Transcriptomic markers meet the real world: finding diagnostic signatures of corticosteroid treatment in commercial beef samples
© Pegolo et al.; licensee BioMed Central Ltd. 2012
Received: 3 August 2012
Accepted: 16 October 2012
Published: 30 October 2012
The use of growth-promoters in beef cattle, despite the EU ban, remains a frequent practice. The use of transcriptomic markers has already proposed to identify indirect evidence of anabolic hormone treatment. So far, such approach has been tested in experimentally treated animals. Here, for the first time commercial samples were analyzed.
Quantitative determination of Dexamethasone (DEX) residues in the urine collected at the slaughterhouse was performed by Liquid Chromatography-Mass Spectrometry (LC-MS). DNA-microarray technology was used to obtain transcriptomic profiles of skeletal muscle in commercial samples and negative controls. LC-MS confirmed the presence of low level of DEX residues in the urine of the commercial samples suspect for histological classification. Principal Component Analysis (PCA) on microarray data identified two clusters of samples. One cluster included negative controls and a subset of commercial samples, while a second cluster included part of the specimens collected at the slaughterhouse together with positives for corticosteroid treatment based on thymus histology and LC-MS. Functional analysis of the differentially expressed genes (3961) between the two groups provided further evidence that animals clustering with positive samples might have been treated with corticosteroids. These suspect samples could be reliably classified with a specific classification tool (Prediction Analysis of Microarray) using just two genes.
Despite broad variation observed in gene expression profiles, the present study showed that DNA-microarrays can be used to find transcriptomic signatures of putative anabolic treatments and that gene expression markers could represent a useful screening tool.
The use of growth promoters in meat production has been banned in the European Union since 1988 due to the potentially adverse effects of hormone residues for the consumer. Council Directive 23/96/EC requires the EU member States to adopt National Monitoring Plans to control the illegal use of these compounds. Despite the ban, these substances are still administered and a black-market for the production, distribution, and use of multiple steroids has flourished . To elude official controls, new anabolic compounds are developed and growth promoters are administered at low doses or combining different substances in hormone cocktails. To keep pace with such a moving target, constant innovation in screening and validation methods is necessary.
Anabolic steroids act on multiple organs and metabolic pathways either through primary interaction or secondary effects. For this reason, indirect approaches, based on the evaluation of perturbations of different biological systems, have been proposed to identify growth-promoter-treated animals. Target organ histology, transcriptomics, proteomics, and metabolomics have been explored as screening tools to better inform confirmative analysis [2–5]. The application of transcriptomics in toxicology has experienced an impressive growth in recent years, leading to the foundation of a new discipline, toxicogenomics . Although toxicogenomics has mostly focused on the effects of pharmacological compounds in model species, it is increasingly applied to monitor the effects of xenobiotics in non-model species (e.g. eco-toxicogenomics). In this contest, gene expression profiles have been used to obtain indirect biomarkers for the use of growth promoters in beef meat production. Initially, quantitative real-time PCR was used to analyze the expression of candidate diagnostic genes. This approach has already been successfully applied in several studies on experimentally treated animals [7–12]. However, good candidate genes are often difficult to identify and the use of single or few gene markers provides a limited and biased view of the biological response to xenobiotics. Using either DNA microarray platforms or RNA-sequencing, it is possible to obtain whole-transcriptome expression profiles, which provide a broad and unbiased (hypothesis-free) picture of the biological response to toxicants. Thanks to the decreasing costs of genomic technologies, transcriptomics is starting to be applied to the identification of gene markers for anabolic treatments in beef cattle. Microarray analysis has been used so far to examine the effects of anabolic hormones in experimentally treated animals, as in the case of skeletal muscle samples from bulls administered with Dexamethasone (DEX) and DEX plus 17β-estradiol , and with trenbolone acetate plus estradiol , or livers from beef cattle after experimental treatment with dehydroepiandrosterone . More recently, the potentiality of RNA- sequencing technology for the detection of growth-promoters abuse in cattle was explored and successfully used to screen for highly regulated genes to be proposed as biomarker candidates for detecting the treatment with trenbolone acetate plus estradiol .
In the present study, the use of transcriptomics is extended for the first time to unknown skeletal muscle samples, collected directly from beef cattle immediately after slaughtering. These samples are part of a larger survey during which also thymus and urine specimens were collected with the aim of detecting putative use of corticosteroids in commercial animals. Histological analyses on thymic samples were used following a validated protocol to classify samples as positive, suspect or negative for corticosteroid treatment and results have been reported previously , showing a relevant percentage of suspect animals for corticosteroid treatment compared to official data. Here, detection of DEX urinary excretion by Liquid Chromatography - Mass Spectrometry (LC-MS) was additionally implemented to provide direct evidence of illicit use of anabolic steroids. All samples (LC-MS positive and negative samples, positive, suspect or negative from histological analysis) were analyzed using a cattle-specific oligo-DNA microarray and compared with known negative controls. The goal of this study was to identify gene expression patterns that could classify unknown commercial samples as negative or putative positive (suspect) by comparison with untreated controls.
Confirmation data of DEX in spiked urine analyzed by LC-MS 3
LC-MS 3 validation
spike level (ng/ml)
repeatibility (R.S.D., %)
linearity (r2, 0.20-10 ng/ml)
LC-MS 3 analysis for dexamethasone (ng/ml)
Microarray experiments for all the 43 bovine skeletal muscle samples were performed.
Spearman’s rho for the set of selected genes used for qRT-PCR validation
Based on these results, the observed dispersion of negative controls is likely to be attributable to the inclusion of Holstein animals, which is a rather different breed than the French breeds or their crosses that are prevalent in the study. However, breed did not explain the separation of all other samples. Two additional variables that are known to largely affect gene expression in the skeletal muscle, gender and muscle type , could be excluded as they were identical for all samples by experimental design.
To avoid possible distortion in the PCA plot due to breed effects, Holstein samples were excluded from the subsequent analyses and the PCA was repeated (Figure 3B). Here, the first two components, which account for 70.9% of the total variance, identified two groups. The x-axis, which explained 56.2% of variance, separated group A, which included the negative controls and some unknown animals, from group B, which comprised all four positive samples and other unknown samples.
To evaluate the statistical significance of the observed clustering, microarray data for the group B were compared to the group A in a two-class unpaired data SAM test. Even enforcing a stringent false discovery rate (FDR, 0%) and a conservative fold-change (FC) threshold (2-fold), a large number of differentially expressed genes (DEG) was obtained, with 2,351 up-regulated and 1,610 down-regulated transcripts (see Additional file 1 and Additional file 2), suggesting a highly significant difference in expression profiles between group A and group B in Figure 3B.
GO Biological Process, GO Molecular functions and KEGG pathways analysis of differentially regulated genes
G-protein coupled receptor protein signaling pathway
Cell surface receptor linked signal transduction
Regulation of transcription, DNA-dependent
Regulation of RNA metabolic process
Pattern specification process
Anterior/posterior pattern formation
tRNA wobble uridine modification
tRNA wobble base modification
Positive regulation of organelle organization
Intrinsic to membrane
Integral to membrane
Contractile fibre part
Intermediate filament cytoskeleton
Transcription factor complex
Olfactory receptor activity
Calcium ion binding
Passive transmembrane transporter activity
Substrate specific channel activity
Ion channel activity
Transcription factor activity
Several sarcomere proteins were differentially regulated as well. High doses of DEX were reported to lead to muscle atrophy, mostly affecting fast twitching fibres . A recent study by Stella and colleagues , comparing protein expression profiles between beef cattle treated with DEX, alone or in association with clenbuterol, using 2D protein gel electrophoresis and mass spectrometry confirmed that the administration of DEX favors a slow fiber phenotype. Stella and co-workers reported that fast fiber specific proteins (e.g. myosin light chain 1, myosin light regulatory chain 2, and isoform 3 of troponin T) and glycolytic enzymes were significantly under-expressed in DEX-treated animals, while myosin light chain 6B (MYL6B) alkali smooth muscle and non-muscle and isoform 1 of troponin T (TNNT1) were significantly over-expressed. In agreement with such evidence, transcripts encoding TNNT1, MYL6B, and myosin binding protein C slow type were up-regulated in group B animals of the present work while three glycolytic enzymes were significantly under-expressed.
Different myosin isoforms (18B, 7B, 16, 9B), myosin heavy chain 7 (MYH7), 15 (MYH15), 4 (MYH4), myosin regulatory light chain interacting protein (MYLIP) and myosin light chain kinase (MYLK) were up-regulated in group B animals, supporting previous studies showing that glucocorticoids can promote myogenic repair and myoblast proliferation. In particular, while high doses of glucocorticoids in vitro impair C2C12 myoblast proliferation rate and differentiation capacity, lower doses increase the myogenic fusion efficiency of C2C12 cells .
Also ion channel activity was found to be modified between the two groups of animals. In particular, several genes involved in calcium signaling pathways were up-regulated. It has been reported that Ca2+-dependent calcineurin signaling mediates skeletal muscle hypertrophy upon stimulation with IGF1 or insulin associated with DEX . Calcineurin is a calcium-activated protein phosphatase which, upon activation, transduces signal by removing specific phophorylation of the cytoplasmic transcription factor nuclear factor of activated T-cells (NFAT). This allows NFAT translocation to the nucleus, where it activates the transcription of IL-4, a cytokine that has a key role in autocrine/paracrine control of mammalian muscle growth . In fact, IL-4 and IL-4 receptor were found to be over-expressed in group B animals sampled at the slaughterhouse as well as in DEX-treated bulls . Muscle cell membrane depolarization due to K+ efflux has also been reported to activate calcineurin-mediated transcriptional responses . Intriguingly, several potassium voltage-gated channels were over-expressed in group B animals.
The putative similarity in the transcriptomic response in the skeletal muscle of beef cattle after low-dose administration of DEX  and in individuals clustering within group B in the present study is statistically relevant. Overall, 206 genes were up-regulated in both sets of samples, showing a highly significant concordance (Fisher Exact test, p < 0.0001). A few “common” transcripts are worth mentioning. Colony stimulating factor 3 receptor (CSF3R) is part of the JAK1-STAT1-STAT3 pathway, which was reported to induce myoblast proliferation . Protein phosphatase 2 (PP2), whose regulatory subunit PP2R1A and catalytic subunit isoform PPP2CA were up-regulated, has an activating role in Wnt signaling, a pathway that was shown to be associated with satellite cell proliferation during muscle regeneration . Actin-related protein 2/3 complex subunit 1A (ARPC1A) is part of a complex that has a critical role in myoblast fusion during either muscle development or regeneration . Elevated expression of Delta-like 1 (DLK1) contributed to hypertrophy in callipyge sheep skeletal muscle , while experimental over-expression of DLK1 in mouse skeletal muscle induced hypertrophy . The highly significant concordance with gene expression modifications reported in DEX-treated cattle suggests that all animals of group B might have been illegally administered with DEX. This can explain, at least in part, the observed marked difference in gene expression profiles between group A and group B.
Two main findings were observed in the present study. First, broad variation in gene expression profiles exists across both known and unknown samples. The causes of such variation are not completely clear, although breed appeared to be the most important. This observation reinforces the obvious, yet often overlooked, need for a most comprehensive representation of biological variation, when indirect biomarkers are applied outside controlled experimental settings. Second, after removing the main source of variation, unsupervised analysis of gene expression profiles showed a highly significant distinction between two groups, one including positive controls and a sub set of commercial samples, the other comprising all negative controls and the remaining unknown individuals. The observed separation was confirmed by a two-class SAM test that identified over 3,900 differentially expressed genes and a class prediction approach that was able to discriminate between the two groups using just two genes, using as a training set positive and negative controls and as test set all unknown samples. Functional annotation of up- and down-regulated transcripts showed several biological processes and molecular pathways that have been already reported in previous proteomic and transcriptomic studies to be altered upon controlled administration of low dosage corticosteroids. Such evidence seems to suggest that unknown animals clustering with LC-MS positive samples might have been administered with glucocorticoids as well. It cannot be completely excluded that all unknown samples in group B showed a transcriptomic profile similar to positive controls for other reasons than being treated with anabolic hormones. On the other hand, such explanation appears quite unlikely as both technical issues (e.g. systematic bias in microarray data) and biological variables known to affect gene expression profiles have been carefully controlled. Therefore, under the working hypothesis that group B samples should considered as suspect, how such hypothesis could be reconciled with the more limited evidence of positive animals (LC-MS) or suspects (histology)? With regard to LC-MS results, one possible explanation is related to the fact that DEX or other corticosteroids are often illegally administered as diluted solution spread on feed and differences in drug uptake can be evidenced across treated animals. Such very low dosages and the rapid metabolism and excretion of these substances make determination of residues extremely difficult , even by using targeted GC/MS and LC/MS/MS methods [32–39]. The urinary parent compound and its metabolites concentration are very low during all the treatment time and completely absent a short time after the interruption of the treatment , while biological effects, including transcriptomic responses, might still be detectable as such effects are caused by anabolic compounds and often persist after the hormone has been eliminated. It is also well-known that corticosteroids are administered as cocktails of different chemical species with similar biological action to reduce the amount of single compounds. This could make LC-MS less sensitive. In particular, when targeted LC-MS analyses are carried out, for instance searching for DEX, sensitivity is decreased when DEX is not the major component of the cocktail or is entirely substituted with other corticosteroids. With regard to the results of histological analyses, it should be noted that while only four samples could be reliable considered as putative positive, others were considered dubious, including some those included in group B. In fact, the efficacy of the histological method has been recently challenged because of the lack of appropriate reference material considering the evolving nature of animal-rearing practices . Furthermore, administration of corticosteroids leads to thymus cortical atrophy and ‘starry sky’ appearance. However, age-associated thymic involution was evidenced in cattle as in other mammals. As the animals included in the present study were mostly all over 12 months old and most around 15–18 months of age, several positive cases might have been considered negative or dubious because natural thymic involution could not be distinguished from the effects of corticosteroid treatment. On the other hand, the biological effects of corticosteroids on gene expression data were proved to persist several days after time of withdrawal [11, 12, 42] and thus may evidence the anabolic treatments even when the active compound and its metabolites have been excreted and they are no longer detectable, as mentioned above.
If the working hypothesis of all animals in group B being putative positive were correct, the number of suspect cases would be quite impressive. However, the actual importance of growth-promoter abuse in beef cattle is known to be underestimated  because highly informative analytical methods are used on just a small number of cases and often such methods are not sensitive enough to detect residues of low dosage cocktails. This prompts for the development and use of more reliable and cost effective screening tools.
The present work showed that it might be possible to use just a few gene markers for highly reliable sample classification. However, global transcriptomic tools (DNA microarrays, RNA-seq) are becoming increasingly affordable and rapid, and soon a whole-transcriptome analysis will be feasible in routine practice. Such an approach avoids a priori selection of candidate markers, allows the identification of complex transcriptomic signatures through functional annotation and enables the comparison with other -omic and analytical/chemical studies. This will lead toward meta-analysis of data from controlled experiments as well as field studies and in turn, to the identification of indirect biomarkers less influenced by unknown variables.
Animals and sampling
Tissue samples from a specific anterior limb muscle (Biceps brachii) were collected, immediately after slaughtering, from 25 beef cattle randomly selected in 10 different batches and stored in 2 mL RNAlater solution (Ambion, Monza, Italy) at −20°C until extraction. Urine samples were collected during slaughtering from the bladder and each sample, without any preservative, was divided in 15 mL aliquots and stored at −20°C until the analysis. Sample collection was part of a monitoring program on anabolic treatment targeted on indirect biomarkers. The sampling was managed by the Regional Veterinary authorities (Veneto, Italy) and carried out by trained veterinarians. The animals were vaccinated against the main respiratory diseases and treated regularly against parasites typical of the species. They were healthy at ante-mortem inspection, passed post-mortem inspection, and their meat was approved for human consumption. All individuals were male, their age ranged between 12 and 24 months, and weight between 300–500 kg. The majority of the commercial samples were Charolais, Limousine, or mixed breed (Charolais×Limousine).
Two additional sets of samples from previous controlled experiments were included as negative controls: one set included 5 samples from mixed-breed (Charolais×Limousine) 18 months old male bulls, ~450 kg mean body weight, while the second one included 13 samples from male Holstein beef cattle between 13 and 18 months old, 500–670 kg body weight. The control animals came from experiments conducted according to the guidelines of Italian law (DL 116/92) and European legislation (11 and subsequent amendments) for care and use of experimental animals, and the studies were approved by the Italian Ministry of Health ethical committee.
All solvents for LC-MS were HPLC or analytical grade and purchased from Carlo Erba Reagenti (Milan, Italy). The water used was purified using a Milli-Q system (Millipore, Bedford, MA, USA). Sodium acetate trihydrate and the enzyme for phase II metabolite deconjugation (Helix pomatia preparation) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Dexamethasone was purchased from Steraloids (Newport, RI, USA). Cortisol-d4 (internal standard) was obtained from CDN Isotopes (Pointe-Claire, Quebec, Canada).
Stock solutions of reference standard (1 mg/mL) were prepared in methanol; working solution (1 μg/mL and 0.1 μg/mL) were prepared, monthly, by successive tenfold dilution with methanol and stored in the dark at −20°C. Standard curves were prepared daily by dilution of working solution in water–methanol-formic acid (50:50:0.1, v:v:v) to obtain seven concentration levels ranging from 0.1 to 25 ng/mL.
After centrifugation, a 5 mL aliquot of urine was spiked with 25 μL of the internal standard solution (1 μg/mL). The pH was adjusted to approximately 5 by adding 15 ml of 1 M pH 5 acetate buffer, then 50 μL of Helix pomatia juice was added to the sample and enzymatic hydrolysis was carried out overnight at 40°C. The preparation was repeated on two separated aliquots, including or omitting the Helix pomatia preparation addition and incubation.
Subsequently, 16 mL of the sample was applied to Strata C18-U SPE cartridges (500 mg, Phenomenex, Bologna, Italy) previously activated with 5 mL of methanol and 5 mL of water. After washing with 5 mL of water and 10 mL of methanol–water (20:80, v/v), analytes were eluted with 2 ml of methanol. Then, the extract was applied to Strata NH2 cartridges (100 mg, Phenomenex, Bologna, Italy), previously activated with 1 mL of methanol. Purified extracts were evaporated and reconstituted in 500 μL of water–methanol-formic acid (50:50:0.1, v:v:v).
An Accela 600 HPLC pump with CTC automatic injector was used (Thermo Fischer Scientific, San Jose, CA, USA). Reversed-phase liquid chromatography was performed on a 100 × 2.1 mm i.d., 1.9-μm Thermo Hypersil Gold column. The mobile phases consisted of (A) 0.1% formic acid (v/v) in water and (B) 0.1% formic acid in methanol. The mobile phase composition (A:B; v/v) was: 90:10 at 0 min, 10:90 at 21 min, and 90:10 from 22 to 25 min; the flow rate was set at 200 μL/min. The sample trays was maintained at 4°C. A sample volume of 10 μL was injected.
Mass spectrometric analysis was performed on a LTQ XL ion trap (Thermo Fischer Scientific, San Jose, CA, USA), equipped with a heated electrospray ionization (HESI-II) probe operating in negative ion mode with the following condition: sheath and auxiliary gas (nitrogen) flow 30 and 10 arbitrary units, respectively; ion spray voltage 2 kV; capillary temperature 275°C; capillary voltage −13 V; tube lens −68 V. Helium was used for collision-induced dissociation.
Molecular weights and diagnostic ions of the investigated corticosteroids
P.I. (m/z) MS2
Product ion (m/z) MS3
292,307 b,325,345 a
Xcalibur (version 2.1) data acquisition software from Thermo was used.
Calibration and quantification
Calibration lines were constructed using pooled urine obtained from samples collected during several experimental plans from bovine of different age and breed.
Pooled urine samples, with no residues of DEX, were spiked with 2.5 ng/mL of IS and with DEX to obtain a concentration range of 0.2-10 ng/mL.
Quantification was based on peak area ratios of the analyte to the IS and a least-squares linear regression analysis was performed to calculate calibration curves.
The validation study was carried out at three concentration levels by the analysis of pooled urine samples spiked with 0.25, 0.50 and 2.5 ng/ml of DEX. To each sample 2.5 ng/mL of the I.S. cortisol-d4 was added, and six replicates of each sample were analyzed on three different days to evaluate the recovery (internal standard corrected) and the precision in term of repeatability (within day) and within laboratory reproducibility (different operators and environmental conditions).
For DEX, the limit of detection (LOD) was calculated as the mean plus three times the standard deviation of the signal-to-noise ratio of 20 representative blank urine samples; the limit of quantification (LOQ) was calculated as the mean plus ten times the standard deviation of the signal-to-noise ratio of 20 representative blank urine samples; the decision limit (CCα) and the detection capability (CCβ) were calculated according to the ISO standard 11843 .
Total RNA was extracted from 30 mg tissues using the RNeasy Mini kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The concentration of RNA samples will be measured using a UV–vis spectrophotometer NanoDrop ND-1000 (Nanodrop Technologies, Wilmington, DE) and RNA a was estimated running each sample on a RNA-chip in an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA). RNA integrity number (RIN) index was evaluated using the Agilent 2100 Expert software. RIN is a numerical assessment of the integrity of total eukaroyte RNA samples based on the entire electrophoretic trace of the RNA sample rather than the ratio of the ribosomal bands and allows to standardize the interpretation of RNA quality. In the present study a conservative threshold was enforced in order to reduce experimental biases due to poor RNA quality. To ensure optimal data quality, only RNA samples with RIN number ≥7.0 were included in the analysis as suggested by protocols for microarray experiments.
RNA amplification, labeling and hybridization
Sample labeling and hybridization were performed according to the Agilent One-Color Microarray-Based Gene Expression Analysis protocol. Briefly, for each sample 200 ng of total RNA were linearly amplified and labeled with Cy3-dCTP. A mixture of 10 different viral poly-adenilated RNAs (Agilent Spike-In Mix) was added to each RNA sample before amplification and labeling, to monitor microarray analysis work-flow. Labeled cRNA was purified with Qiagen RNeasy Mini Kit, and sample concentration and specific activity (pmol Cy3/μg cRNA) were measured in a NanoDrop ND-1000 spectrophotometer. A total of 1,650 ng of labeled cRNA was prepared for fragmentation adding 11 μL 10× Blocking Agent and 2.2 μL of 25× Fragmentation Buffer, heated at 60°C for 30 min, and finally diluted by addition with 55 μL 2× GE Hybridization buffer. A volume of 100 μL of hybridization solution was then dispensed in the gasket slide and assembled to the microarray slide (each slide containing four arrays). Bovine-specific oligo-arrays (Agilent Bovine-Four-Plex G2519F) were used. For most of the transcripts represented on this array, two identical probes are synthesized at two distinct positions on the slide, therefore the average value between the intensities of the two replicate probes was used. The slides were incubated for 17 h at 65°C in an Agilent Hybridization Oven, subsequently removed from the hybridization chamber, quickly submerged in GE Wash Buffer 1 to disassembly the slides and then washed in GE Wash Buffer 1 for approximately 1 minute followed by one additional wash in pre-warmed (37°C) GE Wash Buffer 2. Hybridized slides were scanned at 5 μm resolution using an Agilent G2565BA DNA microarray scanner. Default settings were modified to scan the same slide twice at two different sensitivity levels (XDR Hi 100% and XDR Lo 10%). Microarray data have been deposited in NCBI's Gene Expression Omnibus , and are accessible through GEO Series accession number GSE26318.
Normalization of microarray data
The two linked images generated from the scanned slide were analyzed together, data were extracted and background subtracted using the standard procedures contained in the Agilent Feature Extraction (FE) Software version 9.5.1. The Feature Extraction software returns a series of spot quality measures that enable to evaluate goodness and reliability of spot intensity estimates. Additionally, in each sample, spike-in viral poly-adenilated RNAs were added before sample processing to provide an internal quality control. Each spike-in RNA has a different known concentration following a dilution series and there are 32 replicate probes for each spike-in RNA on the array. All samples (negative, positive, and unknown samples) were then normalized together in a single run to avoid potential biases. After normalization, spike intensities are expected to be uniform across the experiments of a given dataset. Based on the comparison of spike-in probe signal between arrays after normalization, cyclic loess approach was chosen. All control features except for Spike-in (Spike-In Viral RNAs) were excluded from subsequent analyses. After normalization, a further quality control step was performed by removing all probes with intensity values lower than the second lowest spike-in concentration because this value was considered too close to the limit of detection. So, probes with intensity values <4 in at least 30 of the 43 samples were removed from the dataset. Filtering and normalization procedures were performed using R statistical software available at http://www.r-project.org/.
A set of 10 genes (see Additional file 3) was validated by real-time RT-PCR. This validation step provided independent validation of microarray data. For each selected target gene and for the reference gene (RS5), a (q)RT-PCR assay was designed. Gene–specific primers that encompass one intron were defined for each transcript (except for HOXA9 gene) using the program Primer Express version 2. To design intron-spanning primers, putative intron-exon boundaries were deduced from the Genome Browser Database (http://genome.ucsc.edu/cgibin/hgGateway?hgsid=105081852;clade=vertebrate;org=Cow;db=0).
One microgram of the total RNA for each sample was reverse transcribed to cDNA using Superscript II (Invitrogen, Milan, Italy). An aliquot (2.5 μL) of diluted (1:50 or 1:100) cDNA template was amplified in a final volume of 10 μL, containing 5 μL of KAPA SYBR® FAST Universal 2X qPCR Master Mix (Kapa Biosystems, Inc., Woburn, MA, USA), 0.25 μL of each gene specific primer (10 μM). The amplification protocol consisted of an initial step of 2 min at 95°C, followed by 45 cycles of 3 s at 95°C and 40 s at 60°C. All experiments were performed in a LightCycler 480 instrument (Roche Diagnostics, Milan, Italy). To evaluate the efficiency of each assay, standard curves were obtained amplifying two-fold serial dilutions of the same cDNA, which was used as calibrator. For each sample, the Crossing Point (Cp) was used to determine the relative amount of target gene; each measurement was made in duplicate and normalized to the reference gene RS5, which was measured in duplicate as well.
PCA on normalized and filtered gene expression data was carried out using the TMEV suite [48, 49]. Sample classification into two classes was based on visual inspection of PCA plots. A two-class non parametric test for unpaired data was implemented in the program SAM , to identify differentially expressed genes between classes, enforcing a FDR of 0% and a FC of 2. Sample class prediction was carried out using the program PAM , available online at http://www.stat.stanford.edu/~tibs/PAM. All other statistical tests (ANOVA, Spearman rank correlation test, Fisher exact test) were carried out in R using the RCommander GUI .
Enrichment analysis on differentially up- and down-regulated genes was performed using the Functional Annotation tool available in the DAVID Database (http://david.abcc.ncifcrf.gov/). GO terms and KEGG pathways included in the DAVID knowledgebase were considered. For KEGG terms, the following parameters were used: gene count 4, ease 0.05. For GO Biological Process, Cellular Component and Molecular Function BP_FAT, CC_FAT, MF_FAT respectively, with gene count 4, ease 0.05 were used.
The authors wish to thank Dr. Giandomenico Pozza, of the Istituto Zooprofilattico Sperimentale delle Venezia, Legnaro for his help in the collection of samples at commercial slaughterhouses. This work was financially supported in part by the Ministero delle Politiche Agricole e Forestali under project “SAFORISK” to C. M. and L. B. (D.M. 2089/09, 29th of January 2009), by a research grant of the Regione Veneto under project "Valutazione dell'affidabilità dei profili di espressione genica tessuto specifici quali biomarcatori indiretti di trattamento illecito nella filiera del bovino da carne" to C. M (Prot. N. 131062, 16th of March 2011) and from the University of Padua (CPDA077393 and CPDB090779).
- Pinel G, Rambaud L, Monteau F, Elliot CT, Le Bizec B: Estranediols profiling in calves' urine after 17 beta-nandrolone laureate ester administration. J Steroid Biochem Mol Biol. 2010, 121: 626-632. 10.1016/j.jsbmb.2010.01.012.View ArticlePubMedGoogle Scholar
- Mooney MH, Le Bizec B, Elliott CT: Combining biomarker screening and mass-spectrometric analysis to detect hormone abuse in cattle. Trends Anal Chem. 2009, 28: 665-675. 10.1016/j.trac.2009.03.011.View ArticleGoogle Scholar
- Riedmaier I, Becker C, Pfaffl MW, Meyer HH: The use of omic technologies for biomarker development to trace functions of anabolic agents. J Chromatogr A. 2009, 1216: 8192-8199. 10.1016/j.chroma.2009.01.094.View ArticlePubMedGoogle Scholar
- Pinel G, Weigel S, Antignac JP, Mooney MH, Elliott C, Nielen MWF, Le Bizec B: Targeted and untargeted profiling of biological fluids to screen for anabolic practices in cattle. Trends Anal Chem. 2010, 29: 1269-1280. 10.1016/j.trac.2010.06.010.View ArticleGoogle Scholar
- Lancova K, Dip R, Antignac JP, Le Bizec B, Elliott C, Naegeli H: Detection of hazardous food contaminants by transcriptomics fingerprinting. Trends Anal Chem. 2011, 30: 181-191. 10.1016/j.trac.2010.10.013.View ArticleGoogle Scholar
- Choudhuri S: Looking back to the future: From the development of the gene concept to toxicogenomics. Toxicol Mech Methods. 2009, 19: 263-277. 10.1080/15376510902855529.View ArticlePubMedGoogle Scholar
- De Maria R, Divari S, Spada F, Oggero C, Mulasso C, Maniscalco L, Cannizzo FT, Bianchi M, Barbarino G, Brina N, Biolatti B: Progesterone receptor gene expression in the accessory sex glands of veal calves. Vet Rec. 2010, 167: 291-296. 10.1136/vr.b4879.View ArticlePubMedGoogle Scholar
- Lopparelli RM, Zancanella V, Giantin M, Ravarotto L, Cozzi G, Montesissa C, Dacasto M: Constitutive expression of drug metabolizing enzymes and related transcription factors in cattle testis and their modulation by illicit steroids. Xenobiotica. 2010, 40: 670-680. 10.3109/00498254.2010.506562.View ArticlePubMedGoogle Scholar
- Reiter M, Walf VM, Christians A, Pfaffl MW, Meyer HH: Modification of mRNA expression after treatment with anabolic agents and the usefulness for gene expression-biomarkers. Anal Chim Acta. 2007, 86: 73-81.View ArticleGoogle Scholar
- Riedmaier I, Tichopad A, Reiter M, Pfaffl MW, Meyer HH: Identification of potential gene expression biomarkers for the surveillance of anabolic agents in bovine blood cells. Anal Chim Acta. 2009, 638: 106-113. 10.1016/j.aca.2009.02.014.View ArticlePubMedGoogle Scholar
- Toffolatti L, Rosa Gastaldo L, Patarnello T, Romualdi C, Merlanti R, Montesissa C, Poppi L, Castagnaro M, Bargelloni L: Expression analysis of androgen-responsive genes in the prostate of veal calves treated with anabolic hormones. Domest Anim Endocrinol. 2006, 30: 38-55. 10.1016/j.domaniend.2005.05.008.View ArticlePubMedGoogle Scholar
- Giantin M, Lopparelli RM, Zancanella V, Martin PG, Polizzi A, Gallina G, Gottardo F, Montesissa C, Ravarotto L, Pineau T, Dacasto M: Effects of illicit dexamethasone upon hepatic drug metabolizing enzymes and related transcription factors mRNAs and their potential use as biomarkers in cattle. J Agric Food Chem. 2010, 58: 1342-1349. 10.1021/jf9033317.View ArticlePubMedGoogle Scholar
- Carraro L, Ferraresso S, Cardazzo B, Romualdi C, Montesissa C, Gottardo F, Patarnello T, Castagnaro M, Bargelloni L: Expression profiling of skeletal muscle in young bulls treated with steroidal growth promoters. Physiol Genomics. 2009, 38: 138-148. 10.1152/physiolgenomics.00014.2009.View ArticlePubMedGoogle Scholar
- De Jager N, Hudson NJ, Reverter A, Wang YH, Nagaraj SH, Café LM, Greenwood PL, Barnard RT, Kongsuwan KP, Dalrymple BP: Chronic exposure to anabolic steroids induces the muscle expression of oxytocin and a more than fiftyfold increase in circulating oxytocin in cattle. Physiol Genomics. 2011, 43: 467-78. 10.1152/physiolgenomics.00226.2010.View ArticlePubMedGoogle Scholar
- Rijk JC, Peijnenburg AA, Hendriksen PJ, Van Hende JM, Groot MJ, Nielen MV: Feasibility of a liver transcriptomics approach to assess bovine treatment with the prohormone dehydroepiandrosterone (DHEA). BMC Vet Res. 2010, 6: 44-10.1186/1746-6148-6-44.PubMed CentralView ArticlePubMedGoogle Scholar
- Riedmaier I, Benes V, Blake J, Bretschneider N, Zinser C, Becker C, Meyer HHD, Pfaffl MW: RNA-Sequencing as useful screening tool in the combat against the misuse of anabolic agents. Anal Chem. 2012, 84: 6863-6868. 10.1021/ac301433d.View ArticlePubMedGoogle Scholar
- Vascellari M, Pozza G, Poppi L, Capello K, Angeletti R, Ravarotto L, Andrighetto I, Mutinelli F: Evaluation of indirect biomarkers for detecting corticosteroids used as illegal growth promoters in beef cattle. Vet Rec. 2008, 163: 147-151. 10.1136/vr.163.5.147.View ArticlePubMedGoogle Scholar
- Moreno-Sánchez N, Rueda J, Reverter A, Carabaño MJ, Díaz C: Muscle-specific gene expression is underscored by differential stressor responses and coexpression changes. Funct Integr Genomics. 2012, 12: 93-103. 10.1007/s10142-011-0249-9.View ArticlePubMedGoogle Scholar
- Griffin CA, Kafadar KA, Pavlath GK: MOR23 promotes muscle regeneration and regulates cell adhesion and migration. Dev Cell. 2009, 17: 649-661. 10.1016/j.devcel.2009.09.004.PubMed CentralView ArticlePubMedGoogle Scholar
- Livingstone I, Johnson Mastaglia FL: Effects of dexamethasone on fibre subtypes in rat muscle. Neuropathol Appl. 1981, 7: 381-398. 10.1111/j.1365-2990.1981.tb00240.x.View ArticleGoogle Scholar
- Stella R, Biancotto G, Krogh M, Angeletti R, Pozza G, Sorgato MC, James P, Andrighetto I: Protein expression changes in skeletal muscle in response to growth promoter abuse in beef cattle. J Proteome Res. 2011, 10: 2744-2757. 10.1021/pr101255c.View ArticlePubMedGoogle Scholar
- Belanto JJ, Diaz-Perez SV, Magyar CE, Maxwell MM, Yilmaz Y, Topp K, Boso Jamieson CH, Cacalano NA, Jamieson CA: Dexamethasone induces dysferlin in myoblasts and enhances their myogenic differentiation. Neuromuscul Disord. 2010, 20: 111-121. 10.1016/j.nmd.2009.12.003.PubMed CentralView ArticlePubMedGoogle Scholar
- Semsarian C, Wu MJ, Ju YK, Marciniec T, Yeoh T, Allen DG, Harvey RP, Graham RM: Skeletal muscle hypertrophy is mediated by a Ca2+−dependent calcineurin signalling pathway. Nature. 1999, 400: 576-581. 10.1038/23054.View ArticlePubMedGoogle Scholar
- Horsley V, Jansen KM, Mills ST, Pavlath GK: IL-4 acts as a myoblast recruitment factor during mammalian muscle growth. Cell. 2003, 113: 483-494. 10.1016/S0092-8674(03)00319-2.View ArticlePubMedGoogle Scholar
- Carrasco MA, Hidalgo C: Calcium microdomains and gene expression in neurons and skeletal muscle cells. Cell Calcium. 2006, 40: 575-583. 10.1016/j.ceca.2006.08.021.View ArticlePubMedGoogle Scholar
- Sun L, Ma K, Wang H, Xiao F, Gao Y, Zhang W, Wang K, Gao X, Ip N, Wu Z: JAK1-STAT1-STAT3, a key pathway promoting proliferation and preventing premature differentiation of myoblasts. J Cell Biol. 2007, 179: 129-138. 10.1083/jcb.200703184.PubMed CentralView ArticlePubMedGoogle Scholar
- Otto A, Schmidt C, Luke G, Allen S, Valasek P, Muntoni F, Lawrence-Watt D, Patel K: Canonical Wnt signalling induces satellite-cell proliferation during adult skeletal muscle regeneration. J Cell Sci. 2008, 121: 2939-2950. 10.1242/jcs.026534.View ArticlePubMedGoogle Scholar
- Richardson BE, Beckett K, Nowak SJ, Baylies K: SCAR/WAVE and Arp2/3 are crucial for cytoskeletal remodeling at the site of myoblast fusion. Development. 2007, 134: 4357-4367. 10.1242/dev.010678.PubMed CentralView ArticlePubMedGoogle Scholar
- Vuocolo T, Byrne K, White J, McWilliam S, Reverter A, Cockett NE, Tellam RL: Identification of a gene network contributing to hypertrophy in callipyge skeletal muscle. Physiol Genomics. 2007, 28: 253-272.View ArticlePubMedGoogle Scholar
- Davis E, Jensen CH, Schroder HD, Farnir F, Shay-Hadfield T, Kliem A, Cockett N, Georges M, Charlier C: Ectopic Expression of DLK1 Protein in Skeletal Muscle of Padumnal Heterozygotes Causes the Callipyge Phenotype. Curr Biol. 2004, 14: 1858-1862. 10.1016/j.cub.2004.09.079.View ArticlePubMedGoogle Scholar
- Groot MJ, Schilt R, Ossenkoppele JS, Berende PL, Haasnoot W: Combinations of Growth Promoters in Veal Calves: Consequences for Screening and Confirmation Methods. J Vet Med A. 1998, 45: 425-440. 10.1111/j.1439-0442.1998.tb00845.x.View ArticleGoogle Scholar
- Rhea Savu S, Silvestro L, Haag A, Sorgel F: A Confirmatory HPLC–MS/MS Method for Ten Synthetic Corticosteroids in Bovine Urines. J Mass Spectrom. 1996, 31: 1351-1363. 10.1002/(SICI)1096-9888(199612)31:12<1351::AID-JMS431>3.0.CO;2-C.View ArticleGoogle Scholar
- Antignac JP, Le Bizec B, Monteau F, Poulain F, André F: Collision-induced dissociation of corticosteroids in electrospray tandem mass spectrometry and development of a screening method by high performance liquid chromatography/tandem mass spectrometry. Rapid Commun Mass Spectrom. 2000, 14: 33-39. 10.1002/(SICI)1097-0231(20000115)14:1<33::AID-RCM829>3.0.CO;2-R.View ArticlePubMedGoogle Scholar
- Van den Hauwe O, Dumoulin F, Antignac JP, Bouche MP, Elliott C, van Peteghem C: Liquid chromatographic-mass spectrometric analysis of 11 glucocorticoid residues and an optimization of enzymatic hydrolysis conditions in bovine liver. Anal Chim Acta. 2002, 473: 127-134. 10.1016/S0003-2670(02)00754-7.View ArticleGoogle Scholar
- Antignac JP, Le Bizec B, Monteau F, André F: Study of natural and artificial corticosteroid phase II metabolites in bovine urine using HPLC-MS/MS. Steroids. 2002, 67: 873-882. 10.1016/S0039-128X(02)00048-X.View ArticlePubMedGoogle Scholar
- Deventer K, Delbeke FT: Validation of a screening method for corticosteroids in doping analysis by liquid chromatography/tandem mass spectrometry. Rapid Commun Mass Spectrom. 2003, 17: 2107-2114. 10.1002/rcm.1157.View ArticlePubMedGoogle Scholar
- Van den Hauwe O, Campbell K, Crooks SRH, Schilt R, van Peteghem CH: Confirmation of synthetic glucocorticoids with liquid chromatography/mass spectrometry: Organization and results of an international interlaboratory comparison test J. AOAC Int. 2005, 88: 87-94.Google Scholar
- Croes K, Goeyens L, Baeyens W, van Loco J, Impens S: Optimization and validation of a liquid chromatography tandem mass spectrometry (LC/MSn) method for analysis of corticosteroids in bovine liver: Evaluation of Keyhole Limpet β-glucuronidase/sulfatase enzyme extract. J Chromatogr B. 2009, 877: 635-644. 10.1016/j.jchromb.2009.01.028.View ArticleGoogle Scholar
- Tölgyesi A, Sharma VK, Kovacsics L, Fekete J: Quantification of corticosteroids in bovine urine using selective solid phase extraction and reversed-phase liquid chromatography/tandem mass spectrometry. J Chromatogr B. 2010, 878: 1471-1479. 10.1016/j.jchromb.2010.03.041.View ArticleGoogle Scholar
- Vincenti M, Girolami F, Capra P, Pazzi M, Carletti M, Gardini G, Nebbia C: Study of Dexamethasone Urinary Excretion Profile in Cattle by LC−MS/MS: Comparison between Therapeutic and Growth-Promoting Administration. J Agric Food Chem. 2009, 57: 1299-1306. 10.1021/jf803465d.View ArticlePubMedGoogle Scholar
- Groot MJ, Ossenkoppele JS, Bakker R, Pfaffl MW, Meyer HHD: Nielen MWF: Reference histology of veal calf genital and endocrine tissues - An update for screening on hormonal growth promoters. J Vet Med A. 2007, 54: 238-246. 10.1111/j.1439-0442.2007.00927.x.View ArticleGoogle Scholar
- Divari S, Cannizzo FT, Uslenghi F, Pregel P, Mulasso C, Spada F, De Maria R, Biolatti B: Corticosteroid hormone receptors and prereceptors as new biomarkers of the illegal use of glucocorticoids in meat production. J Agric Food Chem. 2011, 59: 2120-2125. 10.1021/jf1044603.View ArticlePubMedGoogle Scholar
- Nebbia C, Urbani A, Carletti M, Gardini G, Balbo A, Bertarelli D, Girolami F: Novel strategies for tracing the exposure of meat cattle to illegal growth-promoters. Vet J. 189, 189: 34-42.View ArticleGoogle Scholar
- Antignac JP, Le Bizec B, Monteau F, Poulain F, Andrè F: Collision-induced dissociation of corticosteroids in electrospray tandem mass spectrometry and development of a screening method by high mass spectrometry. Rapid Commun Mass Spectrum. 2000, 14: 33-39. 10.1002/(SICI)1097-0231(20000115)14:1<33::AID-RCM829>3.0.CO;2-R.View ArticleGoogle Scholar
- Antignac JP, Monteau F, Négriolli J, André F, Le Bizec B: 2004: Application of hyphenated mass spectrometric techniques to the determination of corticosteroid residues in biological matrices. Chromatographia. 2004, 59 (Suppl 1): S13-S22.View ArticleGoogle Scholar
- Verdon E, Hurtaurd-Pessel D, Sanders P: Evaluation of the limit of performance of an analytical method based on a statistical calculation of its critical concentrations according to ISO Standard 11843. Application to Routine Control of Banned Veterinary Drug Residues in Food According to European Decision 657/2002/EC. Accred, Qual Assur. 2006, 11: 58-62. 10.1007/s00769-005-0055-y.View ArticleGoogle Scholar
- Edgar R, Domrachev M, Lash AE: Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res. 2002, 30: 207-210. 10.1093/nar/30.1.207.PubMed CentralView ArticlePubMedGoogle Scholar
- Saeed AI, Sharov V, White J, Li J, Liang W, Bhagabati N, Braisted J, Klapa M, Currier T, Thiagarajan M, Sturn A, Snuffin M, Rezantsev A, Popov D, Ryltsov A, Kostukovich E, Borisovsky I, Liu Z, Vinsavich A, Trush V, Quackenbush J: TM4: a free, open-source system for microarray data management and analysis. Biotechniques. 2003, 34: 374-378.PubMedGoogle Scholar
- Saeed AI, Bhagabati NK, Braisted JC, Liang W, Sharov V, Howe EA, Li J, Thiagarajan M, White JA, Quackenbush J: TM4 microarray software suite. Methods Enzymol. 2006, 411: 134-193.View ArticlePubMedGoogle Scholar
- Thusher V, Tibshirani R, Chu G: Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci USA. 2001, 98: 5116-5121. 10.1073/pnas.091062498.View ArticleGoogle Scholar
- Tibshirani R, Hastie T, Narasimhan B, Chu G: Diagnosis of multiple cancer types by shrunken centroids of gene expression. Proc Natl Acad Sci USA. 2002, 99: 6567-6572. 10.1073/pnas.082099299.PubMed CentralView ArticlePubMedGoogle Scholar
- Fox J: The R Commander: a basic-Statistics Graphical user interface to R. Stat Softw. 2005, 19: 1-42.Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.