Various stresses were applied to the cultures incubated at 37oC with shaking at 210 rpm: ethanol stress - the culture was treated with a final concentration of 4% (v/v) ethanol; salt stress - solid sodium chloride was added to a final concentration of 4% (w/v); butanol stress - the culture was treated with a final concentration of 1% (v/v) butanol (Helmann et al., 2003); heat stress – the culture was shifted from 37oC to 48oC (Helmann et al., 2001); oxidative stress – the culture was treated with hydrogen peroxide in a final sub-lethal concentration of 60 μM (Helmann et al., 2003). Samples were taken in regular time intervals of 5, 10, 15, 20 min after the imposition of the environmental stress factors. For low-temperature growth, the cultures were grown in minimal medium at 37oC to an OD540 of 0.1, and then the cultures were shifted to 15oC. Afterwards, the cultures were sampled at an OD540 of 0.9 and 1.0. Glucose limitation was triggered by growing the strains with growth-limiting amounts of 0.05% (w/v) glucose at 37°C. Oxygen limitation was induced by reducing the shaking speed at OD540 of 0.3 from 210 rpm to 50 rpm. The sampling for both glucose and oxygen limitation was conducted at the time intervals of 15, 30, 45, 60, 90 min after the induction of the limitation.
Growth protocol
The strains were precultured in Luria-Bertani (LB) medium, in case of BSM29 supplemented with 50 μg/ml spectinomycin. For the stress experiments, a defined minimal medium was selected (Stülke et al., 1993)
Extracted molecule
total RNA
Extraction protocol
Total RNA was isolated by the mechanical cell disruption/ acid phenol method (Petersohn et al., 2001).
Label
Cy5
Label protocol
For annealing 12,5 µg of total RNA and 1,25 µl of 2µg/µl random hexamer primer were incubated in 14.5 µl deonized water at 65°C for 10 min. Reverse Transcription was conducted by adding 5 µL of 5 x First Strand Buffer, 2.5 µl of 0.1 M DTT, 1 µL of dNTP mix (5 mM dATP; 5 mM dCTP; 5 mM dGTP; 2 mM dTTP) (Roche Diagnostics), 1 ml of 1 mM Cy3-dUTP or Cy5 dUTP (GE Healthcare), 1 µL SuperScript II reverse transcriptase (200U/µl, Invitrogen). The reaction mixture was incubated at 42 °C for 2 h. Subsequently, the remaining RNA was hydrolyzed by incubation with 5 µL of 1M NaOH at 65 °C for 10 min. After neutralization with 5 µL of 1M HCl and 100 µL of TE buffer (pH 8.0), the labeled cDNA was purified using the CyScribe GFX Purification Kit (GE Healthcare).
Various stresses were applied to the cultures incubated at 37oC with shaking at 210 rpm: ethanol stress - the culture was treated with a final concentration of 4% (v/v) ethanol; salt stress - solid sodium chloride was added to a final concentration of 4% (w/v); butanol stress - the culture was treated with a final concentration of 1% (v/v) butanol (Helmann et al., 2003); heat stress – the culture was shifted from 37oC to 48oC (Helmann et al., 2001); oxidative stress – the culture was treated with hydrogen peroxide in a final sub-lethal concentration of 60 μM (Helmann et al., 2003). Samples were taken in regular time intervals of 5, 10, 15, 20 min after the imposition of the environmental stress factors. For low-temperature growth, the cultures were grown in minimal medium at 37oC to an OD540 of 0.1, and then the cultures were shifted to 15oC. Afterwards, the cultures were sampled at an OD540 of 0.9 and 1.0. Glucose limitation was triggered by growing the strains with growth-limiting amounts of 0.05% (w/v) glucose at 37°C. Oxygen limitation was induced by reducing the shaking speed at OD540 of 0.3 from 210 rpm to 50 rpm. The sampling for both glucose and oxygen limitation was conducted at the time intervals of 15, 30, 45, 60, 90 min after the induction of the limitation.
Growth protocol
The strains were precultured in Luria-Bertani (LB) medium, in case of BSM29 supplemented with 50 μg/ml spectinomycin. For the stress experiments, a defined minimal medium was selected (Stülke et al., 1993)
Extracted molecule
total RNA
Extraction protocol
Total RNA was isolated by the mechanical cell disruption/ acid phenol method (Petersohn et al., 2001).
Label
Cy3
Label protocol
For annealing 12,5 µg of total RNA and 1,25 µl of 2µg/µl random hexamer primer were incubated in 14.5 µl deonized water at 65°C for 10 min. Reverse Transcription was conducted by adding 5 µL of 5 x First Strand Buffer, 2.5 µl of 0.1 M DTT, 1 µL of dNTP mix (5 mM dATP; 5 mM dCTP; 5 mM dGTP; 2 mM dTTP) (Roche Diagnostics), 1 ml of 1 mM Cy3-dUTP or Cy5 dUTP (GE Healthcare), 1 µL SuperScript II reverse transcriptase (200U/µl, Invitrogen). The reaction mixture was incubated at 42 °C for 2 h. Subsequently, the remaining RNA was hydrolyzed by incubation with 5 µL of 1M NaOH at 65 °C for 10 min. After neutralization with 5 µL of 1M HCl and 100 µL of TE buffer (pH 8.0), the labeled cDNA was purified using the CyScribe GFX Purification Kit (GE Healthcare).
Hybridization protocol
The Cy3 and Cy5 labeled cDNA solutions (60 μl) were combined with 43.7 μl nuclease water, 43.5 μl 20x SSC (Ambion), 10.9 μl 2% SDS (diluted 1:10 with nuclease water from 20% SDS, Ambion) and 21.9 μl array block solution (GeneScan mixed). This hybridization solution was then denatured for 1 min at 100°C, cooled down for 5 min on ice and immediately pipetted onto the array. Two gene frames (ABgene, Epsom, UK) had been glued previously on top of the hybridization region creating a 240 μl chamber on the microarray according to the manufacturer’s instructions. The resulting chamber was filled with hybridization solution and closed with a flexible coverslip (ABgene) without air bubbles. The microarray was then inserted into a 50 ml tube covered with aluminum foil and incubated under rotation at 60°C for 16 hours. After hybridization, the gene frames with coverslip were carefully removed, and the array was reinserted in the 50 ml tube. Subsequently, the microarray in the 50 ml tubes was subjected to washing in the following buffers: 2x SSC, 0.1% SDS for 5 min at 65°C, 0.5% SSC for 5 min at 25°C, and 0.1x SSC for 10 min at 25°C. Finally, the arrays were placed in an empty 50 ml tubes and centrifuged at 2500 rpm for drying.
Scan protocol
Scanned on ScanArray Express microarray scanner (Perkin Elmer Lifesciences, Boston, USA; Cy3 at 543 nm, Cy5 at 633 nm). Images were quantified using AArray Vision 6.0 program (GE Healthcare).
Description
dye swap
Data processing
Median trimmed density with the local background substracted Array Vision 6.0 software (GE Healthcare) was used. Thereafter the 3 replicate spots were flagged as present (P), marginal (M), or absent (A). A spot was flagged as present (P) if the S/N ratio >2 and the difference of the median spot intensities between replicate spots was less than 1.25 fold. Marginal spots had to meet the same S/N ratio, but the difference of the median spot intensities between replicate spots had to be less than 1.5 fold. Spots with a S/N ratio<2 or a difference of the median spot intensities between replicate spots greater than 1.5 fold were flagged as absent. Replicate spot intensities were combined using the median of the spots flagged as present or marginal. If none of the replicate spots was flagged as present or marginal, the spot intensity values of the absent flagged spots were taken to obtain a complete data matrix. Generally, correction for systematic effects (e.g., dye effects) in dual dye microarray data is usually achieved using Lowess normalization. In our hybridization design, the wild type and sigB knock-out were always cohybridized, consequently the gene expressions differs widely depending on stress condition and time point. The assumption of Lowess normalization that a fraction of genes/probes (our study 50%) has unchanged expressions is therefore not applicable for most of the hybridizations. Because of this we used the synthetic slide method to normalize our data. Synthetic slides were created from samples having similar/equivalent gene expression pattern. The equivalence of samples was determined using correlation analysis. Therefore we separated the signals of the Cy3 and Cy5 channel for all slides and did a correlation analysis using Pearsons product moment. Then we paired these datasets into synthetic slides, if their Pearsons correlation was above 0.8. To normalize for systematic effects, the paired datasets (synthetic slides) were normalized by Lowess (f = 0.5) using MicroPrep (van Hijum et al., 2003). In the next step we calculated the ratios of gene expression for the wild type strain with respect to the sigB knock-out strain. These ratio values were labelled as "Synthetic slide normalized ratio of wt/mutant (wt168/BSM29)"
