Larvae were either left untreated or treated with 5 µM CuSO4 for 1 h at 28.5 degrees Celsius. Untreated larvae or copper-treated larvae at specified times post-treatment were anaesthetized in Ringer's solution with 600 μM 3 aminobenzoic acid ethyl ester methanesulfonate and their skins removed with fine forceps. Approximately 50 skins were collected for each sample. Skins were placed on ice, and their cells dissociated by incubation with 0.25% trypsin-EDTA (Life Technologies) for 15 min in a water bath at 28.5 ˚C. The samples were then triturated with a P1000 pipet five times or until visibly homogenized. After the trypsin digestion had been quenched with 30% fetal bovine serum and 6 mM CaCl2 in phosphate-buffered saline solution (PBS), the liberated cells were recovered by centrifugation (400 x g for 5 min at 4 ˚C). The pellet was rinsed once with Ca2+-free Ringer’s solution containing 0.5 mg/mL DNaseI (Sigma), resedimented by centrifugation, and resuspended in 100 μl of the same solution. The suspension was kept on ice until just prior to sorting, when it was passed once through a 40 μm filter. Cells were sorted in a flow cytometer equipped with an 85 μm nozzle and 488 nm and 561 nm lasers (FACSAria II, BD Biosciences). Distinct populations of cells were isolated on the basis of forward scattering, lateral scattering, and the intensity of mCherry or GFP fluorescence. Sorted cells were collected in a lysis-buffer solution (RNeasy Micro Kit, Qiagen) supplemented with 130 mM β mercaptoethanol and were stored at 80 ˚C until RNA extraction.
Growth protocol
Fish embryos were produced by natural mating and maintained in system water at 27 degrees Celsius until 3 days post-fertilization, at which point they were moved into E3 medium and shifted to 28.5 degrees Celsius. Samples were collected 4 days post-fertilization.
Extracted molecule
total RNA
Extraction protocol
Total RNA was isolated by a standard protocol (RNeasy Micro Kit, Qiagen). The yield and quality of the product were measured with respectively a spectrophotometer (NanoDrop 1000, NanoDrop Technologies) and a bioanalyzer (Agilent 2100). Only samples with an RNA integrity score greater than 8.0 were selected for the preparation of cDNA libraries. One nanogram of total RNA from each sample was amplified to several micrograms of cDNA (Pico WTA System V2, NuGEN, Inc.) before labeling. In the cases of GFP+ hair cells and mCherry+/GFP+ mantle cells, cells from several sorts were pooled to provide sufficient RNA for each amplification reaction.
Label
Biotin
Label protocol
Amplified cDNA libraries were fragmented and biotin-labeled using a standard protocol (Encore Biotin Module, NuGEN, Inc.).
Hybridization protocol
Biotin-labeled cDNA libraries were hybridized to Affymetrix Zebrafish Gene 1.0 ST Arrays for 16 hours at 45°C as described in the Affymetrix Technical Analysis Manual (Affymetrix, Santa Clara, CA). After hybridization, Gene Chips were stained with streptavidin-phycoerythrin, followed by an antibody solution (anti-streptavidin) and a second streptavidin-phycoerythrin solution, with all liquid handling performed by a GeneChip Fluidics Station 450.
Scan protocol
Arrays were scanned with the Affymetrix GeneChip Scanner 3000 using AGCC, Affymetrix GeneChip Command Console 3.1 core software according to standard protocols.
Description
polyA RNA amplified
Data processing
Data processing was performed using the Partek Genomics Suite, v6.6 (Partek, Inc.). Probeset data were imported using RMA background correction, and were normalized by quantile normalization. Data for comparison of all untreated cell types to each other were imported separately from the time course experiment data (for which untreated mCherry+ and NF cells were re-imported). This was done to prevent cell types not included in the time course (i.e. GFP+ hair cells and mCherry+/GFP+ mantle cells) from affecting quantile normalization of treated samples. In the accompanying submission, data for all untreated samples is as imported for comparison between untreated cell types. Probeset summarization was done by median polish. Imported array data were subjected to principal component analysis, and outliers discarded (not included in this submission). Batch effects due to different sample collection and processing dates were then identified by ANOVA and removed. Probeset data were summarized to transcript-level data for gene-expression analysis; Transcript-level data are included in the accompanying submission. Lists of differentially-expressed genes were compiled according to set cut-offs (i.e. 2-fold up- or down-regulation between sample classes) with Benjamini-Hochberg-adjusted ANOVA-derived P values of less than 0.05.
