ABSTRACT: To identify mRNAs associated with human PUM2 protein, we used a modified Ribonucleoprotein-ImmunoPrecipitation Microarray (RIP-Chip) approach on HeLa S3 cancer cells that express PUM2. PUM ribonucleoprotein (RNP) complexes were captured from cell-free extracts with specific anti PUM2 antibody coupled to protein A sepharose beads, and then eluted with SDS-EDTA. To control for non-specifically enriched RNAs, the same procedure was performed with beads that were not coupled with immunoprecipitating antibodies (mock samples). Total RNA was isolated from cell extracts and immunopurified samples with the mirVanaTM PARISTM kit (Ambion). RNA was quantified with a NanoDrop device (Witeg AG). Poly-adenylated RNAs were amplified in the presence of aminoallyl-UTP with Amino Allyl MessageAmp II aRNA kit (Ambion). For this purpose, 500 ng total RNA from extracts and half (50-100 ng) of the immunopurified RNAs were used for amplification. 8 ug of the amplified RNAs (aaRNA) were fluorescently labeled with NHS-monoester Cy3 and Cy5 dyes (GE HealthSciences), except for mock RNA samples, where an aaRNA amount proportional to the yield obtained from corresponding PUM affinity isolates was used.For PUM2 RIPs, we performed four biological replicates but omitted the dye swaps due to the lower aaRNA obtained after amplification (~10 ug aaRNA from PUM2 RIPs, 9 ug aaRNA from mock RIPs). The Cy3- and Cy5-labeled aaRNA samples were mixed and hybridized to human cDNA microarrays. Detailed methods for microarray experiments are available at http://cmgm.stanford.edu/pbrown/protocols/index.html. cDNA microarrays were produced by the Stanford Functional Genomic Facility and contained 43,197 human probes representing 26,524 Unigene cluster IDs (12,466 ENSEMBL annotated genes) spotted on Corning Ultra GAPS slides. Spotted cDNAs were cross-linked with 65 mJ of UV irradiation on slides, which were then post-processed for 1 hour at 42°C in pre-hybridization solution (5x SSC, 0.1% SDS, 0.1 mg/ml BSA), washed twice in 400 ml of 0.1x SSC for 5 min, dunked in 400 ml ultrapure water for 30 sec, and dried by centrifugation at 550 rpm for 5 min. Slides were used the same day. Cy3- and Cy5-labeled aaRNA probes were mixed and applied to arrays in hybridization solution (3x SSC, 20 ug poly(A) RNA [Invitrogen], 20 ug yeast tRNA [Invitrogen], 20 ug Human Cot-1 DNA [Invitrogen], 20 mM HEPES [pH 7.0] and 0.3% SDS) for 18 h at 65°C. The arrays were then washed sequentially in 400 ml of 2x SSC with 0.1% SDS, 1x SSC, and 0.2x SSC. The first wash was performed for 5 min at 65°C, the subsequent washes were performed for 5 min at RT. The arrays were dried by centrifugation and immediately scanned with an AxonScanner 4200A (Molecular Devices). Data were collected using GENEPIX 5.1 (Molecular Devices). Arrays were normalized computationally by the Stanford Microarray Database (SMD). The data were filtered for signal over background of greater than 1.5 in the channel measuring aaRNA from extract, and only features that met these criteria in > 50% of the arrays were included for further analysis. Log2 median ratios were retrieved and exported into Microsoft Excel. To identify transcripts that were specifically enriched by association with PUM2, we performed two class Significance Analysis of Microarrays (SAM) on median centered arrays. A replicate experimental design type is where a series of replicates are performed to evaluate reproducibility or as a pilot study to determine the appropriate number of replicates for a subsequent experiments. Antibody used in IP: 50 ug rabbit anti-PUMILIO 2 (Bethyl Laboratories, #300-202A) Keywords: replicate_design Computed
