cDNA Arrays: Procedures

  1. Preparation of total RNA
  2. Amplification of cDNAs and miniarray manufacture
  3. Preparation of radiolabelled probes from total RNA
  4. cDNA miniarray hybridization and washing
  5. Data analysis

 Preparation of total RNA

Total RNA were extracted from (i) 1.5 g of roots harvested on 2-month-old cuttings and (ii) 1.3 g of leaves collected on 1-month-old cuttings according to Bugos et al. (1995) Biotechniques 19: 734-737.

 Amplification of cDNAs and miniarray manufacture

 Materials and equipment


  1. Inserts of sequenced cDNAs were amplified by PCR (95°c for 5 min and then, 95° for 1 min, 60°C for 30s, 72°c for 3 min for 30 cycles and a final cycle at 72°C for 15 min) using 20 pmoles of the M13 forward and reverse primers flanking the cDNA.
  2. PCR products (150-300 ng) were diluted in 200 µl 10X SSC and denatured with NaOH (0.2 M final concentration). Prepare JUST PRIOR to use!
  3. Amplified cDNAs were spotted on 12 x 8 cm, Hybond N filters (Amersham) using a Millipore 96-well dot-blot apparatus.
  4. Filters were baked at 80°C for 20 min.

a. The following controls were used: pBluescript plasmid and amplified pBluescript polylinker to assess unspecific hybridization to the cloning vector, and cDNA for human desmin as internal quantification standards.

 Preparation of radiolabelled probes from total mRNA

 Reagents and Equipment


  1. Total RNA was isolated from roots and leaves as described above.
  2. Complex probes were then prepared by reverse transcription using the SMART-PCR cDNA Synthesis kit and the MMLV RTase.
  3. Labeling of the cDNA probes was done in the presence of 30 µCi [32P]dCTP and [32P]dATP, and random hexamers using the 'Prime-a-Gene' kit (Promega) according to manufacturer's instructions.

 cDNA miniarray hybridization and washing


  1. Place the cDNA array filters in a hybridization tube.
  2. cDNA array filters were prehybridized for 4-8 h in 30 ml of hybridization mix (0.6 NaCl, 20 mM PIPES, 5 mM Na2EDTA, 10X Denhardt's solution, 2.5% SDS, 500 µg ml-1 salmon sperm DNA) at 65°C.
  3. Then, the total denatured cDNA probes labeled as above were added to 10 ml of the same hybridization buffer containing 20X Denhardt's solution, 100 µg ml-1 salmon sperm DNA, 20 µg ml-1 pBluescript, 2µg ml-1 polyA and 2 µg ml-1 polyT.
  4. Hybridization was performed for 24 h at 65°C in the presence of the [32P]-labelled probes.
  5. Filters were then washed at 65°C, successively for 3 X 5 min in 2X SSC, 1 X 40 min in 2X SSC, 0.1% SDS, 1 X 40 min in 1X SSC, 0.1% SDS, 1 X 40 min in 0.5X SSC, 0.1% SDS, and 1 X 40 min in 0.1X SSC, 0.1% SDS.
  6. Moist filters were then wrapped in Saran Wrap, exposed to a Kodak imaging plate of a Personnal Molecular Imager FX (Bio-Rad) for varying periods (1h to overnight), after which hybridization signals were scanned.

 Data analysis

  1. The raw image data obtained with the Molecular Imager imaging system were imported into a Apple Macintosh G3.
  2. Identification and quantification of the hybridization signals, as well as the subtraction of local background values, were carried out using the Bio-Rad Quantity One software.
  3. The resulting data were stored and manipulated in the Excel 98 worksheet programme (Microsoft Corporation).
  4. To standardize the relative level of plant and fungal transcripts in the intermingling symbiotic tissues, hybridization signals were normalized to the transcript levels of plant or fungal ribosomal proteins. (a)

a. To determine the reproducibility of hybridizations in which RNAs are present at a wide range of concentrations, we hybridized three copies of a filter arrayed using cDNAs coding for various components of the transduction pathways with two labeled complex probes prepared from the same RNA fungal extract. The filter-to-filter variation was relatively small (± 10%) (data not shown). To assess the variation due to sources other than the hybridization and reading steps, we hybridized samples independently prepared from two different fungal cultures grown in the same conditions. The sample-to-sample variation was slightly larger (± 15%) than observed in the simple hybridization reproducibility experiments.

Special thanks to Catherine Voiblet, Sébastien Duplessis and Denis Tagu (INRA-Nancy), and Gopi Podila (Michigan Tech, USA) for their help in developping these procedures.

Return to Info main page