Bacterial strains and plasmid
Bacterial strains and plasmids used in this study are listed in Table 1.
Culture media and growth conditions
E. coli strains were grown at 37°C in LB medium overnight . S. oneidensis strains were grown at 30°C in either LB or modified R2A (MR2A) medium . Antibiotics were used at the following final concentrations: 15-μg/ml gentamicin (Gm), 50-μg/ml kanamycin (Kan) or 10-μg/ml rifamycin (Rif). For anaerobic growth of S. oneidensis MR-1 and its mutants, the defined medium M1 [per liter: 1.19-g (NH4)2SO4, 0.56-g KH2PO4, 0.1 ml (0.115 M) Na2SeO4 and 10 ml 10X mineral solution) was used as described previously . The electron donor in M1 medium was lactate (20 mM, pH7.0), and electron acceptors were ferric citrate (10 mM), KNO3 (10 mM) or fumarate (10 mM).
In-frame deletion of SO1377 gene
In-frame deletion of the SO1377 gene was generated by the method of Link et al  and its schematic presentation is shown in Figure 1. The deletion process was carried out as follows. In the first step (Fig. 1A), PCR primers were used to amplify 5'- and 3'- end fragments of SO1377 gene, respectively. Primers SO1377-No (5'-CGCGAGCTCGCCTTAGCCCTCCTCATCGG-3') and Ni (5'-tgtttaaacttagtggatgggTCCTTTCGTTTACCTACACA-3') were for the 5'-end fragment, and SO1377-Ci (5'-cccatccactaagtttaaacaTTGAGCCAAAGATAAGTAAT-3') and Co (5'-CGCGAGCTCGCTCCATGGCAAATAGCCGC-3') for the 3'-end fragment. The outside primers (No and Co) harbor a SacI restriction site. The inside primers (Ni and Ci) contain complementary 21-nt tags at their respective 5' termini as shown in lower case letters. The 21-nt complementary tags used in this study also contain a PmeI restriction site. Amplification was performed in a 20-μl volume containing 0.2-μl Taq polymerase (Sigma), 2-μl 10X PCR buffer (Promega), 1-μl each primer (20 pmol/μl), 1.2-μl 25 mM MgCl2, 1-μl genomic DNA of S. oneidensis MR-1 as template (50 ng/μl), 0.5-μl 10 mM dNTP Mix (Clontech) and 13.1-μl ddH2O. The PCR mixture was denatured at 94°C for 2 minutes, followed by 30 cycles of 94°C for 30 seconds, 58°C for 1 minute and 72°C for 1 minute, followed by 7 minutes at 72°C. The PCR products were then purified from agarose gel using QIAquick Gel Purification Kit (QIAGEN Inc.). In the second step (Fig. 1A), the 5'- and 3'- end fragments were annealed at their overlapping region and amplified by PCR as a single fragment, using the outer primers (SO1377-No and SO1377-Co). The PCR amplification conditions were as described above except that the genomic DNA template was replaced by adding 0.5-μl each of 5'- and 3'- end fragments of SO1377 amplified in the first round PCR. The fusion product was purified from agarose gel using QIAquick Gel Purification Kit (QIAGEN Inc.), resuspended in 50-μl of 1X SacI restriction buffer containing 5 U of SacI restriction enzyme, and digested for two hours at 37°C. The digested fragment was gel purified, ligated into SacI-digested and phosphatase-treated suicide vector pDS3.0, electroporated into E. coli S17-1/λpir, spread on LB gentamicin agar plates and incubated at 37°C overnight. The transformants were screened for inserts by PCR with SO1377-No and Co as primers. In the third step, the suicide plasmid construct was integrated into S. oneidensis DSP-10 chromosome by homologous recombination (Fig. 1B). This process was facilitated by conjugal transfer between E. coli S17-1/λpir cells harboring the suicide plasmid construct (donor) with strain DSP-10, a spontaneous rifampicin-resistant (Rifr) derivative of S. oneidensis MR-1 (recipient). E. coli transformants and DSP-10 cells were grown separately in LB medium with proper antibiotics overnight, washed in fresh medium, and mixed in a 1:3 ratio (donor: recipient) by spotting onto 0.2 μm Millipore membrane disks. Following an eight-hour incubation at 30°C, the cells were removed from the filter disks, resuspended in medium, and plated onto LB agar supplemented with Gm (15-μg/ml) and Rif (10-μg/ml). Correct integration of the suicide vector was verified by PCR amplification. PCR confirmation of the inserts was accomplished by comparing the sizes of the products amplified from the wild-type and mutant DNA using primers flanking the insertion sites (Fo and Ro) (Fig 1B). The Fo primer sequence is: 5'-TGCAGCCAAAAGCACAGCAC-3'. The Ro primer sequence is: 5'-CGAACTGTTATGCCATAAAT-3'. In the fourth step, to obtain the deletion mutation of SO1377 gene, the integrated suicide plasmid had to be resolved from chromosome through recombination (Fig. 1C). This was accomplished by growing the strain carrying the integrated suicide vector in NaCl-less LB liquid overnight, followed by 10, 100 and 1000 dilutions of the culture, and plating 100 μl of each dilution onto LB agar containing 5% sucrose. The plates were then incubated at 30°C for two days. Screen for deletion was accomplished by PCR amplification using the outside Fo and Ro primers. Finally, the deletion was verified through DNA sequencing. The sequencing reaction was performed with ABI Prism Dye Terminator Sequencing Kit (Applied Biosystems, Foster City, CA) and analyzed by ABI 3700 sequencer.
Microarray transcriptional expression profiling
Microarray fabrication, hybridization, probe labeling, image acquisition and processing were carried out as described previously [11, 32, 33]. Gene expression analysis was performed using 3 independent microarray experiments, with each slide containing 2 replicate arrays of S. oneidemis MR-1 genome . Total cellular RNA, from the SO1377 gene deletion mutant WG3 and its parental strain DSP-10 were grown to mid-log-phase in the MR2A medium, then isolated and purified using the TRIzol Reagent (Gibco BRL) according to the manufacturer's instruction. The ratios of mutant samples to the wild-type control were normalized by a trimmed geometric mean [11, 32, 33]. Standard t-test was carried out to determine the significance of gene expression.
Proteomic profiling using 2-D PAGE analysis of whole-cell lysates
Mid-log-phase cells of WG3 and DSP-10, grown in aerobic MR2A medium, were collected and washed three times with 50 nM Tris-HCL (pH = 7.4) and the cell pellets were frozen until used. Frozen cell pellets were mixed with five volumes of a solution containing 9 M urea, 2% (v/v) 2-mercaptoethanol, 2% (v/v) pH 8–10 ampholytes (Biorad), and 4% (v/v) Nonidet P40. The lysed cells were centrifuged for 10 min at 400,000 g in a Beckman TL100 ultracentrifuge to sediment particulates. The supernatants were analyzed by 2-D PAGE. Isoelectric focusing (IEF) gels were cast as described by Anderson and Anderson  using a 2:1 mixture of pH 5–7 and pH 3–10 ampholytes (Biorad). Aliquots of samples containing 40 μg of protein were loaded, in triplicate, onto each gel. After IEF, the gels were equilibrated in a buffer containing sodium dodecyl sulfate (SDS) as described by O'Farrell (1975) . The second-dimension slab gels were cast using a linear gradient of 9–17% PAGE. The equilibrated tube gels were secured to the slab gels using agarose and SDS-PAGE as described previously . The proteins were fixed in the gels by soaking in a solution of 20% ethanol (v/v) with 1% formaldehyde (v/v) and detected by silver stain . The 2-DE images were digitized using an Eikonixl412 scanner interfaced with a VAX 4000-90 workstation, and resulting image files were converted to tif format. Data analysis was done using Progenesis software for 2-DE pattern analysis (Nonlinear USA). Statistical analysis of the relative abundance of each matched protein spot across the data set was done using a two-tailed Student t-test as previously described . Proteins showing statistically significant differences in abundance (i.e., P < 0.05) were then identified on the basis of their tryptic peptide masses and amino acid sequences. Proteins to be identified were cut from one to five replicate gels (number of spots required varied with the abundance of individual proteins), reduced at room temperature with tris (2-carboxyethyl) phosphine (Pierce, Rockport, IL), alkylated with iodoacetamide (Sigma), and digested in situ with modified trypsin (Promega Corp., Madison, WI) . The resulting peptides were eluted from the gel with 25 mM ammonium bicarbonate and 5% formic acid in 50% acetonitrile and analyzed by micro liquid chromatography-electrospray ionization tandem mass spectrometry (μ-LC-ESI-MS/MS). For μ-LC-ESI-MS/MS, samples were loaded onto a 365 × 100 μm fused silica capillary (FSC) column packed with 5 μm Zorbax XDB-C18 packing material (Agilent Technologies, Palo Alto, CA) at a length of 7–8 cm. The tryptic peptides were separated with a 30-min linear gradient of 0–60% solvent B (80% acetonitrile/0.02% heptafluorobutyric acid), and then entered a LCQ ion-trap mass spectrometer (Thermo Finnigan, San Jose, CA). Tandem mass spectra were automatically collected under computer control during 30-min LC-MS runs. MS/MS spectra were then directly subjected to SEQUEST [39, 40] database searches to identify proteins by correlating experimental MS/MS spectra to protein sequences predicted by the S. oneidensis MR-1 ORE database available in Genbank.
Assay of hydrogen peroxide challenge
Assay of hydrogen peroxide challenge was carried out as described previously . Cells grown in LB medium to mid-log-phase stage were distributed into 50 ml Erlenmeyer flasks (5 ml each), and hydrogen peroxide was added at concentrations from 0.1 mM to 10 mM. After 20 min of incubation with shaking at 30°C, treatment was stopped by addition of catalase (Roche Molecular Biochemicals) at 400 U/ml and chilling. Samples were diluted in cold 10-2 MgSO4 containing 400 U/ml catalase, followed by plating on LB plates for viable count. Colonies were counted after two days of incubation at 30°C.
Assay for whole-cell electron paramagnetic resonance (EPR) spectroscopy
The procedure for EPR assay was adapted from the description of Keyer and Imlay . After 24 h growth, with shaking in 50-ml LB liquid at 30°C, the optical density (OD600 nm) of the culture was measured. Cells were spun down at room temperature and suspended in fresh 50 ml LB liquid containing 20 mM desferrioxamine (DFO) (Sigma). After incubation with shaking for 20 min at 30°C, cells were spun down, washed with cold 20 mM Tris (pH = 7.4), resuspended in 20 mM Tris (pH = 7.4) containing 10% glycerol to reach the same cell density according to OD600 nm value measured, frozen in 3-mm quartz EPR tubes (Wilmad) on dry ice, and stored at -80°C until EPR measurement was performed. The EPR spectra were recorded using Bruker X-band spectrometer equipped with a Varian TE102 and a Varian temperature controller. Samples were maintained at -125°C during the recording of signal.
Parameters used for low temperature Fe(III) EPR were as follows: central field, 1520 G; sweep width, 500 G; resolution, 1024 points; frequency, 9.44 GHz; microwave power, 21.523 mW; receiver gain, 3.17e+4; conversion, 327.680 ms; time constant, 327.680 ms; sweep time, 335.544 s; number of scan, 4. EPR data was processed using Bruker WinEPR program.
Assay of total iron concentration
Intracellular iron concentration was determined as described previously with some modifications . Inoculated cultures were grown in 20 ml of liquid LB, either with no iron supplement or with an iron supplement of 0.05, 0.1, 0.5, 1, 5 and 10 mM ferric citrate, at 30°C for 12 hours with shaking (120 rpm). The cells were washed four times with fresh cold LB medium (twice with 10 ml and twice with 1 ml), and their OD600 was determined. The cell pellet was then dispensed in 10 ml 1 N HNO3 and incubated in a waterbath at 95°C for 30 min. After spinning down the cellular debris, at 4000 rpm for 10 min, the supernatants were recovered for determination of iron concentration using a Perkin Elmer Optima 3100XL instrument. Iron concentrations were expressed as ppm/109 cells.
Assay of siderophore production
CAS (chrome azurol S) blue agar with iron depletion was used to determine the siderophore secretion of the wild-type and deletion mutant strains . Bacterial cultures grown in LB medium, to mid-log-phase, were collected and washed twice with 0.7% NaCl solution and resuspended in 0.7% NaCl solution. A 5 μl suspension of each strain was spotted on the CAS agar, followed by incubation overnight at 30°C. The excretion of siderophore was indicated by the formation of a yellowish halo area surrounding the colony.
Spontaneous mutation rate of gentamycin or kanamycin resistance
The measurement of spontaneous mutation rate was performed according to the description of Touati et al (1995) . At the amount of 0.1 ml per plate, bacterial cultures grown in LB liquid to mid-log-phase were spread onto LB agar containing 15 μg/ml gentamicin or 50 μg/ml of kanamycin. A serial dilution of the same liquid cultures were made and spread onto LB agar for counting of colony forming unit (CFU) after incubation of two days at 30°C.