and density, especially on day 4 post-5-FU. In addition, the present study showed that 5-FU chemotherapy induced an inflammatory condition in bone by upregulating proinflammatory cytokines and osteoclast regulatory factors. These suggest that 5-FU treatment increases osteoclast presence. Therefore, the higher osteoclast density and bone resorption (but no significant difference on the osteoblast density) could have contributed to the bigger bone loss seen in the secondary spongiosa on day 4, a region where bone remodeling is actively occurring.

The underlying mechanism for 5-FU-induced increased osteoclast formation and density remains to be investigated. Since formation, number, and activity of osteoclasts are regulated by a number of inflammatory molecules or modulators (21), expression of the key osteoclastogenic molecules or modulators at the metaphyseal bone from the treated rats was analyzed in the current study. Our data showed that, consistent with the increased osteoclast density on day 4, there were a significant increase in TNFα, RANK, and OSCAR mRNA expressions in bone. Previously, TNFα has been shown to be a potent stimulator of bone resorption and can promote osteoclastogenesis directly or via the RANKL induction pathway (22, 35). Consistent with the increased osteoclastogenesis and higher osteoclast density in H2O + 5-FU-treated rats observed on day 4, RANK and OSCAR, which are expressed specifically on osteoclast lineage cells and responsible for osteoclast differentiation and activity (30), were upregulated significantly in bone. These findings suggest that 5-FU chemotherapy might be able to create an inflammatory microenvironment in the bone, causing increased levels of proinflammatory cytokines that may be responsible for stimulating osteoclastic bone resorption. Previous studies using rat models have demonstrated that MTX, another chemotherapeutic agent, decreases trabecular bone volume and causes bone loss that is associated with a low osteoblast number, higher osteoclast density on the bone surface, and increased osteoclast progenitor cell number in the bone marrow (12, 50). Similarly, the increase in osteoclast numbers and activity has also been shown to be associated with an increase in proinflammatory cytokines in circulation in patients undergoing chemotherapy (11, 37).

Many chemotherapeutic agents also induce the production of free radicals, which may be another mechanism for chemotherapy drugs to exert toxic effects on normal cells (10, 41). Some clinical studies have documented that oxidative stress might be playing a role in the pathogenesis of bone loss, and the reactive oxygen species formed are potent inducers of proinflammatory cytokines (3, 29). Osteoclasts have been shown to be activated by reactive oxygen species to enhance bone resorption (13, 18). A study has shown that 5-FU induces oxidative stress in bone marrow (31). However, the precise role of oxidative stress caused by 5-FU still remains to be studied, and further studies are required to investigate whether the oxidative stress enhances osteoclast formation via inducing inflammatory cytokines in the 5-FU chemotherapy setting.

In the current study, an inflammatory condition was identified in bone after 5-FU chemotherapy; thus this study also sought to examine whether the nutraceutical EO (which is known to possess a potential anti-inflammatory property) would have some protective effects on bone during 5-FU chemotherapy. Here, EO supplementation (EO + 5-FU) was shown to significantly preserve the height of the primary spongiosa (days 3 and 5) and bone volume at the secondary spongiosa, which were reduced by H2O + 5-FU (days 3 and 4). EO supplementation significantly preserved the osteoblast density especially on day 3 after H2O + 5-FU administration and prevented 5-FU-induced higher osteoclast density in the metaphysis. Although the mechanisms for EO-induced preservation of osteoblast density remain to be studied, it is known that a constant supply of osteoblasts is required for normal bone formation and homeostasis, and any significant reduction in osteoblast numbers and activity can result in reduced bone formation and lead to osteoporosis (8, 49, 51). Although the underlying mechanisms for EO's antiosteoclastic effects remain to be defined, consistent with its effects in suppressing 5-FU-induced osteoclastogenesis, our gene expression studies suggest that EO supplementation can suppress the induction of proinflammatory cytokines, particularly TNFα and osteoclast receptors RANK and OSCAR, which were upregulated by H2O + 5-FU, suggesting its effects in suppressing the local osteoclastogenic signals induced locally in bone by 5-FU chemotherapy. This was consistent with another study whereby topical application of EO to animals has been shown to reduce the levels of TNFα in a model of adjuvant-induced inflammation (53). A study conducted in CD-1 mice has shown that EO's anti-inflammatory property was associated with suppression of the induced expression of proinflammatory cytokines IL-1 and TNFα (53).

Our current study suggests a potential role for EO in the treatment of chemotherapy-induced inflammatory bone loss. Consistently, a previous study has shown that EO supplementation was able to decrease 5-FU chemotherapy-associated inflammation in an intestinal mucosal damage (mucositis) model (24). A number of hypotheses relate to the potential mechanism by which EO exerts its anti-inflammatory effects; 18:3 (n-3) polyunsaturated fatty acids have been shown to suppress TNFα expression levels in humans (23) and to inhibit RANK expression and osteoclast formation (39, 44). It has been suggested that the linolenic acid and oleic acid present in EO may be involved in its anti-inflammatory action (53). However, it has been proposed that the anti-inflammatory properties of EO may not be fully explained by the fatty acid profile and that its other components, such as tocopherols, carotenoids, and flavones, may exert additional anti-inflammatory (42) and antioxidant (45) effects. A recent study has suggested that the radical scavenging properties of EO were most likely due to the minor constituents in the nontriglyceride fraction of the oil (2), which, in combination with the fatty acid components, offered greater protection against oxidation. These antioxidants may have impacted on levels of damaging reactive oxygen species that are generated during chemotherapy. However, since the content of 18:3 (n-3) fatty acids is not particularly high in EO (4), it is speculated that probably the nontriglyceride components of EO might be contributing to its antiosteoclastic effect after 5-FU chemotherapy observed in the current study. The promising anti-inflammatory effects displayed by EO in previous studies (24, 53) and the positive effects of EO in preventing osteoclast formation and bone loss after 5-FU chemotherapy suggest its therapeutic potential in preventing chemotherapy-induced inflammatory condition and bone loss.

In the current study, a set amount of EO (unadjusted by body weight) had been gavaged to the rats (1.0 ml/day). However, the differences in daily body weight change between H2O + Sal and H2O + 5-FU-treated rats were found to be relatively minor, with the maximal difference being only 13 g in average body weight between controls (111 g) and 5-FU rats (98 g) during the trial (which was evident on day 4 after 5-FU); this small body weight difference would have equated to only a 0.12-ml difference between minimum and maximum gavage volumes. Considering that the gavaged dose (1.0 ml/day) had been a relatively large dose for rats, it is believed that such a minor difference in body weight would not have impacted significantly on the outcomes of any of the end point analyses. However, further studies are required to determine the minimal effective dose of EO in preventing 5-FU chemotherapy-induced bone loss. More studies are required to also investigate the active components of EO responsible and the action mechanisms for the anti-inflammatory and antiosteoclastic properties of EO.

In summary, the current study demonstrated that acute 5-FU chemotherapy induced an inflammatory condition in bone, increased osteoclast formation, and caused bone loss as shown by reduced primary spongiosa height and decreased secondary spongiosa bone volume. The present study has also provided some evidence that EO has an antiosteoclastic and osteotrophic property and has shown encouraging data suggesting that EO might be useful in suppressing inflammatory condition and osteoclast formation and preventing bone loss induced during 5-FU chemotherapy. Further studies are required with models for a longer course of chemotherapy, with more time points and an earlier and extended observation period, and possibly in a tumour-bearing model to obtain a more comprehensive profile for both cell types (osteoblasts and osteoclasts) and a better understanding of the mechanisms leading to bone loss and the effects of EO in protecting bone during chemotherapy. In our previous study (48), we observed that, although there were no obvious structural changes of metaphyseal trabecular bone on days 1 and 2 post-5-FU treatment, there were a significant induction of osteoblast apoptosis and a suppression of osteoblast proliferation on these earlier time points. Since the osteoblast number and activity affect osteoclast formation (via production of osteoclastogenesis regulatory cytokines), it is possible that changes in osteoblasts in the earlier time points (48) might have played a role in causing the osteoclast formation and greater bone loss at a later time (day 4; observed in current study). Further studies (e.g., with more time points and extended treatment periods) are required to fully understand the time course of cellular and histological responses, cellular cross-talks, and molecular mechanisms underlying 5-FU-induced bone loss and EO-protective effects.

GRANTS

This project was funded in part by the National Health and Medical Research Council (NHMRC) Australia, the Australian Research Council Linkage Project, the University of Adelaide, and the University of South Australia. R. Raghu Nadhanan is a recipient of Ph.D scholarship from the University of South Australia. S. M. Abimosleh is a recipient of Ph.D scholarship from the University of Adelaide. G. S. Howarth is supported by a Sally Birch Cancer Council Australia Research Fellowship. C. J. Xian is a senior research fellow of NHMRC Australia.

DISCLOSURES

All authors have no conflicts of interest, financial or otherwise.

AUTHOR CONTRIBUTIONS

R.R.N., S.M.A., and Y.-W.S. performed the experiments; R.R.N. and C.J.X. analyzed the data; R.R.N., M.A.S., and C.J.X. interpreted the results of the experiments; R.R.N. and C.J.X. prepared the figures; R.R.N. and C.J.X. drafted the manuscript; R.R.N. and C.J.X. edited and revised the manuscript; S.M.A., G.S.H., and C.J.X. did the conception and design of the research; C.J.X. approved the final version of the manuscript.

https://www.physiology.org/doi/full/10.1152/ajpendo.00587.2011 

ACKNOWLEDGMENTS

This project was supported by Emu Tracks Pty Ltd (Marleston, South Australia, Australia).