Hepatic Metabolic, Inflammatory, and Stress-Related Gene Expression in Growing Mice Consuming a Low Dose of Trans-10, cis-12-Conjugated Linoleic Acid

Dietary trans-10, cis-12-conjugated linoleic acid (trans-10, cis-12-CLA) fed to obese and nonobese rodents reduces body fat but leads to greater liver mass due to steatosis. The molecular mechanisms accompanying such responses remain largely unknown. Our study investigated the effects of chronic low trans-10, cis-12-CLA supplementation on hepatic expression of 39 genes related to metabolism, inflammation, and stress in growing mice. Feeding a diet supplemented with 0.3% trans-10, cis-12-CLA (wt/wt basis) for 6 weeks increased liver mass and concentration of long-chain fatty acids (LCFAs) in liver, while adipose tissue mass decreased markedly. These changes were accompanied by greater expression of genes involved in LCFA uptake (Cd36), lipogenesis, and triacylglycerol synthesis (Acaca, Gpam, Scd, Pck1, Plin2). Expression of these genes was in line with upregulation of the lipogenic transcription factor Srebf1. Unlike previous studies where higher >0.50% of the diet) doses of trans-10, cis-12-CLA were fed, we found greater expression of genes associated with VLDL assembly/secretion (Mttp, Cideb), ketogenesis (Hmgcs2, Bdh1), and LCFA oxidation (Acox1, Pdk4) in response to trans-10, cis-12-CLA. Dietary CLA, however, did not affect inflammation- and stress-related genes. Results suggested that a chronic low dose of dietary CLA increases liver mass and lipid accumulation due to activation of lipogenesis and insufficient induction of LCFA oxidation and VLDL assembly/secretion.

Despite the clear deleterious effects observed in rodents, the European Food Safety authority after an evaluation of the available data concluded that consumption of 3.5 g CLA (or 4.5 g product/d) from a commercial product containing equal proportions of trans-10, cis-12-CLA and cis-9, trans-11-CLA was not deleterious to health [14]. However, at the molecular level, hepatic steatosis in rodents due to trans-10, cis-12-CLA at ≥0.5% of the diet has been linked to alterations in the expression of genes associated with energy expenditure, apoptosis, fatty acid oxidation, lipolysis, differentiation, and lipogenesis [1,15]. Therefore, it is likely that sustained supplementation of the above amount, which is much greater Table 1: Summary of the changes in liver mass, liver TG, and adipose tissue mass observed in previous mouse studies feeding different levels of trans-10, cis-12-CLA compared with feeding linoleic acid (LA) or oleic acid (as canola oil).

Materials and Methods
2.1. Animals, Diets, Sampling, and Fatty Acid Analysis. All experiments were approved by the Institutional Animal Care and Use Committee (IACUC) at Virginia Tech. Male CD-1 mice from Harlan (Madison, Wis, USA), 3 to 4 weeks of age, were divided into two experimental groups of ten mice each. Mice were housed individually with access to food and water at all times. A 12-h light/12-h dark cycle was maintained throughout the study. For 7 d before the start of the study, all mice were fed Harlan Teklad (Global Rodent Diet, 18% protein, 5% fat; Harlan, Madison, WI, USA) with 3% (w/w) high-oleic sunflower oil added. On day 1 of the study, mice were randomly assigned to receive diets containing 3.0% high-oleic sunflower (control) or 2.70% high-oleic sunflower oil + 0.30% trans-10, cis-12-CLA. The free FA form of trans-10, cis-12-CLA (>95% purity) was obtained from Natural Lipids (Hovdebygda, Norway). Mice were fed daily at 1600 hours. At the end of weeks 2 and 6, five mice per dietary treatment were anesthetized using Metofane (Pitman-Moore, Inc., Washington Crossing, NJ, USA) prior to cervical dislocation. The liver was removed, rinsed with diethylpyrocarbonate (Sigma, St. Louis, MO, USA) in distilled water (1 : 1000, v/v), and weighed. A portion of liver was frozen in liquid nitrogen and later stored at −80 • C prior to RNA extraction and gene expression analysis.
Total LCFA in liver were extracted as described previously [17]. Undecenoic acid (Nu-Check Prep, Inc., Elysian, Min, USA), added to extracted lipids prior to formation of fatty acid methyl esters (FAME) [18], served as an internal standard. Using an Agilent 6890N gas chromatograph equipped with a flame ionization detector (Agilent Technologies, Palo Alto, Calif, USA), FAME were separated with a 100 m × 0.25 mm i.d. (0.2 μm film thickness) CP-Sil 88 column (Varian, Lake Forest, Calif, USA) protected by a 2.5 m × 0.25 mm i.d. fused silica retention gap. Ultrapure hydrogen at constant pressure was the carrier gas. Pure methyl ester standards (Nu-Check Prep, Inc., Pa, USA) were used for identification and quantification of sample FA [17].

Additional Materials and
Methods. Specific details of RNA extraction and quantitative real-time RT-PCR (qPCR) are reported in additional file 1 (supplementary material). qPCR performance, primer features, and sequencing are reported in Tables S1-S4 (additional file 1), (see Supplementary material available online at doi:10.1155/2012/571281). Results of internal control gene (ICG) evaluation are shown in Figure S1 (additional file 1).

Statistical
Analysis. Data were analyzed as a completely randomized design with a factorial arrangement of treatments using the MIXED procedure of SAS (Windows version 9.1, Cary, NC, USA). Normalized gene expression data using the geometric mean of the top 4 most-stable ICG ( Figure  S1, additional file 1) were log-transformed prior to statistical analysis. The model for gene expression and liver LCFA profiles included the fixed effects of dietary treatment (0% or 0.30% CLA), time (week 2 and 6), and the interaction of dietary treatment × time. Mouse within dietary treatment × time was the random effect. Least-squares means ± standard error of the means (SEM), (combined observations for weeks 2 and 6) are presented in the tables.

Tissue Weight.
There was no difference in final live body weight between treatments, and the average was 37.1 g Journal of Lipids 3  ( Table 2). Liver weight was ∼19% greater in mice fed CLA and averaged 2.36 g versus 1.99 g in controls. In contrast, the weight of epididymal adipose tissue was 23% lower in CLAfed mice compared with controls, that is, 0.70 g versus 0.91 g. Carcass weight was lower in CLA-fed mice compared with controls, averaging 12.8 g versus 14 g.

Markers of Fatty Acid Uptake in Liver. The expression of
Cd36 was greater (CLA = 1.11 versus control = −1.47; P = 0.001; Table 4) with trans-10, cis-12-CLA and there was no change due to treatment for Slc27a2 expression.

Lipogenesis and TG Synthesis.
We observed that Acaca  SEM: the largest standard error of the mean is shown.

Transcription Regulation and Lipogenesis and TG Synthesis.
Although the expression of glucose-regulated transcription factor Chrebp did not differ between treatments, we observed that Srebf1 expression in mice fed trans-10, cis-12-CLA was upregulated (CLA = 0.14 versus control = −0.22).

Carbohydrate Metabolism.
We observed no treatment effects in the expression of genes associated with glucose transport (Slc2a1), glycolysis (Gck), and gluconeogenesis (Pc).

Lipolysis.
We observed no treatment effects in the expression of the lipolytic genes Lipc and Pnpla2.

Discussion
Several studies have shown that trans-10, cis-12-CLA at ≥0.5% of the diet reduced adipose mass and increased liver mass (summarized in Table 1). In the present study, we fed a low dose of trans-10, cis-12-CLA that may be Journal of Lipids 5  considered more physiologically relevant, compared with previous studies, from a nutritional standpoint to examine any potential links between adipose mass loss, liver mass, and expression of genes related to various aspects of lipid metabolism. Hepatic enlargement has been associated with a 3-to 7-fold increase in TG content (Table 1), and we also observed an increase in liver mass and accumulation of LCFA (Tables 2 and 3).

Trans-10, cis-12-CLA Increased Markers of Fatty Acid
Uptake in Liver. Our results are in agreement with previous data showing two-fold greater hepatic LCFA content [7] and lower serum LCFA concentration in mice fed trans-10, cis-12-CLA [13]. Therefore, upregulation of Cd36 expression with trans-10, cis-12-CLA allows for adipose tissue-derived-LCFA uptake into hepatocytes and, in fact, may represent a mechanism unique to this isomer because it has been shown in vitro that uptake of trans-10, cis-12-CLA in rat hepatocytes was 3X greater than for oleate [19]. Furthermore, a previous study showed not only a significant increase in hepatic Cd36 expression in mice fed trans-10, cis-12-CLA but also a significant association between Cd36 expression and the degree of hepatic steatosis [20]. Therefore, upregulation of Cd36 at low dietary trans-10, cis-12-CLA is part of the mechanism leading to increased hepatic fatty acid uptake and hepatic steatosis.

Trans-10, cis-12-CLA Stimulates Lipogenesis and TG Synthesis in Liver.
Several authors have concluded that de novo fatty acid synthesis via acetyl-CoA carboxylase (Acaca) and fatty acid synthase (Fasn) may play a role in the onset of hepatic steatosis related to CLA supplementation [21,22].
The upregulation of Acaca with CLA suggested that the liver was generating more malonyl-CoA which was then used for palmitic acid synthesis by the multifunctional enzyme Fasn. The numerically greater concentration of 16 : 0 in CLA-fed mice (24.1% versus 22.5%) provides some support for such an effect. Ashwell et al. [15] and Kelley et al. [7] observed greater 16 : 0 in liver tissue of obese mice fed trans-10, cis-12-CLA than those fed a linoleic acid or corn oil. Stearoyl-CoA desaturase (Scd) is the rate-limiting enzyme involved in the synthesis of mono-unsaturated fatty acids (cis-9-16 : 1 or cis-9-18 : 1) from saturated fatty acids (16 : 0 or 18 : 0) [23]. Although not in the glycerol-3-phosphate pathway of TG synthesis, Scd plays an important role in the synthesis of TG [23]. The expression of the mouse Scd gene is regulated primarily at the level of transcription [24], and its expression in adipose tissue decreases as the trans-10, cis-12-CLA content of the diet increases [6]. The upregulation of Scd we observed with CLA confirmed results from Guillén et al. [20] in mice. Furthermore, the upregulation of Scd in our study correlated with the lower concentration of 18 : 0 (substrate) and greater concentration of cis9-18 : 1 (main product), (Table 3). Because a large number of studies have shown that trans-10, cis-12-CLA reduces body fat mass considerably, the upregulation of hepatic Scd mRNA due to feeding trans-10, cis-12-CLA may be a response to handle the influx of 16 : 0 and 18 : 0 derived from adipose tissue lipolysis.
The initial reaction in the pathway of TG synthesis is catalyzed by glycerol-3-phosphate acyltransferase (Gpam) which resides in the outer mitochondrial membrane [25] and it is the first committed step in the synthesis of TG [25] via the glycerol phosphate pathway. In addition, Gpam is an enzyme that can determine the fate of LCFA between β oxidation or glycerolipid synthesis [26]. Subsequent steps of TG synthesis are regulated by Dgat1 and Dgat2, while Pck1 provides a source of glycerol-3-phosphate during glyceroneogenesis [27]. Perilipin (Plin) proteins are vital for cytoplasmic lipid droplet (LD) formation [28]. In isolated hepatocytes, as well as livers from mice and humans, Plin2 levels are proportional to hepatic lipid content [29].
Our results for Gpam expression are contrary to a previous study which found that trans-10, cis-12-CLA did not change the hepatic expression of Gpam in mice [15]. The concerted upregulation of Acaca, Gpam, Pck1, and Plin2 could be taken as a mechanistic response to promote biosynthesis of TG and lipid droplet formation, for example, the upregulation of Acaca stimulates de novo FA synthesis, thus, generating more precursors (i.e., feed-forward effect) for TG synthesis leading to stimulation of both Gpam and Pck1 expression. In addition, Plin protects lipid droplets from cAMP-mediated lipolysis [30], thus, an increase in Plin2 expression represents attenuation of lipolytic enzymemediated TG breakdown within the lipid droplet. Those scenarios also would allow for blood LCFA derived from adipose tissue lipolysis to be esterified into TG within liver. Our results support the hypothesis from studies feeding much higher doses of trans-10, cis-12-CLA that hepatic lipogenesis and esterification are primary factors responsible for liver TG accumulation [5,21,22].

Trans-10, cis-12-CLA Increases VLDL Assembly and
Secretion. In the liver, synthesized TG is either stored in cytoplasmic droplets or secreted as VLDL particles [31]. Most de novo-synthesized TG is stored in cytosolic TG pools and a smaller portion is secreted in the form of VLDL. Apoprotein B 100 (Apob) is the key component whose rate of synthesis in the rough endoplasmic reticulum controls the overall rate of VLDL production [32]. Lipid components are added to apoprotein B by microsomal TG transfer protein (Mttp). Because Mttp catalyzes the transfer of lipids to the ApoB molecule, Mttp may play a crucial role in the overall assembly and secretion of VLDL in the liver. In addition to these proteins, recent work has uncovered a potentially important function for the cell death-inducing DFF45-like effector (Cide) isoform b (Cideb) protein in the process of VLDL synthesis, for example, liver of Cideb-null mice had higher levels of cytosolic TG accompanied by lower levels of VLDL secretion [33]. Cideb is localized to smooth ER and LD and interacts physically with ApoB [33].
Despite the well-described accumulation of TG in the liver due to CLA (Table 1), which leads to greater tissue mass and greater LCFA (Tables 2 and 3), our results suggested that the VLDL assembly/secretion mechanism in mice fed trans-10, cis-12-CLA was not impaired and may have actually been enhanced via the upregulation of hepatic Mttp and Cideb. It was reported previously that in trans-10, cis-12-CLAfed mice the VLDL secretion rate was increased; whereas, the plasma TG concentration was decreased [12]. Taken together, results suggest that the enhanced VLDL secretion rate is insufficient to eliminate excess fatty acids (Table 2) entering the esterification and VLDL assembly/secretion pathways resulted from increased fatty acid uptake, which would account for part of the fat deposited within liver cells. Our finding with Mttp is in agreement with recent kinetic studies showing that the actual amount of hepatic TG-rich lipoprotein secretion rate during nonalcoholic fatty liver disease is actually increased, but it is inadequate to match the increased TG synthesis in the liver [34]. The fact that Cideb requires both its ApoB-binding and LD association domains to allow for secretion of TG-enriched VLDL particles seems to support our findings, that is, Cideb also is required to promote the formation of TG-enriched VLDL particles [33].

Transcription Regulation and Lipogenesis and TG Synthesis.
Sterol regulatory element binding transcription factor 1 (Srebf1) is the key transcription factor regulating hepatic lipogenesis in rodents [35]. Its upregulation with CLA is in agreement with the data on lipogenic genes from the current study as well as previous reports on mice fed higher levels of CLA-mixtures [5]. It is unclear, however, what signal might have triggered upregulation of Srebf1 in our study, but insulin may have played a role. In freshly isolated hepatocytes, Srebf1 mRNA was activated by insulin [36]. Thus, if feeding CLA in our study resulted in greater blood insulin, as previously reported [5,9], it would explain the upregulation of Srebf1. The coordinated greater expression of the Srebf1 target genes Acaca, Scd, and Gpam in mice fed trans-10, cis-12-CLA suggested that the liver was generating more palmitic acid and oleic acid (see Table 3) and esterifying it into TG.
Collectively, these results indicated that low dietary trans-10, cis-12-CLA stimulated Srebf1 leading to an overall increase in fat accumulation in liver.

Pparα and Lipid
Catabolism due to trans-10, cis-12 CLA. Isomers of CLA have some structural features similar to peroxisome proliferators and, importantly, the physiological responses observed in mice fed high doses of CLA (e.g., reduced body mass, hepatic lipid accumulation, hypolipidemia) are characteristic of this group of chemicals [37]. The ligand-activated transcription factor peroxisome proliferator-activated receptor α (Ppara) modulates lipid catabolism partly via activation of Lpl [37], Acox1 [38], Cpt1a, thioesterases, fatty acid binding protein, peroxisomal and mitochondrial β oxidation enzymes, microsomal ωoxidizing enzymes, and apolipoproteins [35]. Trans-10, cis-12-CLA has been reported to be a potent ligand of Ppara [37] and, despite the lack of change in Ppara expression, the upregulation of several of its target genes playing important roles in lipid metabolism (Cd36, Pdk4, Acox1, Ucp2) provided indirect evidence that trans-10, cis-12-CLA might act through Ppara.
Because a decrease in fatty acid oxidation may cause fat accumulation in liver [13], it is intuitive that the mRNA expression of the enzymes related to fatty acid oxidation may be downregulated during liver lipidosis. However, recent investigations of trans-10, cis-12-CLA effects on fatty acid oxidation are controversial. Rasooly et al. [39] found that trans-10, cis-12-CLA supplementation decreased the expression of Cpt1a (−77%), Acox1 (−50%), and Pparα (−65%). However, in the study of Gruffat et al. [19] liver tissue samples of rats incubated with a fatty acid mixture (representative of circulating LCFA) and trans-10, cis-12-CLA or oleate, they found that the rate of CLA, isomer oxidation was 2X higher than that of oleate. Furthermore, expression of hepatic fatty acid oxidation genes in mice also increased in three studies with mixtures of CLA isomers [21,22] and in one study with trans-10, cis-12-CLA [13].
Cpt1a is considered the rate-limiting enzyme for mitochondrial β oxidation [39]. Degrace et al. [13] found an increase in the activity of hepatic Cpt1a in mice supplemented with trans-10, cis-12-CLA accompanied by an increase in Acox1 activity and gene expression. The lack of change in Cpt1a that we observed with CLA may have been associated with the upregulation of Acaca due to CLA, that is, greater flux through Acaca would have generated malonyl-CoA which is a potent inhibitor of Cpt1a [32]. Because mice were fed over a 6-week period in the current study, it is possible that adaptive changes occurred wherein the increase in Cpt1a was prevented by the elevated levels of malonyl-CoA. As a consequence of this effect on Cpt1a, more LCFA would have been available for esterification into TG and accumulation in the cytosol.
In addition to mitochondrial oxidation, peroxisomes also play an important role in LCFA oxidation [32]. Acyl-CoA oxidase 1 (Acox1) catalyzes the first and rate-limiting enzyme of the peroxisomal fatty acid β oxidation pathway [32], and mice lacking Acox1 developed severe steatohepatitis [40]. The upregulation of Acox1 with CLA that we observed suggested an increase in fatty acid β oxidation in peroxisomes. The upregulation of Acox1 might have been due to the relatively lower amount of dietary trans-10, cis-12-CLA supplementation compared with previous studies [5, 9-11, 15, 41]. Furthermore, Acox1 is not affected by malonyl-CoA as Cpt1a and, as such, would represent a mechanism for excess LCFA (e.g., incoming LCFA from adipose catabolism) catabolism to continue on the face of greater lipogenic rates.
Regulation of the pyruvate dehydrogenase complex (PDC) is an important step in fuel selection of energy utilization in animals during different nutritional and hormonal states as the modulation of PDC activity impacts fatty acid as well as pyruvate and glucose metabolism [42]. Phosphorylation of PDC via pyruvate dehydrogenase kinase (Pdk) inhibits its activity [42], thereby reducing the conversion of pyruvate to acetyl-CoA and sparing glucose oxidation in favor of LCFA. The abundance of the Pdk4 isoform, which is highly expressed in liver, heart, and skeletal muscle, is transcriptionally controlled by Pppara [42]. The upregulation of Pdk4 with CLA provided evidence of reduced glucose oxidation and supports the view that peroxisomal oxidation of LCFA might have been increased. Because expression of Pdk4 is induced by high-fat diets and LCFA [43], the greater Pdk4 expression was likely a consequence of trans-10, cis-12-CLA intake and its catabolic effect on adipose tissue (Tables 1  and 2).
Under conditions of increased fatty acid uptake and oxidation the liver produces large amounts of acetoacetate and D-3-hydroxybutyrate during the process of ketogenesis [32]. Ketogenesis is controlled indirectly by Cpt1a (i.e., via provision of acetyl-CoA) and directly by the activity of the mitochondrial key regulatory enzyme 3-hydroxy-3-methylglutaryl-CoA synthase 2 (Hmgcs2), which is a Ppara target [44]. The enzyme 3-hydroxybutyrate dehydrogenase 1 (Bdh1) plays a key role in redox balance and energy metabolism since in the presence of NADH, the hepatic Bdh transforms acetoacetate into D-3-hydroxybutyrate, which is then transported through the blood stream to peripheral tissues, for example, brain, heart, and kidney [45]. Our results support previous findings of greater hepatic Hmgcs2 and Bdh1 expression in mice fed trans-10, cis-12-CLA. Thus, influx of LCFA from adipose lipolysis not only enhanced peroxisomal oxidation but also resulted in increased ketogenesis, which was in agreement with a previous study [46]. Together, data suggested that there was a stimulation of ketone body production in liver due to feeding CLA.
4.6. Trans-10, cis-12-CLA Changes Energy Metabolism in Liver. Uncoupling proteins (Ucp) belong to a family of mitochondrial anion carriers and are present in the mitochondrial inner membrane [47]. Ucp dissipate the proton gradient by allowing the reentry of protons into the mitochondrial matrix during oxidative ATP generation, resulting in the uncoupling of the respiratory chain and heat production [48]. In vivo studies indicated that physiological and pathological elevation of blood LCFA resulting from fasting [49] or high-fat diets [50] induced upregulation of Ucp2. Peters et al. [51] demonstrated that dietary CLA increases Ucp2 expression in murine liver, which we confirmed. The uncoupling 8 Journal of Lipids process might have served as a counterregulatory mechanism to lower cellular ATP levels and decreased metabolic efficiency, thereby serving as a mechanism to reduce fat accumulation in the long term [52]. 4.7. Trans-10, cis-12-CLA, and the Stress Response. According to the complex mechanisms proposed by Jaudszus et al. [6], the antiadipogenic effect of trans-10, cis-12-CLA in adipose tissue may be primarily the consequence of proapoptotic and proinflammatory responses, including the nuclear factor κB-(NFκB-) dependent production of tumor necrosis factor-α, interleukin-6, and interleukin-8 [8,9]. The lack of alteration in the expression of several genes associated with inflammation or stress response was likely related to the relatively low dosage of trans-10, cis-12-CLA that was fed.
In conclusion, our study showed that even at one of the lowest dietary doses of trans-10, cis-12-CLA studied to date, the coordinated upregulation of genes associated with fatty acid uptake, TG synthesis, and lipid droplet formation as well as insufficient induction of VLDL assembly/secretion contribute to enhanced TG accumulation and greater liver mass. Impaired mitochondrial β-oxidation did not appear to be a factor contributing to hepatic steatosis, partly due to a greater contribution of the peroxisomal pathway.