Functional analysis of transcriptome dynamics during growth in Angus cattle longissimus muscle. Bos taurus

The bovine transcriptome dynamics in logissimus muscle (LM) during the post-natal growth remain unknown. Biopsies of LM from Angus steers were harvested at 0, 60, 120, and 220 d from early-weaning. A 13,153 bovine oligonucleotide array was used for transcript profiling. Functional analysis of microarray data was performed using the Dynamic Impact Approach (DIA) by means of KEGG and DAVID databases. During the growing phase, most of the highly-impacted pathways (e.g. Ascorbate and aldarate metabolism, Drug metabolism - cytochrome P450 and Retinol metabolism) were inhibited. The finishing phase, was characterized with the most striking differences with 3,784 differentially expressed genes (DEG; FDR 13,000 annotated sequences developed at the University of Illinois, was used for transcript profiling. Details on the development, annotation, and use of this microarray have been reported previously by Loor et al., 2007 (http://physiolgenomics.physiology.org/content/32/1/105.abstract). Methods for microarray hybridization and scanning were as reported by Loor et al. (2007). Briefly, slides were hydrated, dried, and placed in a UV Stratalinker 1800 (Stratagene, La Joya, CA) for ~5 min. Slides were washed with 0.2% SDS solution, rinsed with MilliQ (Millipore) H2O, and placed in warm prehybridization soln for 45 min at 42 C. The same amount of Cy3- or Cy5-labelled cDNA from muscle and a reference standard RNA pool (made of different bovine tissues) were co-hybridized using a dye-swap design (i.e., two microarrays per sample). Slides were incubated for 48 h at 45 C prior to scanning. Criteria for evaluation of slide quality included: identification of number of spots with a minimum median signal intensity of 3 SD above background; keeping slides with a minimum of 20,000 spots with minimum median signal intensity of 3 SD above background in both Cy3 and Cy5 channels; and keeping slides with a minimum mean intensity of 400 relative fluorescent units in both Cy3 and Cy5 channels across the entire slide. Data from a total of 112 microarrays were normalized for dye and microarray effects (i.e., Lowess normalization and microarray centering) and used for statistical analysis. Data were analyzed using the Proc MIXED procedure of SAS (SAS, SAS Inst. Inc., Cary, NC). Fixed effects were treatment (high-byproduct, high-grain) and time (0, 56, 120, and 220 d), and dye. Random effects included steer and microarray. A covariate adjustment was used to assess treatment and treatment × time interactions, but time effects on gene expression were assessed without covariate adjustment. Raw P values were adjusted using Benjamini and Hochberg’s false discovery rate (FDR). Differences in relative expression due to treatment × time interactions were considered significant at an FDR-adjusted P < 0.25 or at P < 0.01 for time effects. qPCR data were normalized using the median of 4 suitable internal control genes, and were analyzed using the same statistical model described above. Differences were considered significant at P-value of 0.05.