ABSTRACT: It has been postulated that noncoding RNAs (ncRNAs) are involved in the post-transcriptional control of gene expression, and may have contributed to the emergence of the complex attributes observed in mammalians. The complement of ncRNAs expressed from intronic regions of the human and mouse genomes comprises at least 78,147 and 39,660 transcriptional units, respectively. To identify conserved intronic sequences expressed in both humans and mice, we used custom-designed human cDNA microarrays to separately interrogate RNA from mouse and human liver, kidney and prostate tissues. An overlapping tissue expression signature was detected for both species, comprising 198 transcripts; among these, twenty two RNAs map to intronic regions with evidence of evolutionary conservation in humans and mice. Transcription of selected human-mouse intronic ncRNAs was confirmed using strand specific RT-PCR. Altogether, these results support an evolutionarily conserved role of intronic ncRNAs in human and mouse, which are likely to be involved in the fine tuning of gene expression regulation in different mammalian tissues. Keywords: tissue expression pattern 1. Experiment Design: 1.1. Type of experiment: The experiments described here aim to generate expression profiles from human and mouse prostate, kidney and liver tissue samples using a custom-designed cDNA microarray platform with 4,608 unique elements in replicate (9,216) enriched in gene fragments that map to intronic regions of known human genes. The platform consists of 3,355 probes of human cDNA fragments that map in the human genome sequence either to intronic regions of known genes (822), to unannotated regions of the human genome (241) or to known exons of RefSeq genes (2,292). We separately compared the expression profiles obtained from human and mouse samples to identify sequence conserved tissue-specific intronic transcripts. 1.2. Experimental factors: Tissues from liver, prostate and kidney from five adult male mice of the Balb/c strain were obtained and frozen in liquid nitrogen. Likewise, five non-tumor samples of human prostate, kidney or liver from our tumor bank were selected. All samples were obtained from patients who signed informed consent, and approval was received from the ethics committees of the hospitals. Total RNA was purified from tissues and treated with DNAse to reduce the levels of genomic DNA contaminant. Either human and or mouse samples from for each tissue from the five individuals were separately pooled using equal quantities of RNA. 1.3. The number of hybridizations performed in the experiment. RNA pools from each tissue sample was hybridized twice, performing a total of 18 hybridizations. 1.4. The type of reference used for the hybridizations, if any. All hybridizations included external mRNA spikes for quality control of the experiments (detailed below in section 2.5). Hybridizations were performed as single color experiments. 1.5. Hybridization design: Three aliquots of 15ug of the pool of total RNA from each tissue was labeled with Cy5-dUTP after reverse transcription (CyScribe first strand cDNA labeling kit, Amershan Biosciences) and hybridizations were carried out using all labeled-target in an Automate Slide Processor (Amershan Biosciences). Images were obtained using a Generation III Scanner (Amersham Biosciences). The human and mouse obtained expression data were independently analyzed using a highly stringent methodology, including different statistical approaches (see below for a detailed description). 1.6. Quality control steps taken: Integrity of total RNA was checked by electrophoresis using the Bio Sizing total RNA Nano assay in the 2100 Bioanalyzer (Agilent Technologies). Three replicates of each experiment were performed, as describe in the previous section.
