Genome-wide expression profiling in muscle and subcutaneous fat of lambs in response to the intake of concentrate supplemented with vitamin E

The aim of the present study was to assess the effects of adding VE to the fattening concentrate, fed between weaning and slaughter, on the transcriptional changes in the LT and SF. The experimental period ended when the lambs reached the target slaughter LW (22–24 kg), according to the specifications of Ternasco de Aragón Protected Geographical Indication (Regulation (EC) No. 1107/96). The light lamb production is based on two periods: suckling and fattening. The suckling period is usually limited by the time, usually lasting approximately 45 d (in the present study, it was 48 d), while the fattening period is limited by the weight (22–24 kg LW). Thus, lambs with high ADG during lactation period spent fewer days in the fattening period than those with low ADG. To be able to evaluate the potential effects of the inclusion of VE on the transcriptome of the L. Thoracis muscle and subcutaneous fat, we planned to feed at least 30 days of VE concentrate. According to Ripoll et al. [11], supplementation with VE for 30 days was found to increase noticeably concentration of the α-tocopherol content in muscle. In the present study, the ADG during the experimental period was not different, however during the suckling period it was lower in VE treatment (p < 0.01; Table 1). This implies that the lambs of the VE treatment during the suckling period presented lower growth compared to the CON group and needed more days to reach the target weight. Moreover, there was a strong correlation between ADG from birth to slaughter and SA (r ² =−0.9). Therefore, to overcome this unbalance between treatments ADG values were included as covariates in the statistical model used to validate gene expression differences by qPCR, thus avoiding estimation biases. We analyzed LT muscle and SF because meat cuts for human consumption include both intramuscular and subcutaneous fat [20], constituting the total amount of meat fat purchased at retail. The results related to the content of α-tocopherol in the muscle and TBARS confirm what was planned in the methodology. Vitamin E has been successfully used to increase the muscle α-tocopherol and to reduce lipid oxidation in beef and chicken meat [21]. In this study, as expected, there were significant differences between both diets in the α-tocopherol content, lipid oxidation and MMb formation in the LT muscle, as previously found by Ripoll et al. [11]. Moreover, TBARS values were lower in VE lambs which showed that the lipid oxidation process was slower in animals fed this type of diet.

In the present study, we used microarray technology to study changes in gene expression profiles in LT muscle and SF in response to vitamin E supplementation in lambs. Our results, showed that vitamin E supplementation caused different responses in gene expression in LT muscle and SF, suggesting a specific response of tissue to vitamin E supplementation. In our study, we did not measure the concentration of VE in the SF, however the gene expression results might be related to the greater α-tocopherol accumulation in adipose tissues than in skeletal muscles [22, 23].

It has been reported that α-tocopherol can influence a number of biological functions by regulating cell signaling at both the mRNA and miRNA levels [7]. Indeed, our results showed that vitamin E supplementation affected the expression of 29 genes in the LT muscle. The results of functional analysis showed that genes related to the intracellular signaling cascade (CISH, IGF1R, DEF8, and AKAP7) and metabolic processes (ZNF79, MAFB, MYLK2, ACACB, ACAT1, CISH, IGF1R, PGLS, DUSP26, AKR7A2, FBXL4, AKAP7, and RSC1A1) were down-regulated. The enrichment scores were less than 1.3, likely because the number of significant genes in this contrast was low. The most down-regulated gene by vitamin E in LT was CISH (Cytokine Inducible SH2 containing protein). Chen et al. [24] reported that CISH activates protein kinase C (PKC) activity by G-protein coupled receptor protein, which is important for the activation of both the activator protein 1 (AP-1) and nuclear factor- κB (NF-κB) pathways. In a previous study Boscoboinik et al. [25], demonstrated that PKC is inhibited by α-tocopherol. NF-κB proteins are a family of transcription factors and are of central importance in inflammation, immunity and apoptosis [26]. Evidence suggests a role for reactive oxidative intermediates (ROIs) as a common and critical intermediate for various NF-κB-activating signals, based on inhibition of NF-κB activation by a variety of antioxidants [27, 28]. There is a bidirectional relationship between cytokines and oxidative stress. Exposure of myotubes to reactive oxygen species (ROS)-producing agents resulted in an increase in interleukin 6 (IL-6) release through the activation of the redox-sensitive transcription factor, NF-κB [28]. In this sense, IL-6 induces marked increases in expression of CISH, SOCS-1, SOCS-2, and SOCS-3 in tissues, which in turn result in inhibition of the signaling of wide range of cytokines [29]. Thus, CISH expression could be related to oxidative status in the muscle. In this sense, low levels of ROS and ROIs in muscle caused by α-tocopherol could be associated with low expression of CISH. In this sense, lipid oxidation in the LT muscle and metmyoglobin formation were lower when lambs were supplemented with VE. In the same manner, RSC1A, which inhibits a dynamin and PKC-dependent exocytotic pathway of SLC5A1 gene, was down-regulated in the VE group. Interestingly enough CISH and MYLK, also down-regulated in the VE treatment, are involved in inflammation mediated by chemokine and cytokines signaling pathway, which could suggest another possible role for VE that of decreased inflammation.

In our study, alterations of the Ras homolog gene family, member A (RhoA) and actin cytoskeleton signaling were identified (IGF1R and MYLK2 genes). Both IGF1R and MYLK2 were down-regulated in VE treatment. Insulin-like growth hormone 1 (IGF-1) is a protein structurally similar to insulin, and it regulates tissue growth and development in several vertebrates [30]. As a main receptor of IGFs, IGF1R mediates the transduction of metabolic signals of cell proliferation, bone growth, and protein synthesis in the GH/IGF pathway [31]. In our study, IGF1R was down-regulated in VE treatment, and because of IGF1R polymorphisms have been associated to growth traits [32–36], we thought that this effect could be due to the higher ADG in CON animals. However, ADG values were included as a covariate in the statistical model used to validate expression differences by qPCR, thus avoiding estimation biases related to ADG. Therefore these results suggest that supplementation with VE causes down-regulation of IGF1R in LT muscle. Our findings are in agreement with Araujo et al. [37], who showed that VE supplementation reduces IGF1R expression by 17% in hyperthyroid of Wistar rats. In addition, Chuang et al. [38] also found that VE alone significantly and dose-dependently reduced the cell surface expression of IGFIR in HL-60 cells. Moreover, Holzenberger et al. [39] reported that Igf1r ^+/−mice display greater resistance to oxidative stress.

In addition, vitamin E down-regulated two genes related to lipid metabolisms (ACAT1 and ACACB) in the LT muscle. ACAT1 (acetyl-coenzyme A acetyltransferase 1) encodes a mitochondrial localized enzyme that catalyzes the reversible formation of acetoacetyl- CoA from two molecules of acetyl-CoA. ACAT1 is responsible for cholesterol homeostasis and maintain appropriate cholesterol availability in cell membranes, whereas ACACB is the key regulator of the fatty acid oxidation pathways [40]. It controls fatty acid oxidation by means of the ability of malonyl-CoA to inhibit carnitine-palmitoyl-CoA transferase I (CPT1B). As in our work, Shige et al. [41] showed that vitamin E reduced the uptake of modified low-density lipoprotein (LDL) and suppressed ACAT activity, resulting in less cholesterol esterification in macrophages. Interestingly enough ABCC4, an ATP-binding cassette (ABC) transporter, AKR7A2 which is involved in the detoxification of aldehydes and ketones, and finally RSC1A1 which transport carbohydrate across the plasma membrane, were all down- regulated with VE treatment. In hepatocytes, ABCC4 was shown to be induced by oxidative stress through binding of the oxidative sensor nuclear factor E2-related factor 2 (Nrf2) to antioxidant-responsive element sequences in the promoter of ABCC4 [42]. Our results showed for the first time that vitamin E down-regulates ABCC4 expression. Because of significant differences in ADG between treatments, we validated the expression results of the CISH, ABCC4, ACAT1 and IGFR1 genes by qPCR, including in the statistical model ADG as a covariate. ADG was not significant, with a similar FC between the microarray and qPCR results (r ² = 0.99; P = 0.008). Therefore, treatment was the main effect over gene expression.

In addition, we found that the transcription factors FHL3, ZNF777 and MAFB were down-regulated. Considering all together, our results showed that supplementation with VE increased the content of α-tocopherol in the LT muscle and decreased metmyoglobin formation and lipid oxidation. We speculate that α-tocopherol in the LT muscle causes a decrease in the catalytic activity of enzymes involved in cellular transport of fatty acids and carbohydrates, and fatty acid oxidation in the mitochondria. Decreased IGF1R expression might be related to the lower lipid oxidation levels in VE animals. We also found that IGFR1, ABCC4 ACAT1, CISH, ACACB, MYLK2, ZNF777 and MAFB were positively correlated with each other and with the diet, which suggest co-expression processes of these genes. We speculate that transcription factors ZNF777, MAFB and FHL3 could be important players in mediating the effects of VE in regulating the expression of these genes.

To establish whether IGFR1, ABCC4 and ACAT1, CISH are markers of meat oxidation or indirect markers of meat quality, further studies with greater numbers of animals are necessary.

A most dramatic effect of VE was observed on SF gene expression. ALG11, SRPBR, DDX47, SEC23ID and TTC37 were among the most important genes in discriminate fed treatments. These genes were up-regulated and positively correlated with each other and with the diet, which might suggest co-expression processes. Four of the up-regulated genes in the VE group were related to heat shock proteins (HSPs) or chaperonin activity (TTC37, DNAJC16, HSB8, AHSA1). Some HSPs are characterized by various specific functions such as anti-apoptotic or anti-inflammatory effects [43]. In our study, these genes were up-regulated and positively correlated, suggesting putative increased stress protection in VE lambs.

Surprisingly VE treatment showed general up-regulation of almost all significant genes, compared to CON treatment. Lipid biosynthetic processes were among the most enriched functional clusters with major biological significance and importance (PGAP3, EBP, CRLS1, MVD, CYP51A1, GNE, HMGCS1, DPAGT1, LSS, SIGMAR1, LPCAT3, FDFT1, DOLK, PIGM, SQLE, DHCR7, AGPAT9, LTA4H, PCYT2, and HSD17B7). Moreover, genes implicated in sterol, steroid and cholesterol biosynthesis processes (SREBF1, EBP, LDLR, MVD, CYP51A1, SQLE, DHCR7, INSIG1, HMGCS1, LSS, and FDFT1) were up-regulated in the SF of VE animals, compared to the CON group. It has been previously reported that tocopherols inhibited de novo cholesterol synthesis within enterocytes [44], and cause repression of genes (DHCR7 and HMGCS1) involved in the de novo synthesis of cholesterol in testes and adrenal glands [45]. The differences reported in previous studies and ours might be due to the different tissues analyzed. This supports even more our idea that there is a tissue specific response in response to VE supplementation. On the other hand, Wang et al. [46] found oxysterol-specific repressive effects in the CYP51A1 and FDFT1 genes, mediated via direct binding of the ligands to liver X receptor (LXR). Because of α–tocopherol and other antioxidants can inhibit the oxidation of cholesterol [47], the VE group could have a lower quantity of oxysterols and then have inhibited the repression of cholesterol biosynthesis via LXR and oxysterols and up-regulated the sterol, steroid and cholesterol biosynthesis processes. In this sense, several genes implicated in “the stress response” (SOD3, IER3, HMGB2, UACA, LUC7L3) were down-regulated in the VE group, compared to the CON group (Table 3 and Supplementary Table S1). Among them, SOD3 showed higher values of FC. This gene encodes a member of the superoxide dismutase (SOD) protein family that protects the extracellular space from the toxic effects of reactive oxygen intermediates by converting superoxide radicals into hydrogen peroxide and oxygen. This finding contradicts the results of previous studies, which suggested that grass-based treatments elevated the activity of antioxidants, such as glutathione and superoxide dismutase (SOD), compared to grain-feeding [1]. However, Kumar et al. [48] in the myocardium muscle and Strobel et al. [49] in the skeletal muscle of exercise-trained and sedentary rats, found that antioxidant supplements reduced the endogenous antioxidants SOD2 gene and protein and the glutathione peroxidase (GPx) gene and enzyme activity.

Another down-regulated gene in VE treatment was IER3. IER3-deficient NCM460 cells exhibited reduced reactive oxygen species levels, indicating increased antioxidative protection [50]. In our study, IER3 was down regulated in the VE group suggesting an increase in antioxidative protection. The role of HMGB1 in recognizing aberrant or damaged DNA has been shown in multiple in vitro experiments. A recent study directly showed the accumulation of HMGB1 at sites of oxidative DNA damage in live cells, thus defining HMGB1 as a component of an early DNA damage response [51]. A similar function has been attributed to HMGB2 [52]. These authors hypothesized that HMGB1/2 proteins act as a sensor of DNA modification, and their interaction with chemically altered DNA changes the chromatin structure, thus inducing DNA damage responses.

Finally, a cluster related to tRNA metabolic processes was also significant (TRNT1, METTL1, ADAT2, NSUN2, IARS, GARS, EPRS and MARS). Aminoacyl-tRNA synthetases perform an essential function in protein synthesis by catalyzing the esterification of an amino acid to its cognate tRNA (IARS, GARS, EPRS and MARS). Considering all together, we speculate that in the SF, vitamin E exerts anti-inflammatory effect and stress protection by increasing heat shock protein expression. Vitamin E reduces stress response probably as results of reduced reactive oxygen species levels. In addition animals supplemented with VE might have inhibited cholesterol oxidation in SF and enhanced sterol, steroid and cholesterol biosynthesis processes. And finally, we speculate that regulation of these gene expressions it could be mediated through tRNAs IARS, GARS, EPRS and MARS.

Regarding LT, we included ADG values as a covariate in the statistical model used to validate the expression differences by qPCR to avoid estimation biases. In this case, ADG was not significant, and a similar FC between microarray and qPCR results was found (r ² = 0.99; p < 0.0001). Thus, the treatment was also the main effect over gene expression of SF.

Although we found differences in mRNA activity, it did not necessarily cause differences in metabolic processes. An increase in gene expression is not necessarily correlated with an increase in protein concentrations or enzyme activities. There are many processes between transcription and translation, including post-transcriptional, translational and protein degradation regulation, in controlling steady-state protein abundances [53]. Moreover, DNA microarray has been the technology of choice for transcriptome analysis in recent years. Nonetheless, array technology has several limitations which include: using microarray technology limits the researcher to detecting transcripts that correspond to existing genomic sequencing information; and background hybridization limits the accuracy of expression measurements, particularly for transcripts present in low abundance.