Computational protocol: Expression of cell wall related genes in basal and ear internodes of silking brown-midrib-3, caffeic acid O-methyltransferase (COMT) down-regulated, and normal maize plants

Similar protocols

Protocol publication

[…] The maize cell wall macro-array, which consists of gene-specific tags (GSTs) for 651 genes, was described in detail by Guillaumie et al. []. Spotted genes were chosen firstly as maize orthologs of genes expressed during secondary cell wall formation of zinnia (Zinnia elegans Jacq.) tracheary elements [] and secondly from a keyword strategy developed from sequences previously described as involved in primary and secondary cell wall biosynthesis and assembly for all species throughout the plant kingdom. Maize orthologous genes were found from BLAST (blastn and tblastx) searches [] performed against GenoPlante-Info databases [].Membrane hybridizations and data analyses were achieved according to a protocol adapted from Pesquet et al. [] and developed in Guillaumie et al. []. Each GST was diluted to a final concentration of 0.5 mg/ml and then denatured in 50% DMSO. Controls were added in separate 384-well plates. One plate included 384 Tris-EDTA, pH 8.0 (blank background control). Another plate had 30 NPT II fragments (a positive hybridization control), 20 pBluescript plasmids (unspecific hybridization control), and 18 ubiquitin fragments (positive control). All fragments were spotted onto a 20 × 20-cm Nytran SuPerCharge nylon membrane (Schleicher and Schuell, Keene, NH, USA) using a BioGrid spotting robot (BioRobotics) in a 4 × 4 grid organization. Each gene was spotted twice in two different membrane grids corresponding to four replicates. cDNA probes were synthesized according to Guillaumie et al. [] from 10 μg of total RNA for each sample. Membranes were placed in a PhosphorImager cassette (Molecular Dynamics, Amersham- Pharmacia) for 72 h and scanned at 50 mm/pixel by a Storm 820 scanner (Amersham-Pharmacia).Data analysis was performed according to Pesquet et al. [] and Guillaumie et al. []. Macro-array gridding and gene expression levels were measured with ImageQuant 5.0 software (Molecular Dynamics, Amersham-Pharmacia) using 4 × 4 grids. Three independent hybridizations, each corresponding to one of the three groups of ear or basal internodes, were performed for normal, AS225 and bm3 plants. Reproducibility of raw signal intensity values on the membrane was first verified inside each grid and between the two grids for each investigated gene. According to Pesquet et al. [] and Guillaumie et al. [], two threshold values were thus defined so that 95% of the ratios between raw values of two replicates would be within the 0.5–2.0 interval. Aberrant values giving ratios outside of this interval were discarded. Normalization was performed based on the blank background, unspecific hybridization and positive controls (Tris-EDTA pH8, pBluescriptII, ubiquitin and kanamycin-NPTII). Macro-array reproducibility was further investigated by comparing normalized spot intensity values from the three independent hybridizations performed on three independent membranes. As was developed for reproducibility within the membrane, the investigations on the threshold values showed that more than 95% of the ratio values between independent hybridizations were confined within a twofold limit. Out of range values were discarded and 96–99% of the ratios between replicates were found between the two 0.5–2.0 threshold values. Expression data were estimated as the average of normalized intensity signal values of replicates, and comparisons between F2bm3 or AS225 with the normal F2 line were based on expression value ratios. According to Pesquet et al. [] and Guillaumie et al. [], genes with more than a twofold expression ratio in F2bm3 or AS225 compared with F2 were considered as differentially expressed genes. In addition, a variance analysis based on elementary normalized signal intensity values, followed by a bilateral Student t test (risk levels of P < 0.05), was performed. All genes that were considered differentially expressed were also validated by this statistical analysis. The normalized hybridization level under 3000 was similar to the macro-array blank value, and the minimal level of significant hybridizations was thus fixed to a normalized value equal to 6000. For each differentially expressed gene, the mRNA number was searched using a BLAST (blastn) program against all available mRNA sequences in NCBI database.In this investigation, only a subset of genes spotted on the MAIZEWALL macro-array was considered, in order to focus on lignin pathway, cell wall carbohydrate and cell wall protein genes and their transport and regulation factors. This subset thus comprised 105 phenylpropanoid genes, 44 transcription factor genes, 22 auxin related and tissue patterning genes, 33 transport and detoxification process associated genes, 18 cell wall protein genes, and 114 nucleotide sugar and cell wall carbohydrate genes. The exhaustive gene list has been given by Guillaumie et al. []. […]

Pipeline specifications

Software tools BLASTN, TBLASTX
Databases BioGRID GPI JACQ
Application Nucleotide sequence alignment
Organisms Zea mays, Bos taurus