Fresh samples of whole rumen content were withdrawn from a cannulated sheep, farmed at INRA of Clermont-Ferrand-Theix (France), by inertial pumping using a Perspex tube. Groundwater sample was collected by SITA Remediation Company (Lyon, France) by direct pumping at 10 m depth in a piezometer bored near the contamination source and immediately stored on ice until processed.
Extracted molecule
total RNA
Extraction protocol
Rumen sample collected was filtered through nylon membrane of 250 µm porosity and 0.5 g of filtrate was recovered and centrifuged at 21 000 g. Total RNA extraction was performed from the pellet using the optimized protocol described by Culley (Culley, D.E., Kovacik, J.W.P., Brockman, F.J. and Zhang, W. (2006) Optimization of RNA isolation from the archaebacterium Methanosarcina barkeri and validation for oligonucleotide microarray analysis. Journal of Microbiological Methods, 67, 36-43). Groundwater sample was transferred to the laboratory, 300 to 500 mL were filtered under vacuum and collected onto a 47 mm TSTP Millipore filter (3 µm pore size). Filters were then cut in pieces and placed into 900 µL of nucleic extraction buffer (50 mM glucose, 10 mM EDTA, 25 mM Tris-HCl pH 8.0) as described by Vetriani (Vetriani, C., Tran, H.V. and Kerkhof, L.J. (2003) Fingerprinting Microbial Assemblages from the Oxic/Anoxic Chemocline of the Black Sea. Appl. Environ. Microbiol., 69, 6481-6488). In this protocol, freeze-thaw cycles were replaced by beat betting 1 min at 30 Hz with 1 g of glass beads of ≤106 μm in diameter (Sigma-Aldrich). Total RNA were extracted using a standard phenol-chloroform-isoamyl alcohol (25:24:1) method followed by a second extraction with an equal volume of chloroform. Nucleic acids were then precipitated, washed in 70% ethanol and resuspended in nuclease free water. Co-extracted DNA was removed by digestion with 4U of (RNAse free) DNase I (DNA-free, Ambion) at 37°C for 35 min. 8-9 µg of total RNA mixture were subjected to mRNA enrichment protocol using the MicrobExpress kit (Ambion Applied Biosystems), which employs capture hybridization by magnetic separation to remove 16S and 23S rRNA. In a second step, antisense amino-allyl dUTP marked RNA (aRNA) was obtained by amplification with both the MessageAmp II-Bacteria and the Amino-Allyl MessageAmp II kits (Ambion). The entire mmoC (S81887), tceA (AF228507) and vcrA (AY322364) genes sequences associated to SP6 promoter sequence at their 3’ end were synthesized by Biomatik Corporation (USA) and cloned into the pGH vector. Antisense RNAs (aRNAs) were produced by in vitro transcription using the MEGAscript SP6 Kit (Ambion).
Label
Cy3
Label protocol
Antisense mRNA obtained from rumen and groundwater samples were labelled with Cy3 fluorescent dye (CyDyeTM Post-Labeling Reactive Dye Packs, GE Healthcare) following the manufacturer's instructions. Antisense RNAs (aRNAs) from the three synthetic genes mmoC, tceA and vcrA were labelled with Cy3-ULS as described in the Kreatech ULS labelling procedure (Kreatech Diagnostics).
Hybridization protocol
For each hybridization experiment, labeled antisenses RNA (Cy3) were vacuum dried and resuspended in 5.6 µL of water. Then, the hybridization mix (Roche NimbleGen) was realized according to the manufacturer's protocols. The arrays were hybridized on a 4-bay NimbleGen Hybridization System (Roche NimbleGen) at 42°C for 72 hr. Arrays were washed with NimbleGen wash buffers I, II and III according to vendor protocols.
Scan protocol
Samples were scanned using a Scanner Innoscan 900AL (Innopsys, Carbonne, France) at 2 μm resolution. Individual array images were acquired independently, adjusting the PMT gain for each image as recommended using the software Mapix® (Innopsys). Then, for each array image, raw expression data had been extracted using the NimbleScan software v2.1. (Roche NimbleGen) and feature intensities were exported as .pair files.
Description
Total RNA extracted from sheep rumen content farmed at INRA of Clermont-Ferrand-Theix (France)
Data processing
The background noise was then determined using random probes present on the microarrays (8,863 probes in our experiment). This background noise is defined by two components: the background median intensity (Bposition) and its dispersion (Bdispersion). Finally, a modified signal-to-noise ratio called SNR’ and based on the formula of Verdik (Verdick, D., Handran, S. and Pickett, S. (2002) In LLC, D. P. (ed.), In G. Kamberova (ed.), DNA array image analysis: nuts and bolts. Salem, MA., pp. p. 83-98) is calculated as follow in order to reduce-centralize our data: SNR’ = (probe signal intensity - Bposition)/ Bdispersion. However, spatial effect across array surface is a predominant within-slide experimental artifact that needs to be eliminated before any other normalization procedure (Wang, X., He, H., Li, L., Chen, R., Deng, X.W. and Li, S. (2006) NMPP: a user-customized NimbleGen microarray data processing pipeline. Bioinformatics, 22, 2955-2957). That is why, for all array images obtained in this work, surface was segmented to 16 sub-squares according to probe position (X, Y) indicated in pair report. A Perl script was developed to calculate local background noise in all sub-squares and the median SNR’ retrieved from the three replicates of each probe. Finally, another Perl script was implemented to summarize each replicate probe treated and determine the median value between the three replicates. Positive hybridization was considered significant for probes having an SNR’ > 3 (value to avoid all false positives).
Experimental validation of HiSpOD probe design with complex background mRNA or 3 known antiense RNA and expression analysis on a contaminated groundwater sample
Data table header descriptions
ID_REF
VALUE
SNR'-normalized, averaged gene-level signal intensity (by mediane calculation).
