点击显示 收起
1 From the Department of Food Science, Lipid Chemistry and Molecular Biology Laboratory, Purdue University, West Lafayette, IN (BAW, YL, HEL, and SR), and the Department of Anatomy and Cell Biology, Indiana University School of Medicine, Indianapolis (BAW and MFS)
2 Presented at the workshop "The Role of Conjugated Linoleic Acid in Human Health," held in Winnipeg, Canada, March 13–15, 2003. 3 Supported by a grant from the 21st Century Research and Technology Fund, creating the Center for Enhancing Foods to Protect Health. 4 Address reprint requests to BA Watkins, Department of Food Science, Lipid Chemistry and Molecular Biology Laboratory, 745 Agriculture Mall Drive, Purdue University, West Lafayette, IN 47907-2009. E-mail: baw{at}purdue.edu.
ABSTRACT
The philosopher William of Ockham is recognized for the maxim that an assumption introduced to explain a phenomenon must not be multiplied beyond necessity, or that the simplest explanation is probably the correct explanation. The general truth is that conjugated linoleic acids (CLAs) are nutrients. However, the demonstration that these isomers of octadecadienoic acid protect against cancers in rodents stimulated curiosity that directed significant resources to characterize the biological functions of these fatty acids in cell and animal models. The benefits to human subjects given supplements of CLA were at best modest. The disappointing results in humans should be taken as an opportunity to critically evaluate all findings of CLA use and to consolidate the common actions of this nutrient so that future investigations focus on specific isomers and the most reasonable mechanisms. As such, the principal and consistently reported benefits of CLA have been in improving cancer outcomes, reducing body fat in growing animals, and modulating cell functions. Recognizing where related actions of CLA converge in specific disease conditions and physiologic states is how research efforts should be directed to minimize the pursuit of superfluous theories. Here, we briefly review the current biological effects of CLA and attempt to integrate their potential effect on the physiology and health of the skeletal system. Thus, the purpose of this review is to advance the science of CLA and to identify areas of research in which these nutrients affect bone metabolism and skeletal health.
Key Words: Conjugated linoleic acid • n–3 polyunsaturated fatty acids • bone • osteoblast • estrogen deficiency • rats
INTRODUCTION
Biochemical and molecular aspects
Conjugated linoleic acid (CLA) is a group of positional and geometric fatty acid isomers of octadecadienoic acid. The CLA isomers are purported to possess antioxidant properties, inhibit carcinogen-DNA adduct formation, induce apoptosis and cytotoxic activity, modulate tissue fatty acid composition and prostanoid formation, and affect the expression and action of cytokines and growth factors (1). Although numerous biological actions of CLA have been reported, the most consistent findings include anticancer effects in rodents and cancer cells and reduction of body fat in growing animals. In some cases the biological responses observed from CLA isomers were influenced by the amounts of dietary n–6 and n–3 polyunsaturated fatty acids (PUFAs) present (2-4).
The cytotoxic effects of CLA isomers on growth of various human and animal-derived cancer cells seem to be mediated by decreasing the expression of the gene transcription factor Bcl-2 family whose members inhibit apoptotic cell death and/or induce caspase-dependent apoptosis (5-9). CLA also prevented basic fibroblast growth factor–induced angiogenesis (10), a critical process for growth and metastasis of cancers. Evidence for CLA’s anticarcinogenic effects indicates a modifying role in peroxisome proliferator–activated receptor (PPAR) action (11). Moreover, CLA actions on fat and energy metabolism could, in part, be directed through changes in both PPAR and PPAR (12, 13).
Although research demonstrated that CLA reduced body fat accumulation in growing animals (14-16), it seems that not all major CLA isomers contributed to this effect equally as was reported recently in mice (17, 18). In one experiment the trans-10,cis-12 (t10,c12) isomer, but not the cis-9,trans-11 (c9,t11) isomer, induced body fat loss and adipocyte apoptosis in mice (14). Other investigations indicated a fat loss or redistribution when CLA isomers were given to chickens (19), rats (20), and pigs (21), but when tested in human subjects the results were disappointing (22-24). Hence, careful examination of the physiologic states and endocrine effects on energy and lipid metabolism is essential for proceeding with weight loss research with use of CLA.
CLA was reported to alter leptin concentration or expression in animal and cell culture studies (25-29). As a regulator of appetite and lipid metabolism and a proposed neuroendocrine regulator of bone mass, leptin is a likely target of CLA action in mediating its effects on body fat as well as bone growth and bone mass (30, 31).
Specific effects of CLA isomers on activity and expression of enzymes associated with anabolic pathways of lipid metabolism were reported (32). For example, CLA was observed to decrease the mRNA amount of the 9-desaturase enzyme in both liver tissue and hepatocyte cultures (33).
In immune function, CLA diminished the production of an array of proinflammatory products in macrophages through activation of PPAR (12), and it lowered basal and lipopolysaccharide-stimulated interleukin 6 (IL-6) and basal tumor necrosis factor (TNF) production by rat resident peritoneal macrophages (2). Through activation of PPAR, CLA decreased interferon-–induced mRNA expression of cyclooxygenase 2 (COX-2), inducible nitrous oxide synthase, TNF-, and proinflammatory cytokines (IL-1ß and IL-6) in RAW macrophage cell cultures (12).
Certain PUFAs [linoleic acid (LA; 18:2n–6), arachidonic acid (AA; 20:4n–6), eicosapentaenoic acid (EPA; 20:5n–3), and docosahexaenoic acid (DHA; 22:6n–3)] are known agonists or antagonists of COX-2 expression through the activation of PPARs (34), and we reported that dietary CLA isomers reduced ex vivo prostaglandin E2 (PGE2) production in rat bone organ cultures (3). Similar effects of CLA on PGE2 production in various biological systems were demonstrated (35-37). On the basis of these aforementioned actions of CLA isomers, the likely biochemical and molecular targets that integrate their potential biological effect include PPARs, COX enzymes, and other transcription factors. Therefore, consistent with Ockham’s Razor, CLA isomers have potential actions, similar to other PUFA nutrients, to alter metabolism and physiologic functions.
Conjugated linoleic acid effects on bone and joint
Investigations on CLA and bone biology are limited; however, the principal studies focused on bone modeling in rats. In this review, new findings in osteoblasts and estrogen-deficient ovariectomized (OVX) rats are reported. Experiments in growing rats indicate that dietary CLA isomers could decrease bone formation rates and the concentrations of serum biomarkers of bone formation. Our laboratory investigated the effects of 1% dietary CLA in rats. At this concentration, CLA decreased ex vivo PGE2 production in bone organ culture, reduced serum insulinlike growth factor I, and markedly lowered mineral apposition and bone formation rates in tibia compared with the control group (3, 4). In an ensuing study with growing rats, 0.5% dietary CLA was examined on bone formation (29). Serum osteocalcin concentrations and bone-specific alkaline phosphatase (ALP) activity were lowered in rats fed CLA; however, the addition of CLA to moderate dietary n–6 PUFAs resulted in a higher bone formation rate in tibia. These investigations suggest that the effects of dietary CLA could depend on the isomers, the type of fat, the balance of PUFAs in the diet, and the stage of development (29). In agreement with our observations, Kelly et al (38) reported that feeding CLA to growing rats lowered PGE2 production in bone organ culture but did not influence bone formation (osteocalcin) and bone resorption (urinary pyridinium crosslinks) markers or bone mineral mass. In cell culture studies, CLA appeared to enhance calcium absorption (39), and this effect seemed to be improved with n–3 PUFAs (38).
CLA actions appear to be related to or influenced by dietary n–3 PUFAs and modulate key enzymes such as COX that regulates cell function in the osteoblast. In bone, CLA could change COX-2 action/expression to alter prostaglandin-dependent bone resorption, PGE receptor–mediated cell function, and cytokine-induced extracellular release of PGE2 (40, 41).
CLA can also exert effects on bone metabolism through leptin. We reported a lower serum leptin concentration in rats given CLA, which was associated with a higher bone formation rate after 42 d of treatment (29). Yamasaki et al (42) reported that feeding CLA to rats lowered serum leptin concentration compared with the control group supplemented with 8% safflower oil. Leptin can affect bone mass through the central nervous system (43) or directly exert its osteogenic effect on marrow stromal cells (preosteoblastic cells) (44). Emerging evidence suggests that leptin is a systemic hormone that governs bone remodeling (43).
CLA can inhibit inflammatory processes in bone and joint diseases by down-regulating prostanoid production (3, 37, 45). In rats given CLA, bone metabolism can be altered by modulating the production of bone resorptive cytokines and leukotriene B4 (46). IL-1 and IL-6 have long been implicated in the pathophysiology of bone diseases such as rheumatoid arthritis (47) and postmenopausal osteoporosis (48, 49). Dietary CLA was shown to lower basal and lipopolysaccharide-stimulated IL-6 production and basal TNF production by resident peritoneal macrophages in rats (2). Assuming CLA has similar effects on these cytokines in bone cells, together with the fact that CLA reduced the production of PGE2 in bone tissue, at an appropriate dietary amount, CLA can exhibit beneficial effects in the management of inflammatory bone disease.
EXPERIMENTS ON OSTEOBLASTS
Watkins et al (50) reported that fetal rat calvarial cells treated with CLA isomers exhibited an inhibitory effect compared with LA on nodule formation, which is associated with the expression of ALP. The effect of CLA isomers, which generally lowered nodule formation, is consistent with reduced serum concentrations of bone-specific ALP activity in CLA-treated animals (51). Fatty acids can also be either stimulatory or inhibitory on the expression of core binding factor -1 in these cells (50). MC3T3-E1 osteoblastic cells enriched with EPA demonstrated higher ALP activity compared with cells treated with AA (50). Forskolin (FSK), IL-1, TNF-, and PGE2 were demonstrated to induce COX-2 mRNA and protein expression in this cell line (52, 53). Further, IL-1, TNF-, and PGE2 were implicated as possible agents that mediate bone resorption (54). In light of these findings, the following experiments were done to determine whether CLA isomers alter osteoblast function.
Materials and reagents
Chemicals and supplies used in experiments with MC3T3-E1 osteoblast-like cells included fetal bovine serum (Atlanta Biologicals, Atlanta); antibiotics and antimycotics (100 U penicillin, 100 µg streptomycin, and 250 ng amphotericin B/mL), trypsin/EDTA solutions (Invitrogen, Carlsbad, CA); tissue culture plates (Fisher Scientific, Hanover Park, IL); CLA isomers [a 92% CLA mixture (catalog no. 1247) for cell enrichment experiment, 98% pure c9,t11-isomer (catalog no. 1245), and t10,c12-isomer (catalog no. 1249) for PGE2 and COX-2 analyses; Matreya Inc, Pleasant Gap, PA]; fatty acids (>99% purity; Nu-Chek-Prep, Elysian, MN); mammalian protein extraction reagent (Pierce Chemical Company, Rockford, IL); protein assay kit, 10% Tris-HCl sodium dodecyl sulfate–polyacrylamide gel electrophoresis gels, and polyvinylidene difluoride membranes (Bio-Rad, Hercules, CA); PGE2, PGE2 assay kit, and COX-2 polyclonal antibody (Cayman Chemical Company, Ann Arbor, MI); secondary antibodies conjugated to horseradish peroxidase and the enhanced chemiluminescence kit (Amersham Pharmacia Biotech, Piscataway, NJ); other materials (Sigma, St Louis).
Cell culture and fatty acid treatment
MC3T3-E1 cells were cultured in growth medium (GM) consisting of Dulbecco’s modification of Eagle’s medium supplemented with 10% fetal bovine serum, 25 µg/mL ascorbic acid, 1.1 µg/mL ß-glycerolphosphate, and 1% antibiotics and antimycotics at 37 °C in 5% CO2. MC3T3-E1 cultures were maintained in 100-mm tissue culture plates at an initial density of 60 000 cells per plate. Cells were subcultured with use of a trypsin/EDTA solution (0.5 g trypsin, 0.2 g EDTA · 4Na/mL) in calcium/magnesium-free phosphate-buffered saline (PBS). All fatty acids were dissolved in 100-proof ethanol at a final concentration of 100 mg/mL, flushed with nitrogen, and stored at –20 °C until needed. Fatty acid–supplemented medium was prepared by adding an aliquot of fatty acid stock solution (1.0 mmol/L) to either serum-free GM or GM containing bovine serum albumin to obtain a final concentration of 200 µmol/L.
Preparation and analysis of fatty acid methyl esters
MC3T3-E1 cells were subcultured into 150-cm2 flasks and cultured for 24 h to allow attachment. After 24 h, the cells were washed twice with serum-free medium, and CLA-supplemented GM (200 µmol/L) was added (74% c9,t11, 17% cis,cis, and 1% trans,trans isomers; catalog no. 1247; Matreya). At the end of PUFA exposure (24 h), medium was aspirated and the cell layer was washed in PBS. Cells were then scraped into PBS, placed in a glass centrifuge tube, pelleted, resuspended in 3 mL methanol, and frozen at –20 °C until analyzed. Cell suspensions were sonicated (Sonic Bath; Branson, Danbury, CT) for 10 min, lipids were extracted with chloroform:methanol (2:1, by vol), and fatty acid methyl esters were prepared with use of sodium methoxide and analyzed by gas chromatography (4).
Prostaglandin E2 analysis
MC3T3-E1 cells were seeded at 3000 cells/well in 24-well plates and cultured for 3 d in GM until confluent. At confluency, medium was removed, and one of the following culturing/treatment conditions was commenced. Cells were enriched with 200 µmol/L of the indicated fatty acid in GM for 24 h. After 24 h of fatty acid exposure, cells were washed twice with serum-free GM, fresh GM containing FSK (50 µmol/L) was added, and after 24 h medium was collected for analysis (Figure 1A). Cells were enriched with 200 µmol/L of the indicated fatty acid in GM for 72 h. After the 72-h fatty acid enrichment, FSK (50 µmol/L) was added directly to the wells containing fatty acid–supplemented medium for 4 h after which medium was collected for analysis (Figure 1B). Amounts of PGE2 in the medium were analyzed with use of a commercial kit (43% cross-reactivity with PGE3).
FIGURE 1.. Prostaglandin E2 (PGE2) production in MC3T3-E1 osteoblast-like cells enriched with different fatty acids (200 µmol/L) for 24 or 72 h. A: Forskolin (FSK) induction (50 µmol/L, 24 h) was performed after the removal of fatty acid–enriched media. B: Cells were enriched with the indicated fatty acid (200 µmol/L) in growth medium for 72 h. After the 72-h fatty acid enrichment, FSK (50 µmol/L) was added directly to the wells containing fatty acid–supplemented medium for 4 h, after which medium was collected for analysis. As shown in panel A, PGE2 production was lower in cells treated with conjugated linoleic acid isomers and long-chain n–3 polyunsaturated fatty acids [eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA)] than in cells treated with arachidonic acid (AA). As shown in panel B, the PGE2 amount was the highest in cells treated with AA, and the other fatty acids did not significantly affect the PGE2 amount compared with the vehicle (V). The bars represent the mean ± SD for n = 3 per treatment. Bars within a panel with different letters are significantly different at the P value shown (Tukey’s mean separation test). OA, oleic acid; LA, linoleic acid; LnA, linolenic acid.
Cyclooxygenase 2 protein induction
MC3T3-E1 cells seeded at 3000 cells/well in 24-well plates were cultured in GM. Once confluent (3 d), the medium was removed, and one of the following culturing/treatment conditions was commenced. Cells were enriched with 200 µmol/L of the indicated fatty acid in GM for 72 h. After the 72-h fatty acid enrichment, IL-1 (10 ng/mL) was added directly to the wells containing the fatty acid–supplemented medium, and cells were collected and prepared for analysis after 2 and 4 h of IL-1 exposure (Figure 2A). Cells were enriched with 200 µmol/L of the indicated fatty acid in GM for 24 h. After 24 h of fatty acid exposure, cells were washed twice with serum-free GM and fresh GM containing either 50 µmol/L FSK or 10 ng/mL IL-1 was added for 4 h, after which cells were collected and prepared for analysis (Figure 2B). After exposure to inducer, medium was removed, cell layers were washed once with cold PBS, culture plates were placed on ice, and cells were lysed with use of mammalian protein extraction reagent. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis and Western blotting were performed to determine the amounts of COX-2 protein level in the cell lysates (50).
FIGURE 2.. Cyclooxygenase 2 (COX-2) protein level (normalized by ß-actin) in MC3T3-E1 osteoblast-like cells enriched with different fatty acids (200 µmol/L). A: Cells were enriched with the indicated fatty acid (200 µmol/L) in growth medium (GM) for 72 h. After the 72-h fatty acid enrichment, interleukin 1 (IL-1; 10 ng/mL) was added directly to the wells containing the fatty acid–supplemented medium, and the cells were collected and prepared for analysis after 2 and 4 h of IL-1 exposure. B: Cells were enriched with the indicated fatty acid (200 µmol/L) in GM for 24 h. After 24 h of fatty acid exposure, the cells were washed twice with serum-free GM, and fresh GM containing either forskolin (FSK; 50 µmol/L) or IL-1 (10 ng/mL) was added for 4 h, after which the cells were collected and prepared for analysis. As shown in panel A, cells treated with docosahexaenoic acid (DHA) had lower COX-2 protein at 2 h; however, those enriched with conjugated linoleic acid (CLA) isomers did not differ significantly from cells enriched with the vehicle (V) at either time. As shown in panel B, inducing cells with FSK resulted in higher COX-2 protein amounts than did induction with IL-1. Arachidonic acid (AA) enrichment resulted in the lowest COX-2 protein amount compared with other fatty acid treatments in the FSK-induced group. CLA (t10,c12)-enriched cells produced the highest COX-2 protein among the treatments induced by IL-1. The bars represent the mean ± SD for n = 3 per treatment. LA, linoleic acid; LnA, linolenic acid; EPA, eicosapentaenoic acid. Bars with different letters within induction time (panel A) or inducer treatment (panel B) are significantly different at the P value shown (Tukey’s mean separation test). The statistical analyses on the 2 sides of panel A and panel B were performed separately.
Statistical analyses
Statistical analysis for experiments involved either t test or one-way analysis of variance, when applicable, with use of SAS statistical software version 6.12 (SAS Institute Inc, Cary, NC). When differences were detected at an = 0.05 level by the one-way analysis of variance, means were analyzed with use of Tukey’s mean separation test.
RESULTS
Fatty acid enrichment
The most notable change in MC3T3-E1 cells exposed to CLA-supplemented medium was the enrichment of the CLA isomers and reductions in AA and DHA concentrations (Table 1). Interestingly, the proportion of the CLA isomers in the cellular lipids approximated that of the mixture used (74% c9,t11, 17% cis,cis, and 1% trans,trans isomers). In addition, CLA enrichment resulted in lower concentrations of 16:0, 16:1n–9, 18:0, 18:1n–9, and 22:5n–3 in cellular lipids. We have shown that diets containing supplemental CLA caused a higher CLA content in rodent bone tissue compartments (4). The results demonstrated here indicate that local changes in vivo can occur in osteoblasts that would affect osteoblast function. We recently showed that alterations in cellular lipids of MC3T3-E1 with exogenous n–3 PUFAs positively affected osteoblast bone formation markers while decreasing the amounts of PGE2 after stimulation of COX-2 expression (50).
View this table:
TABLE 1. Fatty acid composition of MC3T3-E1 cells enriched with conjugated linoleic acid (CLA)1
Prostaglandin E2 production
MC3T3-E1 cells exposed to CLA-supplemented medium demonstrated a reduction in PGE2 production similar to that of n–3 PUFA–enriched cells after FSK-stimulated COX-2 expression (Figure 1). The CLA attenuating effect was noted with use of both c9,t11- and t10,c12-CLA isomers. Furthermore, when FSK was added directly to cultures containing the fatty acid–supplemented medium, the presence of the CLA isomers did not increase PGE2 production as observed for cells treated with AA (Figure 1B). Thus, the CLA isomers were effective in attenuating PGE2 synthesis after FSK treatment, and the same was observed for cells enriched with n–3 PUFA. We reported a similar effect of CLA on repressing stimulated PGE2 production in mineralizing bone tissue (4) and in MC3T3-E1 cells (55), an effect also noted with use of n–3 PUFAs (50).
In cultures of human saphenous vein endothelial cells a 50:50 mixture of CLA (c9,t11:t10,c12) reduced AA-derived prostaglandin metabolites, an effect also noted when the isomers were supplemented individually at half the concentration (50 µmol/L) of the mixture (56). In addition, a decrease in 2-series prostanoid production was observed in guinea pigs fed diets containing CLA (43% t10,c12; 41% c9,t11) after antigen exposure in lung, trachea, and bladder tissues (57, 58). CLA enrichment of macrophage cultures was also found to diminish PGE2 production (12, 59).
The attenuation of PGE2 production noted in MC3T3-E1 cells could be through displacement of AA (Table 1) or through a reduction in COX-2 expression and activity. Likewise, Yu et al (12) found that 200 µmol/L of c9,t11-CLA decreased COX-2 promoter activity in RAW 264.7 macrophages. Furthermore, several geometric 9,11-CLA isomers and t10,c12 decreased interferon-–stimulated COX-2 mRNA expression in this macrophage cell line. Moreover, c9,t11 was reported to be the most effective CLA isomer at reducing the oxygenation of AA by COX (a measure of enzyme activity) in ram seminal vesicle microsomes (36). Therefore, the CLA isomers could act at multiple points to decrease stimulated PGE2 production, including displacement of AA from cellular lipids as observed in osteoblasts.
Cyclooxygenase 2 protein level
The c9,t11- and t10,c12-CLA isomers increased COX-2 protein level compared with cells enriched with AA (Figure 2). This effect was observed when the COX-2 inducer was added to culture wells containing the fatty acid–supplemented medium (Figure 2A) and after removal of this medium (Figure 2B); however, the CLA treatment was not different from the fatty acid vehicle control (Figure 2A). Nakanishi et al (60) reported an elevated COX-2 mRNA expression concurrent with a reduction in PGE2 production in cerebral tissue from mice fed a source of CLA (31% c9,t11 and t10,c12). Although a reason for these findings was not presented, one explanation could be that the diminished presence of the preferred COX substrate AA in CLA-enriched cells (Table 1) led to an overexpression of COX-2 mRNA. COX-2 activity is thought to rely on endogenous substrates (61), and, if AA is displaced from membrane lipids by CLA, there will be less substrate for COX-2. Two mechanisms are possible for this observation: Cells could undergo a compensatory response to increase protein expression and extend the half-life of COX-2 during periods of diminished substrate availability, or an inhibition of enzyme activity could cause accumulation of COX-2 protein as demonstrated in mammary epithelial cells (34). We are currently investigating these mechanisms in osteoblasts.
EXPERIMENTS ON OVARIECTOMIZED RATS
We investigated the effects of CLA isomers combined with n–3 PUFAs on bone loss in rats (2-mo-old female Sprague-Dawley) that were either sham-operated (Sham) or ovariectomized (OVX) and were given a basal semipurified diet (based on the American Institute of Nutrition Rodent Diet 93G) modified to contain 25% energy from fat (62). Animal care was in compliance with applicable guidelines of the Purdue University Policy on Animal Care and Use. Sham rats (normal control) and one group of the OVX rats (OVX control) were fed the safflower oil diet, and the other 4 OVX groups were given the basal diet containing a mixture of DHA oil [docosahexaenoic acid–rich single-cell oil (42% DHA); Martek, Columbia, MD] and safflower oil or menhaden oil and safflower oil with or without added CLA [0.3% of diet, CLA One (33.4% c9,t11; 34.4% t10,c12; 1.8% cis,cis; and 1.6% trans,trans); PharmaNutrients, Gurnee, IL] at n–6:n–3 PUFAs of 5:1.
The fatty acid composition of bone tissue compartments (femoral periosteum and marrow), where osteoblasts and osteoblastic progenitor cells reside, was greatly modified by the dietary treatments after 12 wk of feeding. In bone periosteum lipids (Table 2), the n–3 dietary treatments [mixture of safflower oil, docosahexaenoic acid–rich single-cell oil, and CLA (DHAC) and mixture of safflower oil, menhaden oil, and CLA (MENC)] had significantly higher concentrations of n–3 PUFAs (20:5n–3, 22:6n–3, and total n–3) compared with the bone compartments of the OVX control group given safflower oil. Menhaden oil treatment (MENC) also elevated the concentrations of 18:3n–3, 20:5n–3, and 22:5n–3 compared with the other 2 groups. The rats given DHA oil had the highest concentration of 22:6n–3. In contrast, the concentrations of n–6 PUFAs (LA, AA, and 22:4n–6) in bone compartments were greatest in rats fed safflower oil compared with those in the OVX control and n–3 groups. The amounts of saturated fatty acids and monounsaturated fatty acids were not affected by the dietary treatments. A similar response of dietary lipid treatments that altered the fatty acid composition of bone marrow was observed (Table 3). Dietary CLA treatments resulted in higher bone concentrations of the 2 primary isomers found in the lipid source; however, the amount of the c9,t11-isomer exceeded that of t10,c12-isomer by 2-fold in both femoral periosteum and marrow.
View this table:
TABLE 2. Fatty acid composition of femoral periosteum of ovariectomized rats fed different dietary lipid treatments1
View this table:
TABLE 3. Fatty acid composition of femoral marrow of ovariectomized rats fed different dietary lipid treatments1
Serum bone turnover markers were differentially affected by the dietary lipid treatments in rats (Table 4). The bone formation markers, bone-specific ALP and osteocalcin, were not affected by the diets. The concentrations of serum deoxypyridinoline crosslinks, a bone resorption marker, were lowered by the menhaden oil treatment relative to the diet rich in n–6 PUFAs. The addition of CLA to the dietary treatments did not influence bone formation or resorption markers in OVX rats. Our results are consistent with Kelly et al (38) which showed that feeding male growing rats diets containing menhaden oil was associated with a reduced extent of bone resorption compared with those fed a soybean oil diet (rich in n–6 PUFAs). In general, the results from these studies indicated a trend that n–3 PUFA treatment lowered bone turnover rates, which is a critical condition that causes bone loss in both OVX rats and postmenopausal women.
View this table:
TABLE 4. Average serum bone biomarkers of female rats1
Bone mineral content and density were determined by dual-energy X-ray absorptiometry (pDEXA Sabre; Norland Medical Systems, White Plains, NY) on excised femur and tibia bones of OVX rats fed the dietary treatments of CLA and n–3 PUFAs. Conservation of bone mineral was observed in rats fed the DHA oil (DHA and DHAC) compared with those fed the safflower oil diet as indicated by the higher bone mineral content values for femur (Table 5). There was no significant effect of dietary treatments on bone mineral density. In this study, the addition of CLA isomers to the dietary lipid treatments did not affect bone mineral content in OVX rats.
View this table:
TABLE 5. Average bone ex vivo dual-energy X-ray absorptiometry measurements of female rats1
APPLICATIONS TO HUMANS
Few studies have reported the effects of CLA isomers in human subjects; however, those available data suggest no effect or only modest effects on blood lipids, body fat, insulin resistance, and immune functions. On the positive side, several beneficial effects of CLA in human studies were reported. In a double-blind study, patients with type 2 diabetes were given either 8 g/d of CLA (76% total CLA, 37% c9,t11-isomer + 39% t10,c12-isomer, in free fatty acid form) or 8 g/d of safflower oil as the control (63). After 8 wk of intervention, it was found that the plasma concentrations of t10,c12-isomer were inversely correlated with body weight changes as well as serum leptin concentrations, indicating a possible role of this CLA isomer in body weight regulation in these patients. Albers et al (64) showed that by giving 1.7 g/d of a CLA mixture of 50:50 c9,t11:t10,c12, the seroprotective rates in healthy subjects that received hepatitis B vaccination were boosted by >2-fold compared with subjects that received about the same amount of either a different CLA mix (80:20 c9,t11:t10,c12) or safflower oil. Noone et al (65) reported benefits of CLA on circulating lipid profiles in normolipidemic human subjects compared with LA. Administering 3 g/d of the CLA mixture of 50:50 c9,t11:t10,c12 for 8 wk lowered fasting triacylglycerol and VLDL-cholesterol concentrations compared with those who took the 80:20 CLA mixture or LA (all supplements were free fatty acids).
In some cases the investigators indicate that consumption of these isomers by humans should not be recommended at present (66). The consistent anticancer findings with CLA isomers in animal models have yet to justify an application in humans. A recent 6.3-y cohort study in the Netherlands reported a weak, but positive, correlation between CLA intake and breast cancer incidence, similar to total trans-fatty acids and saturated fatty acids (67). Although limited, the effects of CLA on plasma lipids and insulin resistance suggest possible effects of individual isomers. Riserus et al (68) reported the results of a double-blind placebo-controlled trial in which 60 obese men with metabolic syndrome were randomly assigned to 1 of 3 groups that received 3.4 g/d of pure t10,c12-CLA, a CLA mixture, or placebo for 12 wk. At the end of the study, supplementation with t10,c12-CLA significantly increased insulin resistance, lowered HDL cholesterol, and elevated the status of oxidative stress and inflammatory biomarkers in obese men. In another 3-mo study, the use of a CLA supplement clearly induced both nonenzymatic and enzymatic lipid peroxidation in human subjects (69). These findings raise some legitimate concerns about the effectiveness and safety of CLA supplements in humans that must be addressed in future research.
Osteoporosis is a condition of decreased bone mass, which is prevalent in postmenopausal women and places them at risk of fractures. It is a significant health problem in the United States, costing $14 billion annually for treatment (70). During menopause, the osteoprotective effect of estrogen is compromised because of diminishing amounts of this hormone that lead to elevated secretions of IL-1 and TNF- from circulating mononuclear cells, which are strong promoters of osteoclastogenesis (71). These cytokines activate osteoclasts and prolong their survival (72, 73). Other reports showed that estrogen blocked osteoblast apoptosis (74) and indirectly inhibited osteoclast function by enhancing osteoblastic nitric oxide production (75). These findings focus on impaired coupling of bone formation and resorption (76). Further, a decline in the amount of bone growth factors deposited into bone matrix with aging leads to lower bone formation rate (77, 78). Because osteoporosis is a condition of abnormal bone mineral loss in the adult, which is linked to reduced bone formation because of a decline in osteoblast numbers, any dietary fatty acid effect on up-regulating the differentiation and activity of osteoblasts would improve this condition. Factors that contribute to bone loss with aging and osteoporosis include impaired bone modeling in the young (ie, failure to attain maximum peak bone mass) and remodeling in the adult, aging and bone loss associated with estrogen and growth factor decline, diseases of catabolic bone loss, and disuse and unloading (loss of weight bearing resulting in muscle and bone atrophy). In addition, diminished capacity of bone cells (osteoblasts) to react to biomechanical forces would reduce the ability of the skeletal system to respond appropriately to endocrine factors and changes in muscle mass and exercise.
Knowledge of prostanoid formation and nonsteroidal anti-inflammatory drug action resides in understanding the substrate availability and regulation and expression of COX. Two isoforms of this enzyme exist: COX-1 and COX-2. COX-1 is a constitutive enzyme responsible for generating prostaglandins that act physiologically, whereas COX-2 is an inducible enzyme expressed in response to cytokine, growth factor, and tumor promoter stimulators. COX-2 is responsible for production of PGE2 that is associated with inflammatory reactions (arthritis), osteoporosis, and cancer (79). COX-2, but not COX-1, mediates mechanotransduction in bone cells (80) and regulates mesenchymal cell differentiation into the osteoblast lineage that is critically involved in bone repair (81). Selective inhibitors of COX-2 were shown to dose-dependently inhibit osteoclast formation, to reduce bone destruction in bone metastasis of mammary carcinoma cell lines (82), and to reduce osteoclastogenesis and bone destruction caused by murine osteolytic 2472 sarcoma cells in the femurs of male mice (83). Inhibiting the activity of COX-2 with use of NS-398 effectively blocked load-induced osteogenesis on the endocortical surface of the tibia (84). Celecoxib, a selective COX-2 inhibitor, completely blocked the calcium release and osteoclastogenesis induced by various inflammatory factors, except PGE2, in cultures of neonatal mouse calvarial bones (54). Therefore, a promising future awaits nutrient formulations that potentiate a modulating action on PGE2 production by way of COX-2 that are without the adverse effects of drugs. CLA isomers could be potential candidates along with long-chain n–3 PUFAs for this application.
SUMMARY
CLA isomers are a group of unusual unsaturated fatty acids, albeit nutrients, that appear to be involved in a variety of biological functions. The biological and physiologic effects of CLA isomers reviewed herein on various health conditions, eg, cancer, obesity, diabetes, immune function, and osteoporosis, point to an integrative approach and some common mechanisms that involve the regulation of PPAR, COX, and specific transcription factors. These observations are consistent with the premise that CLAs are nutrients and behave as such, which fulfills the principle of Ockham’s Razor attributed to William of Ockham. Therefore, recognizing the potential diverse actions of these nutrients provides an opportunity for cooperative scientific endeavor to investigate the functions of CLA isomers for which the effect is far reaching. With respect to bone metabolism and skeletal health, PPARs have been shown to regulate osteoblasts. PPAR, PPAR, and PPAR1 were expressed in MC3T3-E1 cells endogenously, and pharmacologic PPAR activators appear to affect the maturation of osteoblasts and, hence, bone formation and bone mineral density (85). The biological effect of PPAR activators could be dose related, because, at relatively high concentrations of specific PPAR ligands, osteoblast maturation was inhibited. Activation of PPAR2 either promoted adipogenesis or inhibited osteoblastogenesis, depending on the type of ligands, which indicates that the effects of PPAR activators on these processes are mediated by distinct regulatory pathways, depending on the nature of the ligand, and by selective use of the different ligands. Hence, in an effort to understand obesity, lipid metabolism, and bone biology, a promising approach would be to identify isomers that promote bone formation and reduce obesity (86). Because CLA isomers, as well as other PUFAs, are recognized as natural PPAR ligands, investigating how these isomers modulate transcription factors and their target genes could lead to elucidating the potential benefits of CLA in humans.
Future investigations with CLA should take into account the recognized actions and potential applications of these isomers to specific physiologic states and candidate targets for these nutrients. To study bone metabolism, it is necessary to determine how CLA isomers affect osteoblast and osteoclast functions and their interrelationships. Osteoblasts are the primary cell type responsible for new bone formation in the young and adults. Further studies are needed to examine how CLA isomers affect osteoclastogenesis and the activity of these bone-resorbing cells. Another approach is to determine whether CLA isomers regulate osteoclasts directly or through the mediation of osteoblasts, which has shown to have a major regulatory role in osteoclastogenesis. Thus far, no clear-cut conclusion can be drawn about the effects of CLA on bone remodeling in the adult.
ACKNOWLEDGMENTS
All authors contributed to the preparation of the manuscript and have no financial or personal interest in any company or organization sponsoring the research.
REFERENCES