All experimental procedures were approved by the University of Alberta Animal Ethics Committee and conducted in accordance with the Canadian Council on Animal Care. Twenty four male obese JCR:LA-cp rats (cp/cp) were raised in our established breeding colony at the University of Alberta as previously described 29 . At 3 wk of age, rats were transferred from the isolated breeding colony areas to an individually ventilated caging environment (TecniplastTM, Exton PA, USA) and had access to a standard rat chow diet (5001, PMI Nutrition International) (see Additional file 1). At 8 wk of age, rats (n = 8) were randomized and assigned to one of three diets for 16 wk (control and experimental diets) and had free access to food and water. Age and weight matched lean littermates (n = 8) were fed the control diet to mimic cholesterol/high fat feeding under normolipidemic conditions. Food intake and body weight (BW) were monitored weekly throughout the study. At 23 wk of age, a meal tolerance test (MTT) as previously described 30 , was performed in four randomly chosen rats from control and treatment groups in order to determine plasma glucose and insulin concentrations after a meal. We also performed an oral fat challenge test (OFC), as previously described 29 in four additional rats from each group. At the end of the treatment period (24 wk of age), rats were fasted overnight and anesthetized using isofluorane anesthesia. Plasma was sampled from the left ventricle and heart, liver and fat pads were excised, weighted and immediately frozen at -80°C until analysis. Adipose fatty acid composition was measured from total triglyceride on the epidydimal fat pad as previously described 16 .

Three iso-caloric diets were prepared with a constant polyunsaturated to saturated fatty acid ratio (P:S) of 0.4. A control diet (CD) was supplemented (w/w) with 1% cholesterol and contained 42% of energy from carbohydrate, 23.7% from protein and 34.3% from fat. Experimental diets were prepared by adjusting the lipid composition of the CD to provide 1.0% w/w of cis-9, trans-11 CLA alone (CLA), both 1% of VA and 1% w/w of cis-9, trans-11 CLA (VA+CLA). Semi-purified cis-9, trans-11 CLA (G-c9t11 80:20) containing 59.8% of cis-9, trans-11 CLA and 14.4% of trans-10, cis-12 CLA was kindly provided by Lipid Nutrition. The amount of CLA and VA (1% w/w) was chosen based on previous studies allowing for metabolic sufficiency while maintaining a normal dietary fatty acid proportion 15 16 17 31 . Purified VA was synthesized by a chemical alkali isomerisation from linoleic acid-rich vegetable oil 32 . The diet mixture was extruded into pellets, dried at RT°C and stored at 4°C. Fatty acid composition of the three diets was confirmed by gas chromatograph analysis 28 of the fat blend samples (Table 1).

The concentration of biochemical parameters in fasting plasma from lean and obese groups were assessed using commercially available homogenous, enzymatic colorimetric assays. Triglyceride (TG) (Wako Pure Chemical Industries, catalog no. 998-40391, 0.01 mmol/L minimum), Total cholesterol (TC) (Wako Pure Chemical Industries, catalog no. 993-00404, 0.002 mmol/L minimum), LDL cholesterol (LDL-C) (Wako Pure Chemical Industries, catalog no. 993-00404, 0.03-10.4 mmol/L) and nonesterified fatty acids (NEFA) (HR Series NEFA-HR, catalog no. 999-34691, Wako Diagnostics) were measured using direct colorimetric chemical enzymatic reactions. Plasma glucose was measured as per the glucose oxidase method (Diagnostic Chemical, catalog no. 220-32, 0.03-33.3 mmol/L) and plasma insulin was determined using commercially available enzymatic immunoassays for rodents (Ultrasensitive rat insulin ELISA, Mercodia, catalog no. 80-INSRTU-E01, 0.03-1.0 pmol/L). Samples were analyzed in triplicate using assay kits from a single lot and performed in one single batch.

Liver and adipose tissue samples (0.5 g) were homogenized in 200 μL lysis buffer [PBS (pH 7.4) with 1.5% TritonX-100 and 1% protease inhibitor cocktail (Sigma)] and and hepatic TG levels were determined by a commercially available enzymatic colorimetric assay (Wako Pure Chemical Industries, catalog no. 998-40391, 0.01 mmol/L minimum), using an aliquot of the whole homogenate and adjusting by the protein concentration of the homogenate 15 . The remainder of the homogenate was centrifuged at 700 g for 15 min and the supernatant was collected and stored at -80C for western blot and citrate synthase activity analysis.

Citrate synthase activity in liver and adipose tissue samples was determined using a commercially available kit from Sigma (catalog no. CS0720). The coefficient of variation of this assay in our laboratory is < 10%. The citrate synthase activity was expressed as μmol/min/g protein.

Hepatic acetyl-CoA carboxylase-1 (ACC-1) and fatty acid synthase (FAS) were determined by western blot analysis as described elsewhere 33 with few modifications 15 . ACC-1 and FAS relative abundance were normalized based on the respective β-actin protein mass (internal control).

Statistical analysis was performed using the Graph pad Prism software, version 4.0. Data was tested for normal distribution and one-way ANOVA followed by Tukey post-hoc tests were used to identify differences among both lean and obese controls and treatment groups (CLA and VA+CLA). Post-prandial glucose and insulin metabolism as well as post-prandial TG response were assessed by area under the curve (AUC) analysis. Fasting concentrations of these parameters were further subtracted from the total AUC to yield the incremental area under the curve (iAUC). Results are expressed as means ± SEM and the level of significance was set at p < 0.05.

Obese rats fed the combination of VA+CLA showed increased food intake compared to those obese rats fed either the CD or CLA diet. Paradoxically, the VA+CLA fed rats showed reduced body weight (p < 0.001) compared to the CLA group (Table 2). Despite the higher body weight of rats fed the CLA diet as compared to obese control, no difference was observed in absolute and relative heart weights or fat pad deposition (p > 0.05), as measured by the amount of absolute and relative perirenal and inguinal fat pad weights compared with the obese rats fed the CD (Table 2). In contrast, feeding either the CLA or the VA+CLA diet resulted in a lower absolute liver weight by 15% and 26%, respectively, and both diets reduced the ratio of liver weight to total body weight by 22% as compared to obese rats fed the CD (p < 0.001).

The activity of citrate synthase in liver and adipose tissue is shown in Figure 1. There was a higher (p < 0.001) citrate synthase activity in liver and inguinal adipose tissue after feeding either CLA or VA+CLA diets as compared to lean and obese rats fed the CD. Interestingly, the citrate synthase activity in liver and adipose tissue did not differ between lean and obese rats fed the CD (p > 0.05).

The fatty acid composition in adipose tissue triglyceride was shown to directly reflect the dietary fatty acid composition as shown in Table 3. Obese rats fed with the CLA and VA+CLA diets had a markedly increased proportion of the cis-9, trans-11 CLA isomer relative to the obese control group. The content of this isomer, expressed as a percentage of total fatty acids was different between the treated groups. The CLA diet showed the greatest incorporation of the cis-9, trans-11 CLA isomer (70 fold greater than control group) compared to the VA+CLA diet, which was only 46 fold higher than the obese control group. Rats fed VA+CLA diet showed a greater incorporation of trans-11 18:1 (VA) compared to CLA and obese control rats (p < 0.001). Interestingly, rats fed the VA+CLA diet had lower proportions of linoleic acid (18:2 n6), α-linolenic acid (18:3 n3) and arachidonic acid (20:4 n6).

As shown in Table 4, fasting plasma TG, TC and LDL-C were significantly lower in obese rats fed either the CLA or the VA+CLA diet, as compared to the CD. However, feeding the VA+CLA diet further reduced plasma NEFA concentration relative to the CLA fed rats and lowered TG and LDL-C concentrations not different from lean rats fed the CD (p > 0.05). The CLA or VA+CLA diet lowered fasting insulin to that comparable of lean rats (p > 0.05) and reduced total insulin concentration (AUC) after the MTT (p < 0.05). There was no statistical difference between groups for either glucose metabolism (fasting or iAUC) or the relative change (iAUC) in insulin.

Obese rats had a higher post-prandial plasma TG response (iAUC) compared to lean rats following an OFC (Figure 2). AUC analysis showed an improved total TG concentration (p < 0.05) over the 10-h post-prandial period in rats fed either the CLA or VA+CLA diet (35 ± 5 and 38 ± 8 mmol/L.h, respectively), compared to obese rats fed the CD (62 ± 5 mmol/L.h). However, the post-prandial iAUC for TG was not different between obese control and obese treatment groups (CLA and VA+CLA).

Liver TG concentration was higher in obese control rats compared to lean control rats. However, feeding the CLA diet resulted in a liver triglyceride concentration that was 22% lower than the obese rats fed the CD (Table 4). Interestingly, the VA+CLA diet further lowered liver TG concentration by 43% and 27% as compared to obese control and CLA groups, respectively. In addition, CLA and VA+CLA diets resulted in significantly lower hepatic ACC-1 protein abundance relative to obese control (34% and 38%, respectively) and was normalized to concentrations similar to lean rats fed the CD (p > 0.05) (Figure 3A, B). The relative abundance of hepatic FAS protein did not differ between the obese groups (p > 0.05) (Figure 3A, C).

Dyslipidemia and insulin resistance are common features associated with cardiovascular disease and the metabolic syndrome. The homozygous obese (cp/cp), JCR:LA-cp rat, has a complete absence of the leptin receptor. As a consequence, the JCR:LA-cp rat spontaneously develops hyperphagia and obesity, associated dyslipidemia, insulin resistance, and macro-and micro-vascular dysfunction 34 35 36 . In the present study, we used this unique animal model to evaluate hypolipidemic effects of cis-9, trans-11 CLA in combination with VA. Obese rats fed the CLA diet had a higher final body weight relative to obese control rats. However, this was not associated with an increase in fat deposition. Cis-9, trans-11 CLA has been observed to regulate metabolic pathways involved in fatty acid oxidation as well as energy production and thermogenesis 37 38 39 40 ; therefore we investigated whether combination VA+CLA might promote fatty acid oxidation.

Peroxisome proliferator-activated receptor (PPAR)-agonists, such as thiazolidenediones (TZDs), are effective drugs for the treatment of type 2 diabetes by inducing adipogenesis as well as increasing the uptake and metabolism of free fatty acids in adipose tissue. Increased mitochondrial oxidative capacity of white adipose tissue has been observed after treatment with TZD 41 42 . CLA is a natural PPAR ligand 43 44 45 and has been observed to promote fatty acid oxidation 40 or TG synthesis in adipose tissue contributing to lower circulating NEFA and TG concentrations 7 . We observed a significant increase in citrate synthase activity in liver and inguinal adipose tissue following supplementation with either CLA or VA+CLA diet but not in perirenal adipose tissue. Interestingly, citrate synthase activity was not different between lean and obese control rats. It has been reported that insulin resistance is associated with an increase in muscle mitochondrial content and oxidative capacity 46 47 48 . Similarly, it has been demonstrated that mitochondrial biogenesis increases during adipose tissue differentiation 49 . Therefore, it can be proposed that JCR:LA-cp rats maintain similar mitochondrial fatty acid oxidation relative to lean rats but during increased dietary lipid consumption, it is insufficient to prevent TG deposition. Treatment with CLA or VA+CLA may stimulate mitochondrial fatty acid oxidation and in turn this may contribute to improvements in liver and adipose tissue metabolism in the JCR:LA-cp rat.

It is well established that the fatty acid composition of adipose tissue is dependent on dietary intake. However, endogenous synthesis of fatty acids, fatty acid transport and inter-conversion processes (elongation and desaturation) are also significant contributing factors to the composition of adipose tissue 50 . As expected, supplementation with cis-9, trans-11 CLA (CLA diet) and VA (VA+CLA diet) increased the proportion of these fatty acids in adipose tissue from obese rats. We also wish to note that the endogenous synthesis of cis-9, trans-11 CLA from VA may have also occurred in this study. In both humans and animals, the conversion of dietary VA to CLA has been reported to be at a rate of approximately 12-19% [12 and 14]. We also note that the dietary ratio of VA:CLA in this study was ~1.5:1, respectively (Table 1) and that the resultant incorporation of these fatty acids into adipose tissue was found to be 1:1 following supplementation (Table 3). While we cannot infer a rate of conversion from VA to CLA per se from this data, it would support that previously published.

One of the most striking effects of the combined treatment (VA+CLA) in the present study was the reduction in hepatic TG concentration. The improvement in fasting lipid parameters by VA+CLA diet (i.e. TG and LDL-C) suggest an additional benefit with the combination diet and is consistent with previous observations that dietary VA has lipid lowering properties independent from CLA 15 . We also wish to note that both diets (CLA and VA+CLA) consistently contained 0.15 w/w of the trans-10, cis-12 CLA isomer, which is also known for its hypolipidemic properties.

ACC-1 and FAS are two key lipogenic enzymes involved in the synthesis of fatty acids and subsequent TG synthesis. TG is then either stored as lipid droplets within the hepatocyte, secreted into the blood compartment as VLDL or hydrolyzed via oxidation 51 . It is plausible that reduced hepatic lipogenesis may contribute (at least in part) to reduced hepatic TG in rats fed either the CLA or VA+CLA diet, which is supported by a lower hepatic ACC-1 protein abundance relative to obese rats fed the CD. We have reported previously that VA may act in part via ACC-1 and FAS pathways resulting in reduced VLDL secretion to decrease circulating concentrations of plasma TG and LDL-C 15 . However, in conditions of insulin resistance, hepatic steatosis is thought to be caused by an increased free fatty acid flux from adipose tissue into the liver 52 . As discussed above, rats fed the combined treatment (VA+CLA) showed lower circulating NEFA concentrations compared to rats fed the CLA diet alone. We propose that reduced NEFA may also contribute to a further decrease in hepatic TG. In addition, activation of ACC is regulated by phosphorylation/dephosphorylation 53 and thus, it may be possible that CLA and VA+CLA diets may differently regulate post-translational modifications of ACC-1.