In laying hens, hepatic lipid metabolism plays an important role for the production of eggs. Compared with mammals, laying hens have a very high rate of hepatic synthesis of TAG, phospholipids and cholesterol, which plays a crucial role in lipid deposition in egg yolk(Reference Walzem, Hansen, Williams and Hamilton1). Lipids synthesised in the liver are incorporated into TAG-rich lipoproteins that are secreted into the blood. Plasma of laying hens therefore contains extremely high concentrations of TAG, most of which are localised in VLDL(Reference Hermier, Forgez, Williams and Chapman2). VLDL with a particle diameter of 25 to 44 nm are bound to specific oocyte receptors and are deposited in developing egg yolk follicles(Reference Walzem, Hansen, Williams and Hamilton1). An impairment of hepatic lipid synthesis leads to a stop of egg production(Reference König, Kluge, Haase, Brandsch, Stangl and Eder3).
Hepatic lipid metabolism is regulated by transcription factors such as PPARα and sterol regulatory element-binding proteins (SREBP). PPARα is a member of the nuclear receptor superfamily. In mammals, it is highly expressed in tissues with high fatty acid oxidation such as liver or muscle(Reference Desvergne and Wahli4). PPARα regulates the expression of target genes by binding to DNA sequence elements as heterodimers with the 9-cis retinoic acid receptor after activation. PPARα target genes in rodents and human subjects are mainly involved in cellular fatty acid uptake and intracellular fatty acid transport, mitochondrial and peroxisomal fatty acid oxidation, ketogenesis and gluconeogenesis(Reference Mandard, Müller and Kersten5). PPARα is activated by lipid soluble compounds such as eicosanoids, fatty acids or fibrates(Reference Desvergne and Wahli4). In mammals, PPARα activation leads to a strong reduction of plasma and liver TAG concentration due to an enhanced β-oxidation of fatty acids, an inhibition of TAG synthesis and an increased activity of lipoprotein lipase(Reference Staels, Dallongeville, Auwerx, Schoonjans, Leitersdorf and Fruchart6).
SREBP are transcription factors which have been identified and recognised as key regulators of lipid synthesis and homeostasis. SREBP-1c preferentially activates genes required for fatty acid synthesis while SREBP-2 preferentially activates the LDL receptor gene and various genes required for cholesterol synthesis such as 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) reductase(Reference Horton, Goldstein and Brown7). SREBP are synthesised as inactive integral endoplasmic reticulum membrane proteins and are activated by proteolytic cleavages in the Golgi releasing the mature N-terminal domain of SREBP that then translocates to the nucleus and activates transcription of sterol regulatory element-containing genes(Reference Horton, Goldstein and Brown7–Reference Goldstein, Rawson and Brown9).
Recently, it has been shown that both PPARα and SREBP are expressed in chicken liver and that they have a high homology with mouse, rat and human PPARα and SREBP, respectively(Reference Diot and Douaire10–Reference Meng, Li, Zhao and Gu13). We have recently shown that feeding laying hens clofibrate, a synthetic PPARα agonist, causes up-regulation of several genes recognised as typical PPARα target genes in rodents and human subjects which in turn led to dramatically reduced concentrations of TAG in liver and plasma of the hens(Reference König, Kluge, Haase, Brandsch, Stangl and Eder3). Moreover, it has also been shown that SREBP-1 is the major regulator of lipogenesis and that SREBP-2 controls the cholesterol biosynthetic pathway in chick liver as in mammals(Reference Matsuyama, Sato, Nakamura, Suzuki and Akiba14).
In mammals, both SREBP and PPARα are also activated by various naturally occurring lipids such as n-3 fatty acids or conjugated linoleic acids (CLA)(Reference Moya-Camarena, Vanden Heuvel and Belury15–Reference Clarke17). Long-chain n-3 PUFA activate PPARα, which stimulates mitochondrial and peroxisomal β-oxidation of fatty acids, and reduces the transcription of SREBP-1, which leads to an inhibition of de novo fatty acid synthesis due to a down-regulation of lipogenic enzymes(Reference Kim, Takahashi and Ezaki18–Reference Le Jossic-Corcos, Gonthier, Zaghini, Logette, Shechter and Bournot20). These events are involved in the hypotriacylglycerolaemic effects of n-3 PUFA in mammals. CLA have also be shown to be activators of PPARα(Reference Moya-Camarena, Vanden Heuvel and Belury15, Reference Moya-Camarena, Vanden Heuvel, Blanchard, Leesnitzer and Belury16, Reference Ringseis, Saal, Muller, Steinhart and Eder21); in contrast to n-3 PUFA, CLA however do not reduce but rather increase transcription of SREBP-1 and SREBP-2 in mammalian liver cells which leads to an increased de novo synthesis of fatty acids and cholesterol in the liver(Reference Takahashi, Kushiro, Shinohara and Ide22, Reference Ringseis, König, Leuner, Schubert, Nass, Stangl and Eder23). Whether n-3 PUFA or CLA are able to activate PPARα also in laying hens or to influence the action of SREBP, and thereby influence the lipid metabolism of laying hens, has not yet been investigated. Therefore, we performed an experiment in which hens were fed diets with fish oil or a CLA oil in comparison to sunflower oil which was used as a reference. To study the PPARα-activating effect of these oils, we determined mRNA levels of various genes which were shown to be up-regulated by PPARα activation in rat models. These included carnitine palmitoyltransferase (CPT)-1A, acyl-CoA oxidase (ACO) and the peroxisomal bifunctional enzyme (enoyl-CoA hydratase/l-3-hydroxyacyl-CoA dehydrogenase), all genes involved in mitochondrial or peroxisomal β-oxidation(Reference Mandard, Müller and Kersten5). To assess whether fish oil or CLA influence gene expression or proteolytic activation of SREBP, we determined mRNA levels and nuclear concentrations of SREBP-1c and SREBP-2. Moreover, we determined gene expression of insulin-induced genes (Insig), proteins that are involved in the proteolytic activation of SREBP(Reference Yang, Espenshade, Wright, Yabe, Gong, Aebersold, Goldstein and Brown24, Reference Yabe, Brown and Goldstein25) and the important SREBP target genes involved in fatty acid synthesis (acetyl-CoA carboxylase (ACC), fatty acid synthase (FAS)) and cholesterol uptake (LDL receptor) and synthesis (HMG-CoA reductase). To study whether potential effects on PPARα or SREBP lead to phenotypic alterations in the lipid metabolism, we also determined concentrations of TAG and cholesterol in liver, plasma and egg yolk.
Materials and methods
Animals and treatment
An experiment was conducted with thirty Lohmann White layers (64 weeks old) with an average body weight of 1732 (sd 136) g. The hens were allotted to three groups of ten each. They received a nutritionally adequate diet which was supplemented with (30 g/kg) sunflower oil (from a local supermarket; control group), salmon oil (Caesar & Loretz GmbH, Hilden, Germany) or CLA oil (BASF, Ludwigshafen, Germany) (Table 1). The CLA oil contained 60 g CLA isomers per 100 g CLA oil in TAG form with cis-9, trans-11 and trans-10, cis-12 as the main isomers. The fatty acid composition of total lipids of the three experimental diets is given in Table 1.
CLA, conjugated linoleic acids.
* < 0·1 g/100 g total fatty acids.
The basal diet contained (g/kg diet): 465 wheat, 130 extracted soya bean meal, 120 corn, 80 peas, 75 CaCO3, 70 extracted sunflower meal, 30 oil, 12·5 Ca2PO4, 10 vitamin and mineral premix, 5 fibre, 2 NaCl and 0·5 dl-methionine. This diet contained 11·4 MJ metabolisable energy/kg and 169 g crude protein/kg (as determined by the official German VDLUFA methods)(Reference Naumann and Bassler26). Concentrations of essential amino acids, minerals and vitamins were in accordance with recommendations of the German Nutrition Society(27) for laying hens. The amount of vitamin E supplied by the premix was 40 mg/kg (as all-rac-α-tocopheryl acetate).
The hens were kept one bird per cage in an environmentally controlled room at 18°C. The room was lit for 14 h daily. Feed (mash) and water (via nipple drinkers) were available ad libitum. The experiment was conducted over a 5-week period. All procedures followed established guidelines for the care and handling of animals and were approved by the veterinary council of Saxony-Anhalt. Body weight at the start and end of the experiment, total egg production and weekly feed consumption were monitored.
After the end of week 5, overnight-fasted hens were anaesthetised by a strike on the head and were then decapitated. Blood was collected in heparinised tubes; plasma was separated by centrifugation at 1500 g for 10 min at 4°C. Liver was excised, weighed and immediately snap frozen in liquid N2. Aliquots of liver for RNA isolation were stored at − 80°C; other samples were stored at − 20°C. To determine egg yolk weight and concentrations of yolk lipids and fatty acid composition, two eggs from each hen were sampled at the end of week 5. Eggs were cooked in water for 10 min. The two eggs from each hen were pooled for determination of TAG and cholesterol concentrations in egg yolk.
Analysis of TAG, cholesterol and fatty acids
Lipids from liver and cooked egg yolks were extracted with a mixture of n-hexane and isopropanol (3:2, v/v)(Reference Hara and Radin28). For determination of the concentrations of TAG and cholesterol in the liver and egg yolks, aliquots of the lipid extracts were dried and the lipids were dissolved using Triton X-100(Reference De Hoff, Davidson and Kritchevsky29). Concentrations of TAG and cholesterol in plasma, liver and egg yolks were determined using enzymatic reagent kits (VWR International, Darmstadt, Germany, catalogue no. 1.14830, 1.14856). The fatty acid composition of dietary oils was determined by gas chromatography of fatty acid methyl esters which were prepared by methylation with trimethylsulfonium hydroxide(Reference Brandsch, Ringseis and Eder30). Concentrations of CLA isomers in the CLA oil were analysed by Ag+-HPLC-DAD(Reference Sehat, Rickert and Mossoba31).
Total RNA was isolated from livers by TRIZOL reagent (Sigma-Aldrich, Steinheim, Germany) according to the manufacturer's protocol. cDNA synthesis was carried out as described(Reference König and Eder32). The mRNA level of genes was measured by realtime detection PCR using SYBR® Green I and the Rotor Gene 2000 system (Corbett Research, Mortlake, Australia). PCR was performed with 1·25 U Taq DNA polymerase (Promega, Mannheim, Germany), 500 μm desoxyribonucleoside triphosphates and 26·7 pmol of the specific primers (Operon Biotechnologies, Cologne, Germany; Table 2). For determination of mRNA level a threshold cycle (Ct) and amplification efficiency was obtained from each amplification curve using the software RotorGene 4.6 (Corbett Research, Australia). Calculation of the relative mRNA level was made using the amplification efficiencies and the Ct values(Reference Pfaffl33). The housekeeping gene β-actin was used for normalisation.
ACO, acyl-CoA oxidase; ACC, acetyl-CoA carboxylase; CPT, carnitine palmitoyltransferase; FAS, fatty acid synthase; HMG-CoA, 3-hydroxy-3-methylglutaryl-CoA; Insig, insulin-induced gene; SREBP, sterol regulatory element-binding protein.
Nuclear extracts of livers were prepared from fresh tissue samples(Reference Woo, Siow, Pierce, Choy, Minuk and Mymin34). The protein content of the samples was determined by the bicinchoninic acid assay. Bicinchoninic acid reagent was purchased from Interchim (Montfucon, France). Equal amounts of proteins were pooled from five hens per group and 90 μg protein per lane were separated on 10 % SDS-polyacrylamide gels and electrotransferred to a nitrocellulose membrane (Pall, Pensacola, FL, USA). Polyclonal anti-SREBP-1 antibody (Santa Cruz Biotechnology, Inc., Heidelberg, Germany) and polyclonal anti-SREBP-2 antibody (Abcam plc, Cambridge, UK) were used to detect nuclear SREBP-1 and nuclear SREBP-2, respectively, using enhanced chemiluminescence reagent (GE Healthcare, Munich, Germany) and a chemiluminescence imager (biostep GmbH, Jahnsdorf, Germany). Signals were analysed with the Phoretix TotalLab TL100 software. The anti-rabbit-IgG peroxidase conjugate antibody was purchased from Sigma-Aldrich (Steinheim, Germany).
Treatment effects were evaluated by one-factorial ANOVA. For significant F values (P < 0·05), means of the treatments (fish oil, CLA) were compared pairwise with the control group by Student's t test. Means were considered significantly different for P < 0·05. Values are given as means with their standard deviation.
Body and liver weights, feed intake and egg production
Initial body weights of the hens did not differ between the three groups (control, 1747 (sd 146) g; fish oil, 1723 (sd 175) g; CLA, 1719 (sd 95) g, n 10 for each group). Liver weights of hens fed diets containing either fish oil or CLA did not differ from those of control hens (control, 44·0 (sd 8·8) g; fish oil, 40·6 (sd 4·6) g; CLA, 48·6 (sd 5·9) g). However, liver weights of hens fed the diet containing CLA were higher than those of hens fed the diet containing fish oil (P < 0·05). Feed intake of the hens during the experimental period (control, 4059 (sd 372) g; fish oil, 4121 (sd 273) g; CLA, 4041 (sd 348) g) and final body weights after 5 weeks (control, 1735 (sd 158) g; fish oil, 1749 (sd 136) g; CLA, 1701 (sd 143) g) did not differ between the three groups. Egg production rate was not different between the three groups of hens (control, 86 (sd 11) %; fish oil, 86 (sd 10) %; CLA, 87 (sd 9) %). However, eggs of hens fed the diet containing CLA had heavier yolks than eggs of control hens or hens fed the diet containing fish oil (P < 0·05; control, 18·2 (sd 1·1) g; fish oil, 17·7 (sd 0·8) g; CLA, 20·9 (sd 1·5) g).
Expression of PPARα and its target genes in the liver
Gene expression of PPARα was detected in the liver of the hens, but PPARα mRNA level did not differ between the three groups of hens (Fig. 1). Hens fed the diet containing fish oil had significantly higher mRNA levels of ACO (+45 %) and bifunctional enzyme (+50 %) in the liver than control hens (P < 0·05, Fig. 1). mRNA level of CPT-1A was not different between hens fed the diet containing fish oil and control hens (Fig. 1). In hens fed the diet containing CLA none of the PPARα target genes was significantly up-regulated in the liver relative to control hens (Fig. 1). There was however a tendency towards increased mRNA levels of ACO (+29 %, P = 0·10) and bifunctional enzyme (+30 %, P = 0·06) in hens fed the diet containing CLA compared to control hens. The mRNA level of CPT-1A in the liver was not different between hens fed the diet containing CLA and control hens (Fig. 1).
mRNA levels of Insig, SREBP, concentration of nuclear SREBP and mRNA levels of SREBP target genes in the liver
Hens fed the diet containing fish oil showed a moderately increased level of SREBP-1 mRNA compared to control hens (+31 %, P < 0·05; Fig. 2) but concentration of nuclear SREBP-1 protein in the liver was slightly decreased ( − 12 %, Fig. 3 (A)). Hepatic mRNA levels of the SREBP-1 target genes ACC and FAS were unchanged in hens fed the diet containing fish oil compared to control hens (Fig. 2). Hepatic mRNA level of SREBP-2 did not differ between hens fed the diet containing fish oil and those fed the control diet (Fig. 2). The concentration of the nuclear SREBP-2 protein, however, was lower in hens fed the diet containing fish oil than in control hens ( − 42 %, Fig. 3 (B)). Hepatic mRNA levels of the SREBP-2 target genes Insig-1 ( − 36 %, P < 0·05) and HMG-CoA reductase ( − 24 %, P < 0·05) were moderately reduced in hens fed the diet containing fish oil compared to hens fed the control diet; mRNA level of the SREBP-2 target gene LDL receptor, however, did not differ between these two groups of hens (Fig. 2). Hepatic mRNA level of Insig-2 did also not differ between hens treated with fish oil and control hens (Fig. 2).
In hens fed the diet containing CLA, mRNA level of SREBP-1 in the liver was markedly increased in comparison to control hens (+70 %, P < 0·05; Fig. 2), while concentration of nuclear SREBP-1 in the liver of hens fed CLA containing diet was only slightly increased (+23 %; Fig. 3 (A)). However, mRNA levels of the SREBP-1 target genes ACC and FAS did not differ between these two groups of hens (Fig. 2). Hens fed the diet containing CLA did not differ in hepatic mRNA level of SREBP-2 (Fig. 2) and had slightly reduced concentration of the nuclear SREBP-2 protein ( − 17 %, Fig. 3 (B)) and a higher mRNA level of the SREBP-2 target gene LDL receptor (+30 %, P < 0·05; Fig. 2). mRNA levels of SREBP-2 target genes Insig-1 and HMG-CoA reductase as well as that of Insig-2 did not differ between these two groups of hens (Fig. 2).
Concentrations of TAG and cholesterol in liver, plasma and egg yolk
Hens fed the diet containing fish oil had lower concentrations of TAG and cholesterol in plasma than control hens (P < 0·05; Table 3). Concentration of TAG and cholesterol in liver did not differ in hens fed fish oil compared to control hens (Table 3). Hens fed the diet containing fish oil moreover tended to have lower concentrations and absolute amounts of TAG in egg yolk than hens fed the control diet (P = 0·07 and P = 0·06, respectively; Table 3). Concentrations and absolute amounts of cholesterol in egg yolk did not differ between these two groups of hens.
Mean values were significantly different from control group: *P < 0·05, **P < 0·10.
Hens fed the diet containing CLA had higher concentrations of TAG and cholesterol in liver (P < 0·05) and tended to have higher concentration of cholesterol in plasma than control hens (P = 0·06; Table 3). Concentration of TAG in plasma was not altered in hens fed the diet containing CLA compared to control hens. Moreover, both concentration and absolute amount of TAG in egg yolks tended to be lower in hens fed the diet containing fish oil compared to control hens (P = 0·07 and P = 0·06, respectively); concentration of cholesterol in egg yolk did not differ between those two groups of hens (Table 3). Hens fed the diet containing CLA tended to have higher concentrations of TAG and cholesterol in egg yolk than control hens (P = 0·06 and P = 0·10, respectively). Absolute amounts of TAG and cholesterol in egg yolk were higher in hens fed the diet containing CLA than in control hens (P < 0·05, Table 3).
To our knowledge, this is the first study which investigated the effects of fish oil and CLA in laying hens on the expression of genes controlled by PPARα and SREBP, transcription factors that control whole-body lipid homeostasis. Hens received diets containing sunflower oil, fish oil or CLA for 5 weeks. To study whether the dietary oils activated PPARα, we determined mRNA levels of classical PPARα target genes, namely ACO, CPT-1A and the peroxisomal bifunctional enzyme. Feeding fish oil to laying hens caused a significant up-regulation of ACO and bifunctional enzyme compared to feeding sunflower oil which was used as a control fat. Although we did not perform a direct PPARα activation assay, this finding suggests that n-3 PUFA present in fish oil were able to bind to and activate PPARα. It should be noted that, since the up-regulation of these genes was relatively weak, being 45 and 50 % for ACO and bifunctional enzyme, respectively, n-3 PUFA probably caused only a slight activation of PPARα in the liver of hens. This is in contrast to rodents in which administration of fish oil causes a much stronger up-regulation of PPARα target genes in the liver. For example, in the study of Nakatani et al. (Reference Nakatani, Katsumata, Miura, Kamei and Ezaki19), feeding a diet containing fish oil for one week increased mRNA level of ACO in the liver of mice four-fold.
The data of the present study, moreover, show that the CLA mixture administered to the hens, with cis-9, trans-11 and trans-10, cis-12 CLA as the main isomers, did not significantly up-regulate gene expression of PPARα target genes although there were some numerical increases in mRNA levels of ACO and bifunctional enzyme. It is therefore concluded that dietary CLA does not activate PPARα, or at most to a very small extent, in the liver of laying hens under in vivo conditions. This is in contrast to studies in rat liver cells which demonstrated that both of these isomers are able to bind to and activate PPARα(Reference Moya-Camarena, Vanden Heuvel and Belury15, Reference Moya-Camarena, Vanden Heuvel, Blanchard, Leesnitzer and Belury16). The disagreement between the present study and the rat liver cell study may be that rat liver cells have a very high gene expression of PPARα in the liver, which is several-fold greater than in other species(Reference Tugwood, Holden, James, Prince and Roberts35, Reference Holden and Tugwood36).
In rodents, PPARα activation leads to a reduction of liver and plasma TAG concentration(Reference König, Koch, Spielmann, Hilgenfeld, Stangl and Eder37). The increased expression of genes involved in fatty acid β-oxidation is one of the mechanisms underlying the hypotriacylglycerolaemic effect upon PPARα activation(Reference Mandard, Müller and Kersten5). In laying hens, treatment with clofibrate also caused a strong reduction of liver and plasma TAG concentration which suggests that PPARα activation may affect fatty acid and TAG metabolism in a similar way as in rodents(Reference König, Kluge, Haase, Brandsch, Stangl and Eder3). Thus, it is likely that PPARα activation contributed to the reduction of plasma TAG concentrations observed in hens fed the diet containing fish oil. It should be noted, however, that fish oil has several other effects on lipid metabolism such as inhibiting assembly of VLDL or stimulating the expression of PPAR cofactors(Reference Botham, Zheng, Napolitano, Avella, Cavallari, Rivabene and Bravo38, Reference Bassaganya-Riera and Hontecillas39). These effects could also contribute to alterations in plasma lipid concentrations observed in hens fed fish oil.
In mammals, n-3 PUFA cause a down-regulation of SREBP-1 at the transcriptional level which in turn inhibits transcription of genes involved in hepatic de novo fatty acid synthesis such as FAS or ACC(Reference Clarke, Armstrong and Jump40, Reference Geelen, Schoots, Bijleveld and Beynen41). The present study shows that n-3 PUFA do not exert such an effect in the liver of laying hens. It is shown that mRNA level of SREBP-1 is even increased by fish oil in the liver of hens. It has been shown that gene expression and activation of SREBP-1c in the liver are triggered by activation of liver X-receptor, a receptor which is activated by oxysterols(Reference Geyeregger, Zeyda and Stulnig42). Studies in rats have shown that induction of oxidative stress, i.e. by administration of fish oil, enhances the formation of oxysterols in the liver(Reference Ringseis and Eder43). Although the diet used in this study contained an adequate amount of vitamin E, it is possible that administration of fish oil enhanced the formation of oxysterols which in turn caused a liver X-receptor-induced activation of SREBP-1 in the liver of the hens. Nevertheless, an elevated mRNA level of SREBP-1 in the liver of hens fed a diet containing fish oil did not result in increased concentration of nuclear SREBP-1 which was instead marginally decreased. Consistent with that, mRNA levels of SREBP-1 target genes ACC and FAS were also unchanged. These data suggest that fish oil did not inhibit lipogenesis in the liver of hens.
In addition to reduced plasma TAG concentration we found that the TAG concentration of egg yolks also tended to be decreased by fish oil feeding. TAG from the liver are secreted via VLDL into the blood, and they are taken up by oocyte receptors. Thus we suggest that, due to enhanced fatty acid oxidation, VLDL secretion from the liver and uptake by oocyte receptors was reduced leading to reduced TAG concentration in egg yolk.
For SREBP-2, no alteration in mRNA level was observed upon fish-oil feeding but the concentration of nuclear SREBP-2 in the liver was decreased. In agreement with that, mRNA level of the SREBP-2 target gene HMG-CoA reductase and that of Insig-1, whose transcription is also regulated by SREBP, was reduced in hens treated with fish oil. In agreement with reduced expression of HMG-CoA reductase, the key enzyme of cholesterol synthesis, cholesterol concentration in the plasma of hens fed fish oil was reduced. A similar reduction of plasma cholesterol concentration was also observed in another study by feeding fish oil to young chicks(Reference Castillo, Amalik, Linares and Garcia-Peregrin44). In rats, it has been shown that activation of PPARα causes an inhibition of the processing of the immature SREBP-2, leading to a reduced SREBP-2 activity in the liver which in turn causes a reduction of hepatic cholesterol synthesis and plasma cholesterol concentration(Reference König, Koch, Spielmann, Hilgenfeld, Stangl and Eder37, Reference Patel, Knight, Wiggins, Humphreys and Gibbons45). Thus, the possibility exists that processing of SREBP-2 in hens fed fish oil was also inhibited by activation of PPARα.
The present study reveals that feeding CLA up-regulates LDL receptor in the liver of laying hens. This effect that could contribute to the increased cholesterol concentration in the liver observed in hens fed the diet containing CLA. A similar up-regulation of LDL receptor expression by trans-10, cis-12 CLA was also observed in HepG2 cells and was dependent on the up-regulation of SREBP-2(Reference Ringseis, König, Leuner, Schubert, Nass, Stangl and Eder23). In hens fed CLA, level of SREBP-2 mRNA was unchanged and that of nuclear SREBP-2 was marginally decreased indicating that up-regulation of LDL receptor was probably not mediated by SREBP-2 in the liver of hens fed CLA. Consistent with almost unchanged concentration of nuclear SREBP-2, mRNA levels of SREBP-2 target genes HMG-CoA reductase and Insig-1 were also not altered. In mammals, LDL receptor expression is also regulated via specificity protein-1 and activator protein-1 binding sites and by liver X-receptor α(Reference Fu, Huang, Bandyopadhyay, Virella and Lopes-Virella46–Reference Ishimoto, Tachibana and Sumitomo48). Data on effects of dietary CLA on plasma and liver lipids in mammals are inconsistent revealing decreased, unchanged and increased cholesterol levels(Reference Kelley, Bartolini, Warren, Simon, Mackey and Erickson49–Reference Porsgaard, Xu and Mu52). Taken together, the reasons for the increased liver and plasma cholesterol concentrations in hens fed CLA cannot be completely explained by the data of this study.
The present study shows that CLA, like fish oil, caused an up-regulation of SREBP-1 mRNA expression in the liver of hens. Nevertheless, the concentration of mature SREBP-1 protein in the nucleus was only marginally increased and the mRNA levels of both SREBP-1 target genes ACC and FAS remained unchanged. Thus, elevated mRNA levels of SREBP-1 did not result in increased nuclear concentration of the protein. This may be due to the increased expression of Insig-2. Dependent on cholesterol concentration, Insig retard the SREBP precursor proteins in the endoplasmic reticulum thereby preventing their proteolytic activation(Reference Yang, Espenshade, Wright, Yabe, Gong, Aebersold, Goldstein and Brown24, Reference Yabe, Brown and Goldstein25). In mice, Insig-2a, the liver-specific transcript of Insig-2, is regulated by insulin and thus plays a special role in regulation of SREBP-1c maturation allowing fatty acid synthesis even at elevated cholesterol concentrations(Reference Yabe, Komuro, Liang, Goldstein and Brown53). The finding that CLA increases hepatic TAG concentration agrees with recent studies in laying hens(Reference Cherian and Goeger54, Reference Cherian, Holsonbake, Goeger and Bildfell55). In these studies, even a low concentration of a mixture of CLA (5 g/kg diet) caused an increase of the number of lipid vacuoles in the liver of laying hens and an increased concentration of liver total lipids. Nevertheless, the reason for the increased hepatic TAG concentration in hens fed CLA remains unclear. Studies in mice demonstrated that trans-10, cis-12 CLA induces hyperinsulinaemia and fatty liver which may be caused by reduced leptin and adiponectin plasma concentrations and an up-regulation of PPARγ, adipocyte lipid-binding protein, fatty acid transporter mRNA and FAS genes(Reference Clement, Poirier, Niot, Bocher, Guerre-Millo, Krief, Staels and Besnard56, Reference Poirier, Rouault, Clement, Niot, Monnot, Guerre-Millo and Besnard57). It could be that trans-10, cis-12 CLA which made up approximately 50 % of total dietary CLA isomers exerted similar effects in the liver of hens.
Interestingly, we found also an increase in egg yolk weights and in yolk TAG and cholesterol concentrations in hens fed the diet containing CLA. Effects of CLA on egg yolk weight reported in literature are variable. In some studies dietary CLA lowered egg or yolk weights(Reference Raes, Huyghebaert, De Smet, Nollet, Arnouts and Demeyer58–Reference Szymczyk and Pisulewski60) whereas in others CLA did not alter egg or yolk weights(Reference Schäfer, Männer, Sagredos, Eder and Simon61, Reference Ahn, Sell, Jo, Chamruspollert and Jeffrey62). Results about the effects of CLA on egg yolk lipids are also variable(Reference Cherian, Holsonbake, Goeger and Bildfell55, Reference Schäfer, Männer, Sagredos, Eder and Simon61, Reference Ahn, Sell, Jo, Chamruspollert and Jeffrey62). In our study, both TAG and cholesterol concentrations in egg yolks were increased by dietary CLA. It is possible that the increased yolk weights and increased concentrations of TAG and cholesterol in egg yolk are linked to increased TAG concentrations in the liver and increased cholesterol concentrations in liver and plasma. It is possible that VLDL secretion from the liver and uptake into oocytes were increased in hens fed CLA which resulted in increased cholesterol and TAG transfer into egg yolk.
In conclusion, this study shows for the first time that fish oil causes a moderate up-regulation of PPARα target genes in the liver of laying hens but had less effect on the concentration of nuclear SREBP-1 in the liver and its target genes involved in TAG synthesis. Possibly, up-regulation of PPARα target genes contributed to reduced plasma TAG concentration observed in hens fed fish oil. It is also shown that the concentration of nuclear SREBP-2 and expression of HMG-CoA reductase is decreased by fish oil. This suggests that fish oil lowers hepatic cholesterol synthesis, an effect that might contribute to decreased plasma cholesterol concentration in these hens. It is furthermore shown that a mixture of CLA with cis-9, trans-11 and trans-10, cis-12 CLA does not significantly up-regulate PPARα target genes and has less effect on nuclear concentrations of SREBP-1 and SREBP-2 and their target genes. This suggests that CLA do not influence synthesis of TAG and cholesterol in the liver. Nevertheless, hens fed CLA had higher concentrations of TAG in liver and of cholesterol in liver and plasma and a higher concentration of TAG and cholesterol in egg yolks than control hens. The mechanisms underlying these effects have to be elucidated in further studies.
B. K. supervised gene expression analyses, participated in the interpretation of the results and prepared the manuscript; J. S. determined nuclear concentrations of SREBP; K. H. and H. K. planned and carried out the animal experiment; C. B. supervised the lipid analyses; G. S. participated in the design of the study, and interpretation of the results; K. E. conceived the study and its design, coordinated work, participated in the interpretation of the results, and helped to draft the manuscript. There was no specific funding for this study and there is not any potential conflict of interest.