ABSTRACT: Escherichia coli strain MG1655 was grown in a Zn-depleted custom-built chemostat. Culture volume (120 ml), temperature (37 oC) and stirring speed (440 rpm) were maintained. Steady state values for pH and OD600 were 6.9 and 0.6, respectively. Chemostats were grown for 50 h to allow five culture volumes to pass through the vessel and allow an apparent (pseudo-)steady state to be reached. At this point, ZnSO4.7H2O in water was added to a final concentration of 0.2 M in the chemostat. A 10 ml sample of culture was taken using a polypropylene pipette tip immediately prior to Zn addition and 2.5, 7, 10 and 30 min after addition. Samples were harvested into RNAprotect and total RNA was purified using QiagenM-bM-^@M-^Ys RNeasy Mini kit (using the supplierM-bM-^@M-^Ys protocol) prior to use in microarray analysis. A control experiment was also carried out in which water was added. Biological experiments were carried out twice (i.e. control (water added) and experiment (ZnSO4 added) chemostats were grown separately twice), and a dye swap performed for each experiment, providing at least two technical repeats for each of the two biological repeats at all five time points. Slides were analysed so that time points M-bM-^@M-^\with ZnM-bM-^@M-^] were compared to the same time point M-bM-^@M-^\without ZnM-bM-^@M-^] (e.g. 30 min after Zn was added was compared to 30 min after water being added). Cells were grown in glycerol-glycerophosphate medium (GGM) lacking Zn. Final concentrations in GGM are: MES (40.0 mM), NH4Cl (18.7 mM), KCl (13.4 mM), beta-glycerophosphate (7.64 mM), glycerol (5.00 mM), K2SO4 (4.99 mM), MgCl2 (1.00 mM), EDTA (134 microM), CaCl2.2H2O (68.0 microM), FeCl3.6H2O (18.5 microM), H3BO3 (1.62 microM), CuCl2.2H2O (587 nM), CoNO3.6H2O (344 nM), and (NH4)6Mo7O24.4H2O (80.9 nM) in MilliQ water (Millipore). Bulk elements (MES, NH4Cl, KCl, K2SO4 and glycerol in MilliQ water at pH 7.4 (batch growth) or 7.6 (continuous culture)) were passed through a column containing Chelex-100 ion exchange resin (Bio-Rad) to remove contaminating cations. Trace elements (with no added Zn) and a CaCl2 solution were then added to give the final concentrations shown above prior to autoclaving. After autoclaving, MgCl2 and beta-glycerophosphate were added at the final concentrations shown above. All chemicals were of AnalaR grade of purity or higher. Chelex-100 was packed into a Bio-Rad Glass Econo-column (approximately 120 mm M-CM-^W 25 mm) that had previously been soaked in 3.5% nitric acid for 5 d. E. coli strain MG1655 was grown a custom-built chemostat made entirely of non-metal parts. Glass growth vessels and flow-back traps were soaked extensively (approximately two months) in 10% nitric acid before being rinsed thoroughly in MilliQ water. Vent filters (Vent Acro 50 from VWR) were connected to the vessel using PTFE tubing. Metal-free pipette tips were used (MAXYMum Recovery Filter Tips from Axygen). Culture volume was maintained at 120 ml using an overflow weir in the chemostat vessel. The dilution rate (and hence the specific growth rate) was 0.1 h-1. The vessel was inoculated using one of the side-arms. Flasks were placed on KMO 2 Basic IKA-Werke stirrers (arbitrary setting: 400). Stirring speed was calibrated (440 rpm) and matched between vessels using a handheld laser tachometer (Compact Instruments Ltd). Temperature was kept constant at 37 oC. Twice daily, a 2-ml sample was taken and used to test the pH (which stayed constant at pH 6.9), OD600 (0.6 at steady state), glycerol limitation, and contamination on nutrient agar plates. E. coli strain MG1655 was grown in a M-bM-^@M-^\Zn-freeM-bM-^@M-^] chemostat. Chemostats were grown for 50 h to allow five culture volumes to pass through the vessel and allow an apparent (pseudo-)steady state to be reached. At this point, ZnSO4.7H2O in water was added to a final concentration of 0.2 M in the chemostat. A 10 ml sample of culture was taken using a polypropylene pipette tip immediately prior to Zn addition and 2.5, 7, 10 and 30 min after addition. Total RNA was stabilised using RNAprotect (Qiagen). Total RNA was purified using QiagenM-bM-^@M-^Ys RNeasy mini kit. A control experiment was carried out in which water was added. Microarray slides were set up so that time points M-bM-^@M-^\with ZnM-bM-^@M-^] were compared to the same time point M-bM-^@M-^\without ZnM-bM-^@M-^] (e.g. 30 min after Zn was added was compared to 30 min after water being added). For each time point, equal quantities of RNA obtained after addition of zinc and addition of water were labelled by using nucleotide analogues of dCTP containing either Cy3 or Cy5 fluorescent dyes (Perkin Elmer). For each microarray slide, one sample was labelled with Cy3-dCTP, while the other sample was labelled with Cy5-dCTP. RNA (15-20 micrograms) was annealed to 9 micrograms pd(N)6 random hexamers (Amersham Biosciences) by heating at 65 oC for 10 min, followed by 10 min at 22 oC. This was supplemented with 4 microlitres of 5M-CM-^W First-strand buffer (Invitrogen), 2 microlitres dNTP mixture (5 mM each dATP, dTTP, dGTP and 2 mM dCTP), 2 microlitres 0.1 M DTT, 2 microlitres 1 mM Cy3 or Cy5 (Perkin Elmer) and 1 microlitre Superscript II reverse transcriptase (200 U) (Invitrogen). This was incubated at 42 oC for 2 hours. The reaction was terminated by the addition of NaOH (5 microlitres of 1M) and HCl (5 microlitres of 1M) before the addition of TE buffer (200 microlitres). Labelled cDNA was purified using a Qiaquick PCR purification kit (Qiagen). The Cy3-dCTP-labelled sample was mixed with the Cy5-dCTP-labelled sample and re-suspended in 120 microlitres salt-based hybridisation buffer (supplied with the microarray slides). This was heated at 95 oC for 3 min then cooled on ice for 3 min. The mixture was pipetted onto a microarray slide, sealed with a GeneFrame and coverslip in a hybridization chamber and incubated for 18 h at 42 M-BM-0C. Following hybridization, microarray slides (minus GeneFrame and coverslip) were washed in a series of pre-warmed (37 M-BM-0C) SSC buffers for 5 min each at 37oC: 1M-CM-^W SSC/0.1 % SDS, 1M-CM-^W SSC, 0.2M-CM-^W SSC and 0.01M-CM-^W SSC. Microarray slides were dried by centrifugation at 500 M-CM-^W g for 2 min before scanning. Slides were scanned on an Affymetrix 428 scanner. The average signal intensity and local background correction were obtained using a commercially available software package from Biodiscovery, Inc (Imagene, version 4.0 and GeneSight, version 3.5). Spots automatically flagged as bad, negative or poor in the Imagene software were removed before the statistical analysis was carried out in GeneSight. The mean values from each channel were log2 transformed and normalised using the Lowess method to remove intensity-dependent effects in the log2(ratios) values. The Cy3/Cy5 fluorescent ratios were calculated from the normalized values. Biological experiments were carried out twice times, and a dye swap performed for each experiment, providing at least two technical repeats for each of the two biological repeats at all five time points. Data from the independent experiments were combined. Genes that were differentially expressed M-bM-^IM-% twofold and displayed and P value of < 0.05 (as determined by a t test) were defined as being statistically significantly differentially transcribed.