normalization data transformation protocol
To quantify the signals, the images were processed in to GPR (5th) and TXT (7th) files through the GenePix Pro Sofware Suite version 3.0 by using the GAL file for probe annotation, as obtained from Exiqon. The GAL file is based on miRBase version 20.0. These GPR or TXT files were then imported in to R 2.11.0, for quality control and statistical analysis. First, the quality of all arrays was inspected using arrayQC, an in-house quality control pipeline that generates virtual images, boxplots, correlation plots, clustering images, MvA and PCA plots. Spike-ins were used for all arrays as an extra quality check. Within this dataset, no arrays were deviated technically. During data analysis the spot intensities were background-corrected and then filtered for low signals (i.e. intensity < 10). Only the mouse microRNA reporter intensities (661 unique reporters x 4 technical replicates) were used for further analyses. If for a reporter less than 3 of the 4 technical replicates were left, then the reporter intensity on that specific reporter was set to NA. The data were log2 transformed, followed by a quantile normalization and a summarization step. In the summarization step, the median intensity from the remaining technical replicates per condition was calculated. Finally, for each reporter a cut-off filtering was applied using the median of the three biological replicates. Only reporters that had a median intensity ≥50 for at least one condition of the 9 compounds tested, or their respective controls, passed this filtering per time point. After filtering, log2 ratios were calculated between average intensities of the treatment of its respective control. ID_REF = VALUE = Quantile normalized signal intensities
array scanning protocol
After washing, the arrays were scanned using the GenePix 4000A scanner (Axon Instruments, Foster City, CA).
normalization data transformation protocol
All raw datasets from generated CEL files were first imported into R2.11.0 and run through an in-house quality control pipeline (available through ArrayAnalysis.org (Eijssen et al, 2013)) and then used for statistical analysis. R package is freely available for academic use from the Comprehensive R Archive Network (http://www.cran.r-project.org/). The probe sets were re-annotated by use of custom Cell Definition Files (CDF) (version 12.1.0. http://brainarray.mbni.med.umich.edu/Brainarray/Database/CustomCDF/17.0.0/entrezg.asp). This resulted in a list of 16,331 unique gene IDs. After re-annotation, data were normalized by Robust Multichip Average (RMA) normalization and log2 transformed using the R package “affy”. Low expression genes were filtered out by using a log[intensity cutoff] >5. This cutoff was based on the 1st quartile of the distribution of average intensities of all reporters. ID_REF = VALUE = RMA normalized intensity signals
array scanning protocol
standard Affymetrix protocol
To adjust the volume to 200 μl, nuclease free water was added to the labeled samples. Thereafter 200 μl hybridization buffer provided in miRCURY LNA™ microRNA Array, 5th generation kit (Exiqon, Denmark) was added. In a total volume of 400 μl the sample was denatured for 2 minutes at 95ºC, incubated on ice for at least 15 minutes and then spun down. Manual hybridization was performed using an Agilent hybridization SureHyb chamber kit and gasket slide kit at 56 °C for 16h in a hybridization oven with rotation. Following manual hybridization the arrays were automatically washed in a Tecan HS4800 Pro Hybridization station according to Exiqon instructions. For each biological experiment, one hybridization per time point was conducted and one sample per array.
Amplified, biotinylated and fragmented targets were then hybridized on to the arrays. After hybridization, arrays were washed and stained using an Affymetrix fluidics station and scanned by use of an Affymetrix GeneArray scanner. A total of 60 RNA samples was prepared and analyzed on GeneChip arrays (treated and control samples from each time point were at least in triplicate). Normalization quality controls, including scaling factors, average intensities, present calls, background intensities, noise, and raw Q values, appeared to be within acceptable limits for all chips. Hybridization controls BioB, BioC, BioD, and CreX were called present on all chips and yielded the expected increases in intensities.
High-density oligonucleotide GeneChips from Affymetrix were used to measure gene expression levels (Mouse Genome 430 2.0 array (45101 probes)). Targets for these arrays were prepared from 250 ng of total RNA by means of the GeneChip 3’ IVT Express Kit according to the Affymetrix protocol (Affymetrix UK Ltd, High Wycombe, UK).
The miRCURYTM locked-nuclei acid array (LNA), 5th and 7th generation (Exiqon, Denmark) contains probes that detect mature forms of all microRNAs present in miRBase 20.0 (http://www.mirbase.org/) (Griffiths-Jones et al, 2008). The Exiqon platform has been validated and is shown to reliably detect microRNA expression. The 5th and 7th generation microarray contains 9,360 and 14,708 reporters, which represent several control probes and sequences of mature human/mouse/rat specific microRNA (4 technical replicates). In our analysis, 661 unique mouse reporters (measurable on both generation arrays) were used. cDNA was generated using 1 μg of total RNA per sample. RNA was labeled only with Hy3 containing dye (single color) using the mercury LNA™ microRNA Hy3 Power labeling kit (Exiqon, Denmark) according to the manufacturer’s protocol in a total volume of 12.5 μl.
sample treatment protocol
After a recovery period of 40-42h, the Dulbecco's Modified Eagle's (Gibco BRL, Breda, The Netherlands) culture medium was replaced by culture medium containing either one of the 6 compounds or a vehicle control (0.5% of DMSO or PBS). All chemicals were obtained through Sigma-Aldrich, except for TCDD (Cerilliant). A dose causing minimal cytotoxicity after 24h of exposure thus resulting in ±80% viability, was established by the MTT assay (Mosmann, 1983). Next, cells were treated with an IC20-24h dose of aflatoxin B1 (AFB1), benzo(a)pyrene (BaP), cisplatin (CisPl), 2,3,7,8-tetrachloordibenzodioxine (TCDD), cyclosporin A (CsA) or Wy-14,643 (Wy) (as shown in Table 1) for 24 or 48 h before being harvested. Independent biological experiments (at least in triplicate) with hepatocytes from different mice were obtained for every time point.
nucleic acid extraction protocol
At the end of treatment the medium was removed and PMH were harvested in Qiazol (QIAGEN Benelux B.V., Venlo, The Netherlands). Total RNA was isolated using a miRNeasy Mini Kit (QIAGEN Benelux B.V., Venlo, The Netherlands) according to the manufacturer’s protocol and followed by DNase I (Qiagen, Inc) treatment. Following purification, RNA concentrations were measured by means of a NanoDrop® ND-1000 spectrophotometer (Thermo Scientific, Wilmington, USA) at 260 and 280 nm. RNA quality and integrity were assessed by using automated gel electrophoresis on an Agilent 2100 Bioanalyzer system (Agilent Technologies Netherlands B.V., Amstelveen, The Netherlands). Only RNA samples which showed clear 18S and 28S peaks and with an RNA integrity number (RIN) higher than 8 were used. Samples were stored at -80ºC until RNA hybridization.
PMH were isolated from adult male C57BL/6 mice using a perfusion method and cultured between a collagen-collagen sandwich formation in two six wells plates as described by Mathijs et al., 2010.