Total RNA was extracted from lamellar tissues using the TRI Reagent (Ambion, Austin, TX) and purified using the RNeasy mini kit (Qiagen, Hilden, Germany) following the manufacturer’s instructions. Briefly, frozen lamellar tissue was pulverized in liquid nitrogen using a mortar and pestle and transferred into TRI Reagent. The solution was homogenized by five aspirations through a 19-gauge needle followed by five aspirations through a 21-gauge needle. RNA was purified by phenol/bromochloropropane (BCP) extraction. After centrifugation, transition to the RNeasy micro kit protocol for cells was done after mixing the aqueous phase with the same amount of 70% ethanol. At the end of this protocol, RNA was eluted with 40 µl of DEPC-treated water.
Label
Biotin labelling.
Label protocol
See manufacturer`s web side.
Hybridization protocol
See manufacturer`s web side.
Scan protocol
See manufacturer`s web side.
Description
Total RNA isolated from each sample was processed and hybridized to the bovine GeneChip array according to the protocols described in the GeneChip Expression Analysis Technical Manual of 2005 (Affymetrix, Santa Clara, CA). Briefly, total RNA was used to generate double-stranded cDNA by using a T7-linked oligo(dT) primer. Linear amplification-biotin labeling and fragmentation were performed with the one-cycle target labeling kit (Affymetrix, P/N 900493) to have target cRNAs per manufacturer’s recommendation. cRNA yields and purity were assessed by Nanodrop spectroscopy. Samples were hybridized to bovine GeneChips (Affymetrix) per the manufacturer’s instructions. Arrays were then washed and stained with streptavidin-phycoerythrin, before being scanned on an Agilent model G2500A GeneArray scanner at the Penn microarray core facility.
Data processing
Data was normalized according to per chip median in GC-RMA algorithm. "*.CEL" file was used for that normalization. After scanning, array images were assessed by eye to confirm scanner alignment and the absence of significant bubbles or scratches. The 3'/5' ratios for housekeeping genes were confirmed to be within acceptable limits (up to 3, normal range is up to 5), scale factors were 1.2-1.5 (normal range is up to 5) and all spike controls were as expected. In addition, data quality was assessed by review of the average background (background average = 74-84, usually less than 100) and percent present calls (%Present=%19-21) for each array. GeneChip expression data were exported to GeneSpring 7.2 where per chip normalization to the 50th percentile expression level and per gene normalization to the median expression intensity in all samples was performed. Values below 0.01 were set to 0.01. Replicates based cross gene error model was activated in GeneSpring normalizations. Statistical evaluation of replicate experiments: The data for Affymetrix control probe sets were removed prior to analysis. Differentially expressed genes were identified by significance analysis of microarrays (SAM), which is a nonparametric, permutation-based method. Normalized raw data was used as an input for SAM. Lists of differentially expressed genes were described by false discovery rate (FDR) (Benjamini and Hochberg, 1995). Transcript conversion to similar human proteins: There is a very limited database to analyze the gene expression information for bovines, we decided, then, to convert the bovine transcript information to highly similar human protein equivalents to see the biological connection within the significant gene list. Statistically significant bovine transcripts were converted to similar human proteins using the NCBI data base. Transcript sequences were used to find similar human proteins at the level of 80% identity cutoff threshold for blast for match by using the human protein Refseq IDs. The human transcripts which recognize the same protein conversion on human U133 plus ver2 GeneChip has been done to analyze in public databases too. This conversion has been done by using the NetAffx™ Analysis Center of Affymetrix.
