Sampling and analytical procedures:

R. palustris strain CGA009 was grown in batch culture at room temperature in the presence or absence of added iron and molybdate under (i) aerobic chemoheterotrophic growth supplemented with 2 mM ammonium, (ii) photoheterotrophic nitrogen-fixing growth under anaerobic conditions and (iii) photoheterotrophic nitrogen-fixing under anaerobic conditions with added NaS and cysteine as reductants. Aerobic cultures (30 ml) in Nunc flasks (75 cm2) were shaken at 45 rpm. Before the start of the incubations, bacteria were adjusted for >12 generations (> 2 transfers into fresh media) to the growth conditions. Photoheterotrophic anaerobic cultures (10 ml) were grown with a constant light source in Baltch-type glass tubes (25 ml volume) with a nitrogen headspace, sealed with a butyl rubber stopper. For all anaerobic incubations, bacteria were grown for >30 generations (5 transfers into fresh media) in media supplemented with Fe but without Mo. This was necessary to dilute Mo originally present in the medium to levels low enough for a clear measurable Mo growth limitation. All incubations were performed in duplicate. Bacterial growth was monitored spectrophotometrically as optical density at 660 nm. To confirm the absence of oxygen in rhodopetrobactin producing cultures, samples were taken occasionally in a glovebox (COY Vinyl) for measurement of dissolved oxygen with an oxygen probe (Hach HQ40d) and redox potential with an ORP probe (Hana HI-98120) or the redox indicator resazurine. Dissolved oxygen concentrations were always below the detection limit of the oxygen sensor (< 0.02 mg/l). Redox potential measurements with the ORP probe in anaerobic treatments were <90 mV and <150 mV vs. NHE in media without and with added reductants. Reductant additions to anaerobic media clarified the redox indicator resazurine indicating a redox potential <110 mV and had a noticeable H2S smell.

Bacterial incubation samples (1 ml) were collected throughout growth, filtered through 0.2 m syringe filters (Millipore MILLEX GP 0.22 m) and the supernatants were stored at 2208C until analysis. Quantification of rhodopetrobactins was performed on a single quadrupole LC-MS system (Agilent 6120), equipped with a diode array detector. Prior to analysis, the samples were acidified with 0.1% acetic acid and 0.1% formic acid. Samples (100 L) were injected onto a C18 column (Agilent Eclipse Plus C18, 3.5 m, 4.6x100 mm) equipped with a matching guard column. The separation proceeded with A and B solutions (solution A: water, 0.1% formic acid, 0.1% acetic acid; solution B: acetonitrile, 0.1% formic acid, 0.1% acetic acid) over 30 min, at a flow rate of 0.8 ml/min. Using a 6-port valve, the column outflow was diverted to waste for the first 5.25 min ensuring that the sample was desalted before introduction into the mass spectrometer. For quantification, UV/Vis traces were extracted at 294 nm, and peak areas corresponding to the elution of rhodopetrobactin A and B were determined using MassHunter software (Agilent). Concentrations were determined with standard solutions of 3,4-dihydroxybenzoic acid and isolated rhodopetrobactins. The detection limit for rhodopetrobactins was approximately 0.1 M.

Iron concentrations in the supernatants were measured by inductively coupled plasma-mass spectrometry (Thermo iCAP-Q in KED mode) after acidification (10% HNO3) and dilution (1:5). The detection limit for iron (⁵⁶Fe) was approximately 0.065 M. 