| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL PRODUCTION |
* Division of Animal Sciences, and
and
Department of Statistics, University of Missouri, Columbia 65211
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: deoxyribonucleic acid microarray • fescue • gene expression • mouse
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Fescue toxicosis leads to reduced feed intake and decreased milk yield, reproduction, and weight gain (Schmidt and Osborn, 1993; Thompson and Stuedemann, 1993; Paterson et al., 1995). Grazing endophyte-infected tall fescue decreases serum prolactin, cholesterol, progesterone, and LH in cattle (Burke et al., 2001). The major toxins associated with endophyte-infected grasses and with fescue toxicosis are ergopeptine alkaloids, primarily ergovaline and lysergic acid amides.
Laboratory mice were used previously as a model for fescue toxicosis because these animals exhibit reduced growth, reproduction, and lactation when fed endophyte-infected tall fescue seed (Zavos et al., 1987; Godfrey et al., 1994; Miller et al., 1994). However, the molecular mechanisms by which the toxins affect animals are not clear. Mice selected for sensitivity to fescue toxins (Hohenboken and Blodgett, 1997) were used in this study with the objective of identifying some of the molecular pathways by which endophyte-infected seeds may cause fescue toxicosis. In mammals, the liver plays an important role in detoxification and metabolism of toxins, and is a prime target of tissue injury in response to various physiological challenges. Consequently, we use microarray-based expression profiling to study the transcriptional response of mouse liver genes to the ergopeptine alkaloids.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The Animal Care and Use Committee, University of Missouri—Columbia, approved all procedures and protocols.
Mice used in this study were developed at Virginia Polytechnic Institute and State University (Blacksburg, VA) by Hohenboken and Blodgett (1997). Briefly, mice from the ICR strain (Harlan Sprague Dawley, Indianapolis, IN) were divergently selected for 8 generations on an index reflecting the impact of a diet containing endophyte-infected fescue seed on growth rate after weaning. Two lines of mice were produced; one, designated sensitive, had slow growth on the endophyte diet, and the other, designated resistant, had rapid growth. Only mice from the sensitive line were used in the present experiment for gene expression analysis.
In the current study, pups from 11 litters were weaned at 21 d of age. Twenty-seven mice (14 male, 13 female) were randomly allocated to receive diets containing 50% endophyte-infected (E+; Seed Research of Oregon, Inc., Corvallis, OR) or endophyte-free (E–; Missouri Southern Seed, Rolla, MO) fescue seed for a period of 2 wk, from 47 to 60 d of age, under thermoneutral conditions (24 ± 1°C). Fescue seeds and rodent chow (Formulab Diet # 5008; PMI Feeds, St. Louis, MO) were ground to pass a 1-mm screen and were mixed in equal parts.
All mice were housed in individual cages with relative humidity maintained at 35 to 50% and a 12:12 light-dark cycle, with lights on at 0700. All mice were weighed on d 1, 7, and 14 of the experiment. At the end of the experiment, mice were euthanized with carbon dioxide gas followed by cervical dislocation. Liver tissues from all mice were weighed and snap frozen in liquid nitrogen.
RNA Extraction
Extraction of RNA was done by using the RNAqueous-Midi kit according to the manufacturer’s instructions (Ambion Inc., Austin, Texas). Concentration of RNA was determined by using microcon-30 filters according to the manufacturer’s instructions (Millipore Corp., Bedford, MA). To remove contaminating DNA, the RNA was DNase-treated by using a DNA-Free kit (Ambion, Inc.). The DNase-treated RNA was then treated with phenol:choloroform:isoamyl alcohol (25:24:1) in a phase-lock tube (Eppendorf, Hamburg, Germany). Concentrated RNA was checked for quality and integrity using agarose gel electrophoresis. The RNA was quantified by taking optical density readings on an ND-1000 Spectrophotometer (Nanodrop Technologies, Wilmington, DE). The DNase-treated RNA samples were kept at –80°C until used.
Microarray Preparation
Microarrays were prepared by printing 1,353 oligos, 50-mers in length, representing rat genes expressed in liver, on Pan epoxy glass slides (MWG Biotech AG, Ebersberg, Germany). As external controls, the arrays also contained 10 Arabidopsis genes, each of them printed 10 times across the array (Spot report oligo, Stratagene, Cedar Creek, TX). Blank spots and 3x SSC were used as negative controls. Mouse Cot-1 (Applied Genetics Laboratories Inc., Melbourne, FL) and poly-A DNA were used to detect nonspecific hybridization. A robotic microarray printer with 16 printing tips was used to spot the oligos. The rat liver oligo arrays had a total of 1,920 spots and were organized in 16 blocks arrayed in 4 rows and 4 columns. Each block had 120 spots arranged in 10 rows and 12 columns. Slides were stored in the dark until used.
Microarray Protocol
Samples of RNA from mice in the study and from a reference RNA pool (Universal Mouse Reference RNA, Stratagene, La Jolla, CA) were used in the hybridization. This Universal Mouse Reference RNA is a pool of RNA extracted from 11 mouse cell lines. Sample or reference RNA (15 µg) containing 10 different Arabidopsis mRNA were added at varying concentrations as external controls and were primed using 1 µL of oligo dT (10 µg/µL; IDT DNA, Coralville, IA), and 1 µL of random hexamers (10 µg/µL; IDT DNA) at 70°C for 10 min and then chilled on ice for 10 min. The cDNA synthesis was done by adding 3.0 µL (50 units/µL) of Stratascript reverse transcription (StrataScript RT, Stratagene), 3.0 µL of 10x Stratascript RT buffer, 3.0 µL of dithiothreitol (0.1 M), 0.6 µL of 50x amino-allyl dUTP (Sigma Chemicals, St. Louis, MO), and 5 µL of DNase- and RNase-free water. The reverse transcription was done at 42°C for 2 h. Degradation and hydrolysis of RNA was done by adding 10 µL of 1 N NaOH and 10 µL of 0.5 M EDTA to the mix, then the reaction was incubated at 65°C for 15 min followed by neutralization with 25 µL of 1 M Tris, pH 7.4.
The cDNA that was obtained was purified using micro-con-30 filters (Millipore Corp., Bedford, MA) and dried in a Centrivap concentrator (Labconco Corp., Kansas City, MO). Dried cDNA was resuspended in 10 µL of 0.5 M sodium bicarbonate buffer, pH 9.0, at room temperature for 10 to 15 min. The cDNA concentrations were quantified by using an ND-1000 spectrophotometer (Nanodrop Technologies, Wilmington, DE). Labeling of cDNA was done by indirect labeling with Cy5 or Cy3 dyes (Amersham Biosciences, UK). The Cy3 and Cy5 dyes were prepared by addition of 10 µL of DMSO to 1 mg of dye; 1.5 µL of the mixture was placed into a new tube, dried in the Centrivap, and stored for future use. The labeling reaction was carried out for 1 h at room temperature in darkness. The removal of unincorporated dyes and concentration of labeled cDNA were done by using a Qia-quick PCR purification kit (Qiagen, Valencia, CA).
Microarray Hybridization and Washing
Labeled cDNA from the samples and the reference were mixed and dried in the Centrivap. The dried mixture was then resuspended in a hybridization buffer containing 2.0 µL of poly-A (10 µg/µL; IDT DNA), 6.25 µL of mouse Cot-1 DNA (3.2 µg/µL; Applied Genetics Laboratories Inc., Melbourne, FL), 3.0 µL of 20x SSC, 0.45 µL of 10% SDS, 0.5 µL of 1x HEPES (pH: 7.0), 8.25 µL of water, and subsequently hybridized to the arrays for 12 to 14 h at 65°C in a humidified hybridization chamber. This temperature of hybridization was chosen to ensure high specificity hybridization of mouse RNA to rat oligonucleotides.
After hybridization, the slides were washed in solution 1 (340 mL of water, 10 mL of 20x SSC, and 1 mL of 10% SDS) for 5 min, washed in solution 2 (350 mL of water, 1 mL of 20x SSC) for 5 min, and then centrifuged for 5 min to dry the slide. Extracted RNA from each of the 6 (3 male, 3 female) randomly selected E+ mice and 7 (4 male, 3 female) E– mice treatment groups were individually hybridized to the array with reference RNA in a reference microarray design. In a reference design, each experimental sample is hybridized against a common reference RNA sample (Churchill, 2002). Two or 3 replicates (arrays) were done per animal, out of which 1 was done in a dye swap design. A total of 35 arrays were done across all the animals and treatment groups.
Microarray Scanning and Data Acquisition
Slides were scanned using an Axon GenePix 4000B scanner (Axon Instruments Inc., Union City, CA) at 5-µm resolution, and the image was labeled and stored in the BioArray Software Environment (BASE) database (Saal et al., 2002). Gridding was done using GenePix Pro 4.0.1.12 software (Axon Instruments Inc., Union City, CA). Data files were linked to the appropriate image file and then stored in the BASE database. The raw data files on which this paper is based have been deposited with National Center for Biotechnology Information Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/), with series number GSE2134 and platform number GLP1786.
Statistical Analyses of Microarray Data
Microarray data were filtered in the BASE database to remove control spots, 3x SSC, and blank spots. The probe intensities from each array were treated as statistically independent by the ANOVA model (Kerr and Churchill, 2001; Wolfinger et al., 2001). The log base 2 intensities for each array spot were analyzed using a mixed model approach consisting of 2 steps, following the procedures of Kerr and Churchill (2001) and Wolfinger et al. (2001). The first step involved normalization across genes and the second step involved normalization within gene. The normalized data were then analyzed in a gene-specific analysis to test the effect of E+ vs. E– diets on expression of individual genes. The normalization model in the first step was
 | [1] |
in which Yagp represents the Log 2 of the observed fluorescent intensity signal from each gene on the array; µ is the overall mean value; Aa is the main effect of array a; Pp is the main effect of printing pin p (1 of the 16 pins used to print oligos on the array); (AP)ap is the effect of pin p within array a; and 
agp is the stochastic error. The residual (Ragp) from step 1 was obtained by subtracting the fitted values for the effects from the Yagp values. Array and pin accounted for 25% of the variance in the data. The second step of the statistical analysis consisted of gene-specific models for the residuals (Ragp) obtained from the normalization approach discussed above. These models were
 | [2] |
or
 | [3] |
in which µg is the mean value for gene g; (GD)gd is the gene g by dye d interaction on array a; (GT)gt is the interaction term for gene g by treatment t (these values were used to do gene-specific t-tests and to find differentially expressed genes between treatment groups); (GAn(T))gnt is a random effect for animal An within treatment t by gene g; (GAn)gn in Equation 3 is the random effect for gene g by animal An; and dgnt and dgn are the stochastic errors obtained from the gene-specific models.
These analyses were computed by using the Mixed Procedure of SAS (SAS Inst. Inc., Cary, NC). To obtain lists of differentially expressed genes, analyses were done using equal and unequal variances between treatment groups. The (GAn)gn values obtained for genes from equation 3 were used in a significance analysis of microarray (SAM, Tusher et al., 2001). Gene lists from the 3 analyses (Table 1) were used to identify differentially expressed genes due to the E+ vs. E– treatment, and later to select genes for quantitative real-time PCR analysis (qPCR). The Ragp values obtained from the stage 1 normalization model were also used in a nonparametric Wilcoxon test. The probability values obtained from these analyses were used to find genes that had significantly different expression between treatment groups. Hierarchical clustering and support tree analyses of differentially expressed genes were done using TIGR multi-experiment viewer software (Saeed et al., 2003). Gene ontology of the genes shown in Tables 3 and 4 were obtained using DAVID (Dennis et al., 2003) and LocusLink (http://www.ncbi.nlm.nih.gov/LocusLink/) from the National Center for Biotechnology Information [LocusLink has been superceded by Entrez Gene (http://www.ncbi.nih.gov/entrez/query.fcgi?db=gene)].
|
|
|
Liver RNA extracted from E+ and E– mice was used in these experiments. Genes for qPCR were chosen based on the gene lists obtained using ANOVA, a nonparametric Wilcoxon test, the P values of genes, and the false discovery rate of genes obtained from SAM analysis. Of 11 gene expression profiles measured using qPCR, 5 were done using Taqman probes and the remainder by SYBR green assay.
From each mouse, 10 µg of total RNA was reverse transcribed using Stratascript RT (Stratagene, La Jolla, CA) with oligo dT and random hexamer primers. Then, 1.25 ng of cDNA was added to a 25-µL PCR reaction to get a final concentration of 0.05 ng/µL of cDNA. For the Taqman assay, final concentrations of the forward and reverse primers were 300 nM, and the probe concentration was 200 nM. For the SYBR green assay, final concentrations of the forward and reverse primers were 100 nM. The list of forward and reverse primer sequences is shown in Table 2. Reactions were performed by using the Brilliant QPCR Plus Core Reagent kit, and Brilliant SYBR Green QPCR Master Mix for Taqman and SYBR Green assays, respectively (Stratagene, La Jolla, CA). Primers were designed using Primer Express (Applied Biosystems, Foster City, CA), with an annealing temperature of 60°C and amplification size of less than 150 bp. Glyceraldehyde-3-phosphate dehydrogenase (GAPD) and 18S ribosomal (Rn18s) genes were each evaluated for use as an endogenous control by comparing their expression across 4 different E+ and E– samples at 3 different template concentrations per sample and selecting the gene that had the lower SD and CV. Glyceraldehyde-3-phosphate dehydrogenase was chosen as the endogenous control gene in our experiments as it had a lower SD (0.47 to 0.80) and lower CV (0.02 to 0.04) across different samples and template concentrations compared with the Rn18s gene (SD, 1.01 to 1.40; and CV, 0.06 to 0.08).
|
CT method (user manual #2, ABI Prism 7700 SDS) and qPCR was done in an ABI prism 7500 sequence detection system (Applied Biosystems, Foster City, CA). In the 2–CT analysis, the threshold cycle (CT) from E– mice was used as a calibrator sample. The threshold cycle was determined manually for each gene across all samples such that it fell in the exponential phase of PCR (user manual #2; ABI Prism 7700 SDS). Dependent variables, CT, and delta CT values for the individual genes in E+ and E– mice were fitted to a model including effects of sex, endophyte, and their interactions using GLM of SAS. |
RESULTS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Mice fed the E+ diet gained less weight (P < 0.01) than those fed E– from d 1 to 7 of the experiment (0.13 and 0.85 ± 0.11 g for E+ and E–, respectively). Mice consuming E+ gained 0.78 g from d 7 to 14 compared with 1.01 (± 0.11) g for mice consuming E– (P = 0.16). Overall, BW gain from d 1 to 14 on the E+ diet (0.91 ± 0.15 g) was less (P < 0.01) than on the E– diet (1.85 ± 0.16).
Microarray Experiments
Microarray analyses were conducted to find differential gene expression between E+ or E– fed mice. These analyses revealed 9 and 13 differentially expressed genes using ANOVA with equal and unequal group variances, respectively, using P 
0.02 between treatment groups as the cutoff for detecting differential expression. Significance analysis of microarrays done on (GAn)gn values obtained from the second stage ANOVA model identified 9 genes as downregulated in E+ mice compared with E– mice at a false discovery rate of less than 40%. Genes identified using ANOVA with equal, unequal group variance and SAM were compared to identify differentially expressed genes (Table 1
). Microarray analyses, using ANOVA with unequal group variances, identified 36 genes as differentially expressed when P 0.05 was used as a cutoff for significance between treatment groups. The E+ diet resulted in upregulation of 13 genes, whereas 23 genes were downregulated (Tables 3 and 4). The hierarchical clustering of these 36 genes was able to separate the animals into 2 groups that precisely reflect the E+ or E– treatments with 100% support from the bootstrap test. The genes were also separated into 2 groups of upregulated or downregulated genes with 100% support (Figure 1).
|
Quantitative Real-Time PCR
Expression of 11 genes were measured using qPCR in 14 E+ and 12 E– mice liver samples. Among these samples, only 6 E+ and 7 E– mice were used to generate microarray data. Nine of the 11 genes were present on the microarray and qPCR results of 7 of these genes were in agreement with microarray results in terms of direction of change and magnitude. Four of the 7 genes (Atpb5, Hsd17b2, Hsd3b5, expressed sequence AU018778) had concordance in terms of statistical significance obtained using a 2-way Student’s t-test (P < 0.05). When Scd1 expression was measured by qPCR on the samples hybridized to the array (n = 13), both methods agreed on the direction of change (34% upregulation) in E+ samples. However, when Scd1 expression was measured on all animals (n = 26), it showed that Scd1 was downregulated in the E+ group by 38% compared with the control (P < 0.05). For genes Abcb4 and mt-Atp8, qPCR had the same direction of fold change as in the microarray experiment (Abcb4, downregulation in E+ group with microarray = 29%, qPCR = 13%; mt-Atp8, downregulation in E+ group with microarray = 23%, qPCR = 7%), but differences were not detected between treatment groups (P = 0.70, P = 0.41, for Abcb4 and mt-Atp8, respectively). Two of the genes (IL-6 and Rbbp7) showed gene expression changes in the opposite direction to that of microarray results (Figure 2). One of the genes, IL-6, was expressed at very low levels and amplimer was not detected until cycle 33, even when the template concentration was increased to 1 ng/µL. There was an effect of sex for gene Hsd3b5, with 81% downregulation in females compared with males, whereas gene Hsd17b2 was up-regulated in female mice by 45% (P < 0.01) compared with males. There was an interaction of sex and diet for Hsd3b5, where E+ resulted in 93% reduction in Hsd3b5 expression in females when compared with male mice fed E+ diet (P < 0.01). There was no interaction of sex and diet (P = 0.10) for gene Hsd17b2. The greatest difference in gene expression detected by qPCR was for Hsd3b5, which was 86% downregulated in E+ fed animals. Expression of 2 genes in which probes were not present on the arrays was analyzed using qPCR. The expression of the steroidogenic acute regulatory protein (Star) gene, which controls transport of cholesterol into mitochondria for conversion into different kinds of steroids, was not different between the E+ and E– groups. Expression of Hsd3b3 was downregulated due to the E+ diet by 50% compared with E– control mice (P = 0.01). Means and SE of cycle threshold obtained from qPCR between E+ and E– mice along with the P-values are shown in Table 5.
|
|
|
DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Carbohydrate Metabolism
Janssen et al. (2000) described metabolic changes in a dietary subacute toxicity experiment with the ergot alkaloid 
-ergocryptine in Sprague-Dawley rats. Total plasma cholesterol and high-density lipoprotein (HDL) cholesterol were decreased in a dose-dependent manner in females, and both triglyceride and glucose concentrations were decreased in the greatest dose groups in both sexes. The ergot alkaloid -ergocryptine also influenced carbohydrate metabolism in their study, with increases in insulin, glucagon, and observed glycogen storage in liver. The phosphorylase kinase, gamma 2 (testis/liver), codes for phosphorylase kinase. This important regulatory enzyme of glycogen metabolism was downregulated in E+ mice by 34% compared with E– mice (Table 4). Mutations and deficiency of this gene cause increased glycogen storage in liver of human and rat, and is one of the conditions seen in rats fed the ergopeptine alkaloids, ergocryptine (Janssen et al., 2000) or ergometrine (Peters-Volleberg et al., 1996). Thus, downregulation of the phosphorylase kinase, gamma 2 (testis/liver) could possibly induce some of the carbohydrate metabolism perturbations after ingestion of ergopeptine alkaloids.
Cholesterol Trafficking
The expression of ATP synthase H+ transporting gene (Atp5b) was increased by 100% (qPCR) in mice fed the E+ diet when compared with the control. The beta chain of Atp5b is present on the ectopic surface of the hepatocyte cell and can act as a receptor subject to stimulation by apoA I, and triggers endocytosis of HDL particles by an energy dependent process (Martinez et al., 2003). In liver, HDL endocytosis by the hepatocytes is the main transport mechanism of cholesterol import from peripheral tissues. Cholesterol concentration in serum is also known to decrease in animals fed endophyte-infected fescue (Bond et al., 1984; Stuedemann et al., 1985; Nihsen et al., 2004). The upregulation of Atp5b might be part of a feed back mechanism in the hepatocytes. The cells, sensing a decreased serum level of cholesterol, might be trying to compensate by increasing HDL endocytosis through an increase in HDL receptor protein expression. The Paraoxinase 1 (PON1) gene was downre gulated by 20% in the E+ when compared with E– mice. The protein PON1 is known to be quantitatively associated with HDL. Furthermore, overexpressing PON1 in transgenic mice protected HDL integrity and function (Oda et al., 2002).
The expression of the stearoyl-CoA desaturase 1 (Scd1) gene was downregulated due to E+ diet by 38% when measured in all animals (n = 26) by qPCR (Figure 2 and Table 5). The Scd1 catalyzes a rate-limiting step in the synthesis of monounsaturated fatty acids and plays an important role in fat metabolism (Ntambi, 1999). It is 1 of 3 genes induced more than 1.5-fold in response to oxidative stress and heat shock in different human cell types (Murray et al., 2004). This gene is also heavily regulated by the diet. For example, cholesterol and polyunsaturated fatty acids up-regulate Scd1, whereas a diet rich in carbohydrate downregulate it (Ntambi, 1999; Ntambi and Miyazaki, 2004). In liver, the majority of fatty acid metabolism is directed toward triglyceride synthesis and exocytosis. Changes in the expression of Scd1 are thus very likely to affect the fatty acid composition of the cells. This gene is needed for the synthesis of oleic acid, itself required for the esterification of cholesterol in the liver. Esterified cholesterol is then ready for export to the peripheral tissues in the form of very low-density lipoprotein. A decrease in Scd1 expression would likely translate into a decrease in very low-density lipoprotein synthesis and cholesterol export out of liver. Therefore, it seems that mice fed endophyte-infected fescue are trying to keep cholesterol inside the liver and limit its availability to the peripheral tissues.
The aldehyde dehydrogenase family 1, subfamily A1 and A2 genes were upregulated by 70 to 100% in E+ mice and these genes are involved in aldehyde dehydrogenase activity, bile acid synthesis, fatty acid, and glycerolipid metabolism. The ATP-binding cassette subfamily B (MDR/TAP) member 4 (Abcb4) and gene expressed sequence AU018778, which belongs to type-B carboxylesterase/lipase family, are involved in lipid transport and metabolism, respectively. The Abcb4 gene was downregulated by 29%, whereas AU018778 was downregulated by 35% due to the E+ diet when compared with the control, indicating cholesterol and lipid metabolism changes in the liver.
The disagreement of results obtained from microarray and qPCR for genes IL-6, Rbbp7 could be due to low levels of expression of these genes. Array probes with low intensities have been shown to yield greater technical variance, thus increasing the chance of being mistakenly identified as differentially expressed (Quackenbush, 2002). Thus, the low expression of IL-6 could yield a false microarray result. Another reason for the discrepancy between microarray and qPCR results for IL-6 and Rbbp7 is that each technique may detect different splicing variants. Because the sequences of the oligos on the array are not known, we cannot rule out this possibility.
Sex Steroid Hormone Metabolism
Expression of 3 of the genes affected by the E+ diet play a role in sex steroid hormone metabolism. A flowchart showing the gene expression changes in the sex steroid metabolism pathway is shown in Figure 3. The Hsd3b5 gene product catalyzes the formation of the relatively inactive steroid androstanediol from active dihydrotes-tosterone (DHT), and was reported to have sex-specific expression only in male mice (Abbaszade et al., 1995). In the current study, expression of this gene was also detected in female mice and was downregulated in E+ mice by 49% compared with E– treated mice. This would reduce the conversion of testosterone and DHT to the less active androstanediol. The steroid activating isoform, Hsd3b3, which catalyzes the reverse reaction of Hsd3b5/b4, was also downregulated in E+ mice by 50% compared with E– mice. This enzyme is essential to sex steroid hormone synthesis, as it catalyzes production of progesterone from pregnenolone, androstenedione from dehydroepiandrosterone and testosterone from androstenediol (Wong and Sarjeet, 2002). It is interesting to note that earlier reports indicate that consumption of an E+ diet caused a decrease in circulating progesterone levels (Estienne et al., 1990). The third gene, Hsd17b2, was downregulated in E+ mice by 34% compared with E–treated mice. This enzyme catalyzes conversion of the highly active 17B-hydroxyl forms of sex steroids into less potent 17-keto steroids such as estradiol into estrone, testosterone into androstenedione, and DHT into 5-androstenedione, respectively. This enzyme is likely involved in the rapid degradation and excretion of steroids in surface epithelial cells and hepatocytes in the intestine and liver, respectively (Mustonen et al., 1998).
|
; Dalla Valle et al., 2002) or mice (Choudhary et al., 2003). Furthermore, the P459arom gene is expressed in bone, brain, vascular tissue, adipose tissue, skin, and fetal liver of humans (Bulun et al., 2004), and in liver, pituitary, muscle, kidney, and adipose tissues of sheep (Sharma et al., 2004). The function of these proteins in these tissues is unknown but is likely important to the animals, especially as related to the conservation of expression throughout related species.
We do not know if metabolic intermediates such as pregnenolone or dehydroepiandrosterone are likely to accumulate because we did not measure expression of P450scc and CYP17 genes. These results agree with previous studies indicating that cattle fed endophyte-infected fescue had reduced levels of estradiol, progesterone, and LH in circulation, and in CL (Browning et al., 1998; Burke et al., 2001).
Other Differentially Expressed Genes
The probasin gene was found to be downregulated due to the E+ diet. This protein belongs to the lipocalpin family and expression of this gene is androgen-regulated (Rennie et al., 1993). Two genes, small inducible cytokine A6 and complement component C6 were differentially expressed and are involved in chemokine activity, immune response, and complement activation. Genes involved in apoptosis, like Caspase 6, and structural protein like crystalline-mu, were differentially expressed. The "structural maintenance of chromosome 1-like 1 protein (SMC-protein)" gene was downregulated in E+ mice. This gene is involved in chromosome cohesion during the cell cycle and in DNA repair and has chromatin, ATP binding and ATP-binding cassette transporter activity.
|
Footnotes |
|---|
2 Corresponding author: AntoniouE{at}missouri.edu
Received for publication March 18, 2005. Accepted for publication January 5, 2006.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. R. Brown, G. A. Anderson, K. Son, G. Rentfrow, L. P. Bush, J. L. Klotz, J. R. Strickland, J. A. Boling, and J. C. Matthews Growing steers grazing high versus low endophyte (Neotyphodium coenophialum)-infected tall fescue have reduced serum enzymes, increased hepatic glucogenic enzymes, and reduced liver and carcass mass J Anim Sci, February 1, 2009; 87(2): 748 - 760. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. de Haan, R. Wehrens, S. Bauerschmidt, E. Piek, R. C. v. Schaik, and L. M. C. Buydens Interpretation of ANOVA models for microarray data using PCA Bioinformatics, January 15, 2007; 23(2): 184 - 190. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Bhusari, Z. Liu, L. B. Hearne, D. E. Spiers, W. R. Lamberson, and E. Antoniou Expression Profiling of Heat Stress Effects on Mice Fed Ergot Alkaloids Toxicol. Sci., January 1, 2007; 95(1): 89 - 97. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
3') used in real-time PCR (accession numbers are listed in Tables 3
