A comparative transcriptomic study of vitamin E and an algae-based antioxidant as antioxidative agents: Investigation of replacing vitamin E with the algae-based antioxidant in broiler diets

A comparative transcriptomic study of vitamin E and an algae-based antioxidant as antioxidative agents: Investigation of replacing vitamin E with the algae-based antioxidant in broiler diets R. Xiao,*†1 R. F. Power,*† D. Mallonee,*† C. Crowdus,* K. M. Brennan,*† T. Ao,*† J. L. Pierce,*† and K. A. Dawson*† *Center for Animal Nutrigenomics and Applied Animal Nutrition, Alltech, Nicholasville, KY 40356; and †Alltech–University of Kentucky Nutrition Research Alliance, Lexington 40506 that expression of 542 genes of the breast muscle were altered (P < 0.05, fold change >1.2) by dietary treatments, of which a significant part were commonly regulated by EcoE and VE (especially the control diet + 50 IU of VE/kg). In addition to the process of cellular oxidation, gene ontology analysis indicated the involvement of EcoE and VE on cell morphology, skeletal and muscular system development and function, immune response, and multiple metabolic processes, including lipid, carbohydrate, and drug metabolism. Results of this experiment indicate that the biological roles of high VE, including its activity as an antioxidant, can be greatly mimicked at the transcriptional level by EcoE, and they suggest a relationship of functional redundancy between VE and EcoE in the broiler diets.

Key words: antioxidants, EconomasE, gene expression, vitamin E 2011 Poultry Science 90:136–146 doi:10.3382/ps.2010-01018

INTRODUCTION

lated to high VE have been credited to the improved antioxidant status of animals, but the exact mechanisms involved are far from clear. Increased prices of VE, which have resulted from the rising cost of raw materials and energy in recent years and the potential environmental impact related to the manufacturing of VE and the growing demand for this functional antioxidant in the feed industry (Feedinfo News Service, 2008, 2009), ensure the necessity of research on cheaper but functionally equivalent products to replace VE. Our previous study indicated that inclusion of an algae-based antioxidant containing Se yeast (EconomasE, Alltech Inc., Nicholasville, KY; EcoE), significantly reduced the amount of VE required in the broiler diet for the best growth performance and meat quality (Pierce et al., 2009). The ability of EcoE to replace VE in broiler diets may be explained by its high content of antioxidants such as Se, an essential trace element that plays an important role in the antioxidative system (Combs, 1981; Choct et al., 2004), or through

Vitamin E (VE, α-tocopherol), which is one of nature’s most effective lipid-soluble antioxidants, protects cell membranes from oxidative damage (Brigelius-Flohé and Traber, 1999). The nutritional requirements for VE in many animal species and the risks related to its deficiency are well established (Wilson et al., 1984; Bendich et al., 1986; Hakkarainen et al., 1986; Richter et al., 1987). In the feed industry, however, much higher levels of VE (5 to10 times the NRC-recommended amount; Leeson et al, 2007) are typically added in animal diets to achieve optimized growth performance, reproduction, and meat quality (Kennedy et al., 1992; Coetzee and Hoffman, 2001). Many of the benefits re©2011 Poultry Science Association Inc. Received July 16, 2010. Accepted October 3, 2010. 1 Corresponding author: [email protected]

136

Downloaded from http://ps.oxfordjournals.org/ by guest on January 12, 2015

ABSTRACT Previous study indicated that inclusion of an algae-based antioxidant as an antioxidative agent [EconomasE, Alltech, Nicholasville, KY; EcoE] significantly reduced the amount of vitamin E (VE) required in broiler diets without compromising performance and meat quality. To assess the mechanisms related to the VE-saving activity of EcoE, as well as other potential functions related to EcoE and VE supplementation, we analyzed gene expression profiles of breast muscle from broilers fed a control diet, the control diet + 50 IU of VE/kg, the control diet + 100 IU of VE/kg, or the control diet + 200 g of EcoE/ton. Evaluation of the serum antioxidant capacity indicated that dietary supplementation of either a high level of VE (50 or 100 IU of VE/ kg) or EcoE significantly improved bird antioxidant status. Analysis of gene expression profiles indicated

137

TRANSCRIPTOMIC STUDY OF VITAMIN E AND AN ALGAE-BASED ANTIOXIDANT

MATERIALS AND METHODS Bird Handling and Diets A total of 640 one-day-old Cobb 500 broiler chicks were randomly divided into 4 treatment groups and assigned to 1 of the following diets: a corn- and soy-based diet (control; Table 1) or the control diet supplemented with 50 IU of VE/kg (E50 diet), 100 IU of VE/kg (E100 diet), or 200 g of EcoE/ton of diet (EcoE diet). The control, E50, and E100 diets were further fortified with 0.3 mg/kg of Se from selenite, which brought all the diets in this experiment to the same Se level (Table 1). After 6 wk of feeding, 7 chicks from each diet group were killed and blood samples were taken, via cardiac puncture, for serum. The right side of the breast muscle was used for the VE content analysis. Samples of the left side breast muscle were snap-frozen in liquid N and stored in a −80°C freezer for later microarray analysis. All procedures were approved by the University of Kentucky Institutional Animal Care and Use Committee.

Total Antioxidant Capacity Serum total antioxidant capacity was determined by using a Cayman antioxidant assay kit (Cayman Chemical Co., Ann Arbor, MI) and the protocol supplied (Miller et al., 1993). This assay relies on the ability of antioxidants in the sample to inhibit the oxidation of 2,2′-azino-di-(3-ethylbenzthiazoline sulfonate). The capacity of the antioxidants in the sample to prevent 2,2′-azino-di-(3-ethylbenzthiazoline sulfonate) oxidation is compared with that of Trolox, a water-soluble tocopherol analog, and is quantified as millimolar Trolox equivalents.

Table 1. Ingredients and composition of the basal Item Ingredient (%)   Corn   Soybean meal (48%)   Corn oil   Limestone   Dicalcium phosphate   Salt   Vitamin premix2 (no vitamin E)   Mineral mix2 (no Se)   dl-Methionine   Mycosorb3   Total Nutrient composition   ME (kcal/kg)   CP (%)   Ca (%)   Available P (%)   Lysine (%)   Methionine (%)   Methionine + cysteine (%)   Na (%)

Starter (d 1 to 21)

Grower (d 22 to 42)

54.35 36.5 4.80 1.33 1.76 0.45 0.25 0.25 0.21 0.10 100   3,160 22 1.00 0.45 1.24 0.56 0.91 0.20

60.93 31.50 3.65 1.23 1.54 0.45 0.25 0.25 0.10 0.10 100   3,150 20 0.90 0.40 1.11 0.42 0.75 0.20

1Analytical dietary vitamin E concentrations (IU/kg) of starter diets were 23.9, 81.9, 160.0, and 23.3 in the control diet, control diet + 50 IU of vitamin E/kg (E50), control diet + 100 IU of vitamin E/kg (E100), and control diet + 200 g/ton of an algae-based antioxidant (EconomasE, Alltech, Nicholasville, KY; EcoE), respectively. Vitamin E concentrations (IU/kg) in the grower diets were 15.5, 63.8, 104.0, and 14.9, respectively. Dietary Se concentrations (ppb) of grower feed were 272, 274, 255, and 307 in the control, E50, E100, and EcoE groups, respectively. 2Supplied per kilogram of diet: 11,025 IU of vitamin A (retinyl acetate), 0.0882 mg of vitamin D3 (cholecalciferol), 0.91 mg of vitamin K3 (2-methyl-1,4-naphthoquinone), 2 mg of thiamine, 8 mg of riboflavin, 55 mg of niacin, 18 mg of Ca pantothenate, 5 mg of vitamin B6 (pyridoxine), 0.221 mg of biotin, 1 mg of folic acid, 478 mg of choline, and 28 µg of vitamin B12 (cyanocobalamin); 80 mg of iron as FeSO4·H2O, 60 mg of manganese as MnSO4·H2O, 0.35 mg of iodine as KIO3, 20 mg of CuSO4·5H2O and 0.15 mg of ZnSO4·H2O. 3Mycosorb (Alltech, Nicholasville, KY).

Vitamin E Analysis Breast muscle samples for each diet group were saponified using a methanol-potassium hydroxide mixture and the content of α-tocopherol was measured using an HPLC method (Barella et al., 2004).

Microarray Gene Expression Profiling Muscle samples were stored at −80°C until RNA extraction. Tissues were disrupted and homogenized with Qiagen TissueRuptor (Qiagen, Valencia, CA). Total RNA isolation was performed with an RNeasy Minikit (Qiagen). Cytoplasmic RNA preparation, hybridization, and scanning were performed following the standard protocols recommended by Affymetrix (Affymetrix, Santa Clara, CA). Gene expression data were obtained for each of 7 birds of each diet group using the Chicken Genome Array (900592, Affymetrix).

Microarray Data Analysis GeneSpring GX 10.0 software (Silicon Genetics, Redwood, CA) was used to validate and normalize the mi-

Downloaded from http://ps.oxfordjournals.org/ by guest on January 12, 2015

other biological functions common to EcoE and high dietary VE. However, no such study that focuses on the mechanisms of this functional redundancy between VE and EcoE has yet been done. Advances in nutrigenomics, especially the application of high-throughput genomics technology to nutritional science, has created unprecedented opportunities for increasing our understanding of how nutrients modulate gene and protein expression to influence cellular metabolism, health consequences, and the related performance (Dawson, 2006; Kussmann et al., 2006; Crujeiras et al., 2008). The microarray technique is an ideal tool for gene expression profiling of a large number of genes in a single assay by using species-specific array platforms. Taking advantage of this technique, we had as our objective in this study to compare the genome-wide gene expression profiles of breast muscle from chicks fed EcoE or high VE diets, and to deduce the molecular mechanisms associated with high VE- and EcoEsupplemented diets and their functional redundancy at the level of transcriptomics.

diets1

138

Xiao et al.

croarray data and to perform statistical and gene expression pattern analyses. To minimize the possibility of misleading findings, probe sets with very low signal intensity and that were labeled as “absent” by the Affymetrix MAS5 algorithm across samples were excluded from further analysis. Only genes with P < 0.05 and a corresponding signal intensity fold change >1.2 were deemed as changed.

Bioinformatics Analysis

Real-Time Reverse Transcription-PCR To confirm the gene expression results from representative genes, 0.5 μg of total RNA was reverse transcribed into cDNA using a High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA) according to the manufacturer’s instructions. Real-time PCR was performed in triplicate using a Power SYBR Green PCR Master Mix (Applied Biosystems) and a 7500 Real-Time PCR System (Applied Biosystems). Primers were designed using Primer Express software version 2.0 (Applied Biosystems), and the corresponding sequences were as follows: sterol carrier protein 2 [SCP2; forward (F): 5′-CAGCTGTTTTTCCTGATGGA-3′, reverse (R): 5′-CCCAAAGGGAGAGTGTGGAAT-3′]; protein phosphatase 3 (F: 5′-GTGTGCTTCAATGCCCTTCTTTA-3′, R: 5′-AAAAGGATAATTGCTGCCAAATTAGTA-3′). Translin-associated factor X (F: 5′-TCTCCTTTGATGACGTCTTTTAAATC-3′, R: 5′-CCTGGTGTCGAGTTCCAGCT-3′) was selected as a housekeeping gene to account for any nonbiological variations that occurred in the process. The relative quantification was calculated as a ratio of the target gene to the control gene by using the delta-delta cycle

Statistical Analysis Measurements of serum TAC and muscle VE concentration were subjected to ANOVA by using the linear model of Statistix version 8 (Analytical Software, Tallahassee, FL). Significant differences among means were determined using Fisher’s least significant difference test. Real-time PCR results were analyzed using the GLM procedure of the SAS statistical package, version 9.1 (SAS Institute Inc., Cary, NC). Differences in mRNA fold changes were determined using Student’s t-test. Values are presented as means ± SEM, in which differences between treatment means were considered significant at P < 0.05.

RESULTS Effects of EcoE and High VE on Antioxidant Capacity of Broilers To study the effect of high VE or EcoE supplementation on the tissue antioxidant status, the antioxidant capacity of the serum collected at 3 and 6 wk was measured (Figure 1A). As shown in the figure, both high VE and EcoE significantly increased the value of total antioxidant capacity. Supplementation of VE increased dietary VE contents from 23.3 to 160 IU/kg in starter broiler diets, or from 14.9 to 104 IU/kg in grower diets (Table 1). However, assays on the VE content of breast muscle showed no significant dietary effects in this tissue in either 3- or 6-wk-old birds (Figure 1B). This indicates that the improvement in bird antioxidant status by VE and EcoE supplementation was through a mechanism that did not directly relate to VE content as measured in the muscle. In addition, the similar dietary Se levels between the diets used in this study indicated that dietary Se concentration was not the reason, or at least the sole reason, for the activities related to EcoE (Table 1).

Gene Expression Profiles in Breast Muscle Out of 28,000 genes represented in this gene array, approximately 50% of these were expressed in breast muscle. Compared with unsupplemented controls, the expression levels of 312 transcripts were altered in chickens fed E100 (179 upregulated, 133 downregulated), whereas birds fed E50 exhibited changes in 324 mRNA levels (166 upregulated, 158 downregulated). The broilers fed the EcoE-supplemented diet displayed altered expression in 285 transcripts (169 upregulated, 116 downregulated). Effects of dietary treatment on the expression patterns of these genes were illustrated

Downloaded from http://ps.oxfordjournals.org/ by guest on January 12, 2015

To analyze pathway information and networks of genes that were differentially expressed by supplementation of EcoE or a high level of dietary VE, the data set containing gene identifiers and corresponding expression fold changes were uploaded into Ingenuity Pathways Analysis software (Ingenuity Systems, Redwood City, CA). Each identifier was mapped to its corresponding gene object in the Ingenuity Knowledge Base (IKB). A network analysis was performed whereby focus genes (imported genes that were eligible for generating interaction networks based on IKB) were overlaid onto a global molecular network developed from information contained in the IKB. Networks of focus genes were then algorithmically generated based on their connectivity. A functional analysis was performed to determine the biological functions that were most significant to the genes in the data set, whereas the canonical pathway analysis identified pathways that were significant to the altered genes. Fisher’s exact test was used to calculate a P-value determining the probability that the association between the genes and the given network, biological function, or canonical pathway was due to random chance.

threshold (ΔΔCt) method. Conditions for PCR were as follows: 50°C for 2 min, 95°C for 10 min, 40 cycles of 95°C for 15 s, and 60°C for 1m, followed by a hold at 4°C.

TRANSCRIPTOMIC STUDY OF VITAMIN E AND AN ALGAE-BASED ANTIOXIDANT

139

in unsupervised hierarchical clustering based on both treatment (array, shown by the top dendrogram in Figure 2) and normalized gene expression (Figure 2). Further comparison of genes regulated by EcoE and VE

indicated that different as well as common genes were detected (Figure 3). Details of the genes changed by dietary supplementation of E100, E50, and EcoE are listed in Supplemental Tables 1, 2, and 3, respectively.

Figure 2. Unsupervised hierarchical clustering of differentially regulated genes in the breast muscle of broilers. Genes differentially regulated by E100, EcoE, or E50 (P < 0.05, fold change >1.2) were subjected to unsupervised hierarchical clustering based on both arrays and genes. In the heatmap, normalized gene expressions are shown in colors that reflect the expression changes compared with the mean value of each gene, where blue, red, or yellow colors represent decreased, increased, or no change in the level of expression intensity, respectively. The dendrogram on the top reflects the extent of similarity of expression profiles between arrays, whereas the dendrogram on the left side represents the changes in expression patterns of individual genes across the array. Values shown in the figure are threshold distances (measures of dissimilarity between clusters represented by the distance from the node to corresponding cluster leaves), where a smaller value indicates a higher similarity between expression profiles. E100 = control diet + 100 IU of vitamin E (VE)/kg; EcoE = control diet + 200 g/ton of an algae-based antioxidant (EconomasE, Alltech, Nicholasville, KY); E50 = control diet + 50 IU of VE/kg.

Downloaded from http://ps.oxfordjournals.org/ by guest on January 12, 2015

Figure 1. Effects of vitamin E (VE) and an algae-based antioxidant (EconomasE, Alltech, Nicholasville, KY; EcoE) on the total antioxidant capacity (TAC) and breast muscle VE concentration of broilers. Chicks from each diet group of the control diet, the E50 diet, the E100 diet, and the EcoE diet were killed and blood samples were taken via cardiac puncture for serum after 3 or 6 wk of feeding, respectively. A) Serum TAC was determined by using an antioxidant assay kit (Cayman Chemical Co., Ann Arbor, MI) and was quantified as millimolar Trolox equivalents on the ability of antioxidants in the sample to inhibit the oxidation of 2,2′-azino-di-(3-ethylbenzthiazoline sulfonate). B) Breast muscle VE concentration (ppm). Values presented are means ± SEM, n = 7. *P < 0.05, compared with the control. E50 = control diet + 50 IU of VE/kg; E100 = control diet + 100 IU of VE/kg; EcoE = control diet + 200 g/ton of an algae-based antioxidant (EconomasE).

140

Xiao et al.

VE-Regulated Genes and the Effects of EcoE on Their Expression If the biophysiological benefits related to a high level of VE are reflections of transcriptional changes, genes significantly changed by VE are likely to play key roles in mediating these effects. In support of this, several genes related to VE function, including heme oxygenase 2 (Nier et al., 2006), interferon γ receptor 2 (Han et al., 2006; Crujeiras et al., 2008), SCP2 (Cho et al., 2009),

protein phosphatase 3, and calpastatin, were among those significantly changed by E100 or E50 (Table 3). Real-time reverse transcription-PCR analysis also confirmed this dietary effect on the expression change of selected genes, including SCP2, a gene that plays important roles in maintaining cellular morphology and lipid metabolism (Atshaves et al., 2000; Seedorf et al., 2000; Fuchs et al., 2001), and protein phosphatase 3, which has been linked with growth factor-stimulated cell proliferation and signaling cascades, such as im-

Table 2. Top networks and biofunctions associated with genes commonly altered by the E100 and EcoE diets in the breast muscle of broilers1 Associated network functions2 Lipid metabolism, molecular transport, infection mechanism Energy production, molecular transport, nucleic acid metabolism Cell cycle, cellular development, skeletal and muscular system development and function Protein degradation, protein synthesis, cell-to-cell signaling and interaction Lipid metabolism, molecular transport, small molecule biochemistry Top biological functions Molecular and cellular functions   Carbohydrate metabolism   Lipid metabolism   Molecular transport   Nucleic acid metabolism   Cell cycle Physiological system development and function   Nervous system development and function   Organ development   Connective tissue development and function   Auditory and vestibular system development and function   Cardiovascular system development and function

Score2 26 26 23 21 21 Altered genes (no.) 8 6 11 6 10 6 4 7 1 4

1A total of 145 genes commonly changed by the EcoE and E100 diets were applied for Core analysis, which interpreted the data set in the context of biological processes, pathways, and molecular networks (Ingenuity Systems, Redwood City, CA). EcoE = control diet + 200 g/ton of an algae-based antioxidant (EconomasE, Alltech, Nicholasville, KY); E100 = control diet + 100 IU of vitamin E/kg. 2The 3 most significant functions for each network are listed. Network score was based on a P-value calculation, which calculated the likelihood that the network eligible molecules that were part of a network were found therein by random chance.

Downloaded from http://ps.oxfordjournals.org/ by guest on January 12, 2015

Figure 3. Venn diagram of genes differentially regulated by E50, E100, or EcoE in the breast muscle of broilers. Venn diagram representation of overall genes differentially regulated (P < 0.05, fold change >1.2) because of the E100 (312), E50 (324), or EcoE (285) treatment. A) Thirtyfive genes were commonly downregulated by all 3 diet treatments, whereas 54 genes were commonly suppressed by E100 and EcoE. B) Fifty-three genes were upregulated by E100, E50, and EcoE supplementation, whereas expression of 92 genes was commonly induced by E100 and EcoE. E50 = control diet + 50 IU of vitamin E/kg; E100 = control diet + 100 IU of vitamin E/kg; EcoE = control diet + 200 g/ton of an algae-based antioxidant (EconomasE, Alltech, Nicholasville, KY).

TRANSCRIPTOMIC STUDY OF VITAMIN E AND AN ALGAE-BASED ANTIOXIDANT Table 3. Select gene expression changes mediated by the E100 or EcoE Gene symbol

FC_ EcoE

FC_ E100

ABCC6

−1.4

−1.377

AGTR2 CDC34

1.262 −1.2

1.456 −1.366

CTSZ CYB5R4 DIO2

1.202 −1.31 −1.239

1.222 −1.321 −1.526

EXT2 FUBP1

2 1.3

1.825 1.457

GPD2

1.462

1.691

GSG2 GUCY2C

1.5 1.65

1.371 1.614

HEXA

1.266

1.476

Gene name

Associated biological functions

ATP-binding cassette, sub-family C, member 6 Angiotensin II receptor, type 2 Cell division cycle 34 homolog (Saccharomyces cerevisiae) Cathepsin Z Cytochrome b5 reductase 4 Deiodinase, iodothyronine, type II

Lipid metabolism, infection mechanism, molecular transport Protein trafficking, molecular transport, lipid metabolism Cell cycle, cellular assembly and organization

Exostosin 2 Far upstream element (FUSE) binding protein 1 Glycerol-3-phosphate dehydrogenase 2 (mitochondrial) Germ cell associated 2 Guanylate cyclase 2C (heat stable enterotoxin receptor) Hexosaminidase A (α polypeptide)

−1.41

−1.335

IFNGR2

1.25

1.235

KDM5B LSS MOCS2 PRPS2 PSEN1

1.6 −1.549 −1.44 −1.2 1.276

1.449 −1.385 −1.414 −1.577 1.35

PTN

1.238

2.021

Pleiotrophin

RBL1

1.979

1.212

Retinoblastoma-like 1 (p107)

SCP2

−1.818

−1.69

Sterol carrier protein 2

SKIL SLC38A1

−1.7 1.316

−1.525 1.459

SKI-like oncogene Solute carrier family 38, member 1

1.473

1.336

TRPC1

Hyaluronoglucosaminidase 1 Interferon gamma receptor 2 (interferon gamma transducer 1) Lysine (K)-specific demethylase 5B Lanosterol synthase Molybdenum cofactor synthesis 2 Phosphoribosyl pyrophosphate synthetase 2 Presenilin 1

UMPS HMOX2

−1.4 −1.43

−1.507 −1.2

Transient receptor potential cation channel, subfamily C, member 1 Uridine monophosphate synthetase Heme oxygenase 2

PPP3CA

−1.77

−1.42

Protein phosphatase 3

1.34

1.35

CAST

Calpastatin

Antigen presentation, cell morphology Carbohydrate metabolism, small molecule biochemistry Cell signaling, molecular transport, vitamin and mineral metabolism, drug metabolism, posttranslational modification Carbohydrate metabolism, cellular compromise Cell cycle Molecular transport, carbohydrate metabolism, nucleic acid metabolism Cell cycle Cell signaling, molecular transport, small molecule biochemistry, nucleic acid metabolism Molecular transport, carbohydrate metabolism, lipid metabolism, cell morphology Carbohydrate metabolism, cellular growth and proliferation, drug metabolism Cell signaling, innate immune responses Cell cycle Molecular transport, lipid metabolism Vitamin and mineral metabolism Nucleic acid metabolism, small molecule biochemistry Cell signaling, vitamin and mineral metabolism, carbohydrate metabolism, cellular growth and proliferation, cell cycle Cell signaling, cellular growth and proliferation, cell death, cell morphology Cellular growth and proliferation, cell cycle, cell morphology Molecular transport, vitamin and mineral metabolism, carbohydrate metabolism, lipid metabolism, cell morphology, cellular assembly and organization Cell cycle Molecular transport, small molecule biochemistry, amino acid metabolism Molecular transport, vitamin and mineral metabolism, carbohydrate metabolism, lipid metabolism Nucleic acid metabolism, small molecule biochemistry Cell signaling, oxidative stress response, inflammatory response Cell cycle, skeletal muscle fiber development, vitamin and mineral metabolism Cell cycle, negative regulation of catalytic activity

1Transcripts

altered (P < 0.05 and fold change >1.2, compared with the control diet) by the EcoE and E100 diets that were associated with the top molecular and cellular functions (cell signaling, molecular transport, vitamin and mineral metabolism, carbohydrate and lipid metabolism, and others) are shown in the table. Associations between gene and function is based on the Ingenuity Knowledge Base (Ingenuity Systems, Redwood City, CA). EcoE = control diet + 200 g/ton of an algae-based antioxidant (EconomasE, Alltech, Nicholasville, KY); E100 = control diet + 100 IU of vitamin E/kg.

mune responses and muscle development (Camps et al., 2000; Aramburu et al., 2004; Wang et al., 2008), whereas calpastatin, a gene that encodes a protease inhibitor and that was increased by E50 and EcoE on the microarray, was not different by reverse transcriptionPCR, although a similar pattern of dietary effects was observed (Figure 4). The reason for this discrepancy is not clear, but it may be related to the variations in and small changes of expression. Because the principal interest of this study was to understand the mechanisms by which EcoE saves VE,

the study focused on the effects of EcoE on genes differentially regulated by VE, and it may provide a clearer view of this functional overlap. As shown in Figure 5, expression pattern analysis focusing on genes that were changed by E100 indicated a substantial similarity between the effects of EcoE and E100 on these genes, which is visually evident and is indicated by the smallest threshold distance (a measure of dissimilarity between clusters) between their expression profiles. In contrast, the expression intensity of most of these genes in birds fed E50 was somewhere in between that of the

Downloaded from http://ps.oxfordjournals.org/ by guest on January 12, 2015

HYAL1

141

diet1

142

Xiao et al.

control diet and the E100 diet. Hence, further pathway analysis focused on genes that were regulated by those 2 treatments.

Gene Pathways Altered by EcoE or High VE

DISCUSSION Many claims have been made about the potential of VE to promote human health and prevent and treat diseases such as cancer, cardiovascular diseases, and neurodegenerative diseases (Chan, 1998; Bunout, 2000; Capuron et al., 2009; Saremi and Arora, 2010). In the animal industry, high-level dietary supplementation of VE is extensively applied to achieve better growth and reproductive performance, feed efficiency, and preferred meat product quality (Kennedy et al., 1992). The mechanism by which VE might provide these benefits is popularly credited to its strong antioxidative capability (Waylan et al., 2002; Guo et al., 2006). Indeed, measurement of total antioxidant capacity indicated that high-level VE improved the antioxidative capability of the birds. Further, gene expression analysis of breast

Figure 4. Real-time reverse transcription-PCR validation of microarray results. The RNA used for real-time reverse transcriptionPCR was from the same birds (n = 7 per diet group) as those used for microarray analysis. Values represented are least squares means ± SEM; selected genes are sterol carrier protein 2 (SCP2), protein phosphatase 3 (PPP3CA), and calpastatin (CAST). An asterisk (*) indicates a significant difference (P < 0.05) compared with control, whereas a carat (^) indicates a tendency of the effect (0.05 < P < 0.1). E50 = control diet + 50 IU of vitamin E/kg; E100 = control diet + 100 IU of vitamin E/kg; EcoE = control diet + 200 g/ton of an algaebased antioxidant (EconomasE, Alltech, Nicholasville, KY).

Downloaded from http://ps.oxfordjournals.org/ by guest on January 12, 2015

To explore the nature of the biological effect related to high VE and EcoE in broilers, genes that were differentially regulated by VE and EcoE were categorized according to gene ontology attributes by use of Ingenuity Pathways Analysis software (Figure 6). Genes altered by E100 were significantly enriched in biofunctional categories, including cell signaling, cell morphology, nervous system development and function, skeletal and muscular system development and function, and numerous others (Figure 6A). Perhaps the most interesting observation of this analysis, however, was that multiple metabolic processes, such as lipid metabolism, carbohydrate metabolism, and vitamin and drug metabolism, were significantly altered by E100. Analysis of EcoE-altered genes indicated that a similar number of functional categories was significantly affected in breast muscle (Figure 6B), which included 15 of the 28 biofunctions changed by E100, including those related to lipid and carbohydrate metabolism, cell death, morphology, and others. Further, EcoE supplementation caused alterations related to gene expression and amino acid metabolism in muscle cells, which might indicate a role of EcoE in skeletal muscle protein synthesis and metabolism. Because of the great similarity of individual genes changed by E100 and EcoE supplementation and the apparent overlay on functional analysis, 145 genes commonly changed by EcoE and E100 (Figure 3) were used for further functional analysis. As indicated in Table 2, the top gene networks and biofunctions represented are those related to cellular function and maintenance, cell signaling, lipid metabolism, nucleic acid metabolism, energy production, protein synthesis, and protein degradation. Genes involved in these cellular functions are summarized in Table 3.

muscle indicated that expression of genes related to cellular oxidation and response, such as heme oxygenase 2 and SCP2 (Ewing and Maines, 1997; Kannenberg et al., 1999; Krönke et al., 2003), were significantly altered by high VE in breast muscle. In addition to its activities as an antioxidant, VE is involved in other cellular processes, such as immune function, metabolic processes, and many other cell-signaling pathways (Brigelius-Flohé and Traber, 1999). It has even been suggested that the most important function of VE is as a signaling molecule (Azzi, 2007). In the current study, pathway analysis of the genes altered by VE supplementation revealed that gene networks relevant to cell cycling, morphology, cellular assembly and organization, nervous system development, and skeletal and muscular development and function were significantly changed, indicating the extensive effect of VE supplementation at the transcriptional level. The genes altered by VE that were related to immune function, lipid metabolism, carbohydrate metabolism, drug metabolism, and vitamin and mineral metabolism may play key roles related to the activities of VE supplementation. Alterations in gene expression and corresponding bioinformatics indicated that cellular oxidation and other gene networks are responsible for the mechanisms related to the actions of high VE supplementation. However, the most interesting observation in this study was the great similarity between the breast muscle gene expression profile in response to EcoE or VE supplementation. Because EcoE is a product that aims to

TRANSCRIPTOMIC STUDY OF VITAMIN E AND AN ALGAE-BASED ANTIOXIDANT

143

improve the capability of oxidative resistance (Pierce et al., 2009), it was not surprising to see the increased serum total antioxidant capacity value, a response similar to that of VE supplementation. A side-by-side comparison of the genes regulated by VE and EcoE indicated that most genes changed by VE were also sensitive to EcoE supplementation (Figure 5). Because EcoE is a product containing Se yeast and extensive overlapping of biological functions exists for VE and Se (Combs et al., 1975; Awad et al., 1994; Hill et al., 2001), one may wonder what the role of Se is in the VE-saving effect of EcoE. However, the similar Se level in all the diets used in this study, including the control, suggests that dietary Se concentration was likely not the primary reason for the observed actions of EcoE. Because the main focus of the current study was to investigate the effects of high VE on muscle transcriptional profiles and the mechanisms related to the VEsaving effects of EcoE, genes commonly regulated by these supplementations were further categorized into biologically relevant networks based on the functional

class of the gene product. As described above, the top gene networks represented by these genes are those related to cellular function and maintenance, such as lipid metabolism, nucleic acid metabolism, vitamin and mineral metabolism, protein synthesis and degradation, immune responses, and cell morphology (Figure 6). Potential links of these alterations with activities seen in high-level VE and EcoE supplementation may be explained by using the following examples. Lipid oxidation is considered the primary mechanism causing the deterioration of quality in meat products (Buckley et al., 1995; Cortinas et al., 2005). Different nutritional strategies have been explored in an attempt to delay the onset of lipid oxidation and improve the quality of poultry meat (Lewis et al., 2002; Kennedy et al., 2005). Supplementing chicken feed with VE protects against lipid oxidation in poultry meat (Young et al., 2003). Alterations in genes important to lipid metabolism, such as angiotensin receptor 2, hexosaminidase A, lanosterol synthase, transient receptor protein 1, and SCP2 (Table 3), may at least partly explain the longer

Downloaded from http://ps.oxfordjournals.org/ by guest on January 12, 2015

Figure 5. Similar effects of EcoE on expression of E100-regulated genes. A total of 312 genes differentially regulated (P < 0.05, fold change >1.2) by E100 were included in an unsupervised hierarchical clustering analysis. In the heatmap, normalized gene expression is shown in colors that reflect the expression changes compared with the mean value of each gene, where blue, red, or yellow color represents decreased, increased, or no change in the level of expression intensity, respectively. The dendrogram on the top reflects the extent of similarity of expression profiles between diets, while the dendrogram on the left side represents the changes of expression pattern of individual genes across the diets. Values shown in the figure are threshold distances, where a smaller number indicates a higher similarity between expression profiles. E100 = control diet + 100 IU of vitamin E/kg; EcoE = control diet + 200 g/ton of an algae-based antioxidant (EconomasE, Alltech, Nicholasville, KY); E50 = control diet + 50 IU of vitamin E/kg.

144

Xiao et al.

shelf life observed for meat products related to VE and EcoE supplementation (Sanders et al., 1997; Pierce et al., 2009). Among these genes, SCP2 is a peroxisomeassociated thiolase that plays a crucial role in the trafficking and metabolism of cholesterol and other lipids in mammalian cells. Overexpression of SCP-2 has been related to downregulation of proteins involved in cholesterol transport (e.g., liver fatty acid binding protein and scavenger receptor B1), cholesterol synthesis (e.g., sterol-responsive element-binding protein 2 and 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase), and bile acid oxidation and transport (Atshaves et al., 2009). There is evidence indicating that SCP2 may facilitate lipid hydroperoxide transfer between membranes, and thereby potentially enhance dissemination of peroxidative damage under oxidative stress, as suggested by a faster loss of mitochondrial membrane potential and the development of apoptosis (Vila et al., 2004; Kriska et al., 2006). The lower expression of SCP2 in chicks fed the VE or EcoE diet may indicate a lower oxidative

stress state in these cells as well as changes in membrane lipid synthesis and transport. The connections between VE or EcoE supplementation and the genes responsible for cell morphology, which are important in maintaining the integrity of the cell membrane under stress conditions, may also be important in activities associated with VE and EcoE. Although expression changes of SCP2 directly affect the structure of the plasma membrane, genes in this category also included angiotensin receptor 2, presenilin 1, cathepsin Z, hexosaminidase, pleiotrophin, and retinoblastoma-like 1 (Table 3). The regulation of genes involved in immune response and cellular anti-inflammatory diseases confirmed the role of VE as a booster of the immune system (Bendich et al., 1986; Capuron et al., 2009). Increased expression of interferon γ receptor 2, a gene that encodes the non-ligand-binding β chain of the γ interferon receptor, may indicate an enhanced resistance capability to mycobacterial diseases (Dorman et al., 1999; Hwang et al., 2006).

Downloaded from http://ps.oxfordjournals.org/ by guest on January 12, 2015

Figure 6. Biological functions associated with the genes altered by dietary treatments in the breast muscle of broilers. Genes that were differentially expressed (compared with the control) in broilers fed the E100 (A) or EcoE (B) diet were categorized according to their gene ontology attributes and functional connectivity defined by Ingenuity Pathways Analysis software (Ingenuity Systems, Redwood City, CA). The P-value associated with a biological process or pathway is calculated with a right-tailed Fisher’s exact test. A P < 0.01 [shown as −log (P-value) >2 in the figure] was deemed a significant enrichment of altered genes in a specified biological function. Gene classes with bold name were commonly enriched in the E100 and EcoE groups. E100 = control diet + 100 IU of vitamin E/kg; EcoE = control diet + 200 g/ton of an algae-based antioxidant (EconomasE, Alltech, Nicholasville, KY). Color version available in the online PDF.

TRANSCRIPTOMIC STUDY OF VITAMIN E AND AN ALGAE-BASED ANTIOXIDANT

ACKNOWLEDGMENTS The authors thank other members of the Alltech– University of Kentucky Nutrition Research Alliance for assistance with birds and tissue collection. We also thank Kate Jacques (Alltech, Nicholasville, KY) for helpful discussions and editing of the manuscript.

REFERENCES Aramburu, J., J. Heitman, and G. R. Crabtree. 2004. Calcineurin: A central controller of signalling in eukaryotes. EMBO Rep. 5:343–348. Atshaves, B. P., A. L. McIntosh, G. G. Martin, D. Landrock, H. R. Payne, S. Bhuvanendran, K. K. Landrock, O. I. Lyuksyutova, J. D. Johnson, R. D. Macfarlane, A. B. Kier, and F. Schroeder. 2009. Overexpression of sterol carrier protein-2 differentially alters hepatic cholesterol accumulation in cholesterol-fed mice. J. Lipid Res. 50:1429–1447. Atshaves, B. P., O. Starodub, A. McIntosh, A. Petrescu, J. B. Roths, A. B. Kier, and F. Schroeder. 2000. Sterol carrier protein-2 alters high density lipoprotein-mediated cholesterol efflux. J. Biol. Chem. 275:36852–36861. Awad, J. A., J. D. Morrow, K. E. Hill, L. J. Roberts 2nd, and R. F. Burk. 1994. Detection and localization of lipid peroxidation in selenium- and vitamin E-deficient rats using F2-isoprostanes. J. Nutr. 124:810–816. Azzi, A. 2007. Molecular mechanism of alpha-tocopherol action. Free Radic. Biol. Med. 43:16–21.

Barella, L., P. Y. Muller, M. Schlachter, W. Hunziker, E. Stocklin, V. Spitzer, N. Meier, S. de Pascual-Teresa, A. M. Minihane, and G. Rimbach. 2004. Identification of hepatic molecular mechanisms of action of alpha-tocopherol using global gene expression profile analysis in rats. Biochim. Biophys. Acta 1689:66–74. Bendich, A., E. Gabriel, and L. J. Machlin. 1986. Dietary vitamin E requirement for optimum immune responses in the rat. J. Nutr. 116:675–681. Brigelius-Flohé, R., and M. G. Traber. 1999. Vitamin E: Function and metabolism. FASEB J. 13:1145–1155. Buckley, D. J., P. A. Morrissey, and J. I. Gray. 1995. Influence of dietary vitamin E on the oxidative stability and quality of pig meat. J. Anim. Sci. 73:3122–3130. Bunout, D. 2000. Therapeutic potential of vitamin E in heart disease. Expert Opin. Investig. Drugs 9:2629–2635. Camps, M., A. Nichols, and S. Arkinstall. 2000. Dual specificity phosphatases: A gene family for control of MAP kinase function. FASEB J. 14:6–16. Capuron, L., A. Moranis, N. Combe, F. Cousson-Gelie, D. Fuchs, V. De Smedt-Peyrusse, P. Barberger-Gateau, and S. Laye. 2009. Vitamin E status and quality of life in the elderly: Influence of inflammatory processes. Br. J. Nutr. 102:1390–1394. Chan, A. C. 1998. Vitamin E and atherosclerosis. J. Nutr. 128:1593– 1596. Cho, J. Y., D. W. Kang, X. Ma, S. H. Ahn, K. W. Krausz, H. Luecke, J. R. Idle, and F. J. Gonzalez. 2009. Metabolomics reveals a novel vitamin E metabolite and attenuated vitamin E metabolism upon PXR activation. J. Lipid Res. 50:924–937. Choct, M., A. J. Naylor, and N. Reinke. 2004. Selenium supplementation affects broiler growth performance, meat yield and feather coverage. Br. Poult. Sci. 45:677–683. Coetzee, G. J. M., and L. C. Hoffman. 2001. Effect of dietary vitamin E on the performance of broilers and quality of broiler meat during refrigerated and frozen storage. S. Afr. J. Anim. Sci. 31:158–173. Combs, G. F. Jr 1981. Influences of dietary vitamin E and selenium on the oxidant defense system of the chick. Poult. Sci. 60:2098–2105. Combs, G. F. Jr., T. Noguchi, and M. L. Scott. 1975. Mechanisms of action of selenium and vitamin E in protection of biological membranes. Fed. Proc. 34:2090–2095. Cortinas, L., A. Barroeta, C. Villaverde, J. Galobart, F. Guardiola, and M. D. Baucells. 2005. Influence of the dietary polyunsaturation level on chicken meat quality: Lipid oxidation. Poult. Sci. 84:48–55. Crujeiras, A. B., D. Parra, F. I. Milagro, E. Goyenechea, E. Larrarte, J. Margareto, and J. A. Martinez. 2008. Differential expression of oxidative stress and inflammation related genes in peripheral blood mononuclear cells in response to a low-calorie diet: A nutrigenomics study. OMICS 12:251–261. Dawson, K. A. 2006. Nutrigenomics: Feeding the genes for improved fertility. Anim. Reprod. Sci. 96:312–322. Dorman, S. E., G. Uzel, J. Roesler, J. S. Bradley, J. Bastian, G. Billman, S. King, A. Filie, J. Schermerhorn, and S. M. Holland. 1999. Viral infections in interferon-gamma receptor deficiency. J. Pediatr. 135:640–643. Ewing, J. F., and M. D. Maines. 1997. Histochemical localization of heme oxygenase-2 protein and mRNA expression in rat brain. Brain Res. Brain Res. Protoc. 1:165–174. Feedinfo News Service. 2008. Vitamin E Feedgrade: Market to Stay Tight This Summer. Feedinfo News Service, Labège, France. Accessed June 30, 2008. http://www.feedinfo.com/console/PageViewer.aspx?page=925674&public=yes. Feedinfo News Service. 2009. Exclusive Interview: BASF Discusses the Vitamin E and A Markets. Feedinfo News Service, Labège, France. Accessed Oct. 6, 2009. http://www.feedinfo.com/console/PageViewer.aspx?page=1421596&public=yes. Fuchs, M., A. Hafer, C. Munch, F. Kannenberg, S. Teichmann, J. Scheibner, E. F. Stange, and U. Seedorf. 2001. Disruption of the sterol carrier protein 2 gene in mice impairs biliary lipid and hepatic cholesterol metabolism. J. Biol. Chem. 276:48058–48065. Guo, Q., B. T. Richert, J. R. Burgess, D. M. Webel, D. E. Orr, M. Blair, G. E. Fitzner, D. D. Hall, A. L. Grant, and D. E. Gerrard.

Downloaded from http://ps.oxfordjournals.org/ by guest on January 12, 2015

Similar effects of EcoE and VE on muscle transcriptional profiles and their connections to a series of gene networks that are functionally associated with the biophysiological roles of high VE supplementation may explain why the addition of EcoE to the diet reduced the amount of VE required for broilers. Differences between VE and EcoE in the gene expression profile were also apparent. For example, genes involved in drug metabolism and hepatic system development and function were significantly enriched in genes altered by VE, whereas this activity was not apparent in EcoE. On the other hand, genes related to amino acid metabolism and several other functional classes were significant only in genes altered by EcoE (Figure 6). Further studies focusing on these product-specific activities may provide information on the nonredundant biophysiological roles related to VE and EcoE. In conclusion, transcriptional profiling and pathway analysis revealed differential expression with high VE supplementation when compared with a low-VE control diet and identified gene networks that likely play key roles related to the nutriphysiological functions of VE. The EcoE product mimicked many of the effects of VE on breast muscle gene expression profiles, which may explain why inclusion of EcoE decreased the required amount of VE in broiler diets. This study also indicates that microarray and gene expression studies can be a useful model in the study of molecular mechanisms associated with nutritional factors, and the results will facilitate the development of novel dietary strategies to meet the needs of the animal health industry.

145

146

Xiao et al. Lewis, S. J., A. Velasquez, S. L. Cuppett, and S. R. McKee. 2002. Effect of electron beam irradiation on poultry meat safety and quality. Poult. Sci. 81:896–903. Miller, N., C. Rice-Evans, and M. J. Davies. 1993. A new method for measuring antioxidant activity. Biochem. Soc. Trans. 21(2):95S. Nier, B., P. D. Weinberg, G. Rimbach, E. Stocklin, and L. Barella. 2006. Differential gene expression in skeletal muscle of rats with vitamin E deficiency. IUBMB Life 58:540–548. Pierce, J. L., T. Ao, R. F. Power, K. A. Dawson, A. J. Pescatore, A. H. Cantor, and M. J. Ford. 2009. Investigation of replacing vitamin E with EconomasE® in broiler diet. Poult. Sci. 87(Suppl. 1):97. (Abstr.) Richter, G., A. Lemser, G. Jahreis, and G. Steinbach. 1987. [Vitamin E requirement of chicks and young hens]. Arch. Tierernahr. 37:1029–1039. Sanders, S. K., J. B. Morgan, D. M. Wulf, J. D. Tatum, S. N. Williams, and G. C. Smith. 1997. Vitamin E supplementation of cattle and shelf-life of beef for the Japanese market. J. Anim. Sci. 75:2634–2640. Saremi, A., and R. Arora. 2010. Vitamin E and cardiovascular disease. Am. J. Ther. 17:e56–e65. Seedorf, U., P. Ellinghaus, and J. Roch Nofer. 2000. Sterol carrier protein-2. Biochim. Biophys. Acta 1486:45–54. Vila, A., V. V. Levchenko, W. Korytowski, and A. W. Girotti. 2004. Sterol carrier protein-2-facilitated intermembrane transfer of cholesterol- and phospholipid-derived hydroperoxides. Biochemistry 43:12592–12605. Wang, K., Y. Song, D. B. Chen, and J. Zheng. 2008. Protein phosphatase 3 differentially modulates vascular endothelial growth factor- and fibroblast growth factor 2-stimulated cell proliferation and signaling in ovine fetoplacental artery endothelial cells. Biol. Reprod. 79:704–710. Waylan, A. T., P. R. O’Quinn, J. A. Unruh, J. L. Nelssen, R. D. Goodband, J. C. Woodworth, M. D. Tokach, and S. I. Koo. 2002. Effects of modified tall oil and vitamin E on growth performance, carcass characteristics, and meat quality of growing-finishing pigs. J. Anim. Sci. 80:1575–1585. Wilson, R. P., P. R. Bowser, and W. E. Poe. 1984. Dietary vitamin E requirement of fingerling channel catfish. J. Nutr. 114:2053– 2058. Young, J. F., J. Stagsted, S. K. Jensen, A. H. Karlsson, and P. Henckel. 2003. Ascorbic acid, alpha-tocopherol, and oregano supplements reduce stress-induced deterioration of chicken meat quality. Poult. Sci. 82:1343–1351.

Downloaded from http://ps.oxfordjournals.org/ by guest on January 12, 2015

2006. Effects of dietary vitamin E and fat supplementation on pork quality. J. Anim. Sci. 84:3089–3099. Hakkarainen, J., J. Tyopponen, and L. Jonsson. 1986. Vitamin E requirement of the growing rat during selenium deficiency with special reference to selenium dependent- and selenium-independent glutathione peroxidase. Zentralbl. Veterinarmed. [C] 33:247–258. Han, S. N., O. Adolfsson, C. K. Lee, T. A. Prolla, J. Ordovas, and S. N. Meydani. 2006. Age and vitamin E-induced changes in gene expression profiles of T cells. J. Immunol. 177:6052–6061. Hill, K. E., A. K. Motley, X. Li, J. M. May, and R. F. Burk. 2001. Combined selenium and vitamin E deficiency causes fatal myopathy in guinea pigs. J. Nutr. 131:1798–1802. Hwang, J. H., W. J. Koh, E. J. Kim, E. H. Kang, G. Y. Suh, M. P. Chung, H. Kim, and O. J. Kwon. 2006. Partial interferon-gamma receptor deficiency and non-tuberculous mycobacterial lung disease. Tuberculosis (Edinb.) 86:382–385. Kannenberg, F., P. Ellinghaus, G. Assmann, and U. Seedorf. 1999. Aberrant oxidation of the cholesterol side chain in bile acid synthesis of sterol carrier protein-2/sterol carrier protein-x knockout mice. J. Biol. Chem. 274:35455–35460. Kennedy, D. G., D. A. Rice, E. A. Bruce, E. A. Goodall, and S. G. McIlroy. 1992. Economic effects of increased vitamin E supplementation of broiler diets on commercial broiler production. Br. Poult. Sci. 33:1015–1023. Kennedy, O. B., B. J. Stewart-Knox, P. C. Mitchell, and D. I. Thurnham. 2005. Vitamin E supplementation, cereal feed type and consumer sensory perceptions of poultry meat quality. Br. J. Nutr. 93:333–338. Kriska, T., V. V. Levchenko, W. Korytowski, B. P. Atshaves, F. Schroeder, and A. W. Girotti. 2006. Intracellular dissemination of peroxidative stress. Internalization, transport, and lethal targeting of a cholesterol hydroperoxide species by sterol carrier protein-2-overexpressing hepatoma cells. J. Biol. Chem. 281:23643–23651. Krönke, G., V. N. Bochkov, J. Huber, F. Gruber, S. Bluml, A. Furnkranz, A. Kadl, B. R. Binder, and N. Leitinger. 2003. Oxidized phospholipids induce expression of human heme oxygenase-1 involving activation of cAMP-responsive element-binding protein. J. Biol. Chem. 278:51006–51014. Kussmann, M., F. Raymond, and M. Affolter. 2006. OMICS-driven biomarker discovery in nutrition and health. J. Biotechnol. 124:758–787. Leeson, S. 2007. Vitamin requirements: Is there a basis for re-evaluating dietary specifications? World’s Poult. Sci. J. 63:255–266.