Phenazines and other redox-active antibiotics promote microbial mineral reduction

Abstract

Natural products with important therapeutic properties are known to be produced by a variety of soil bacteria, yet the ecological function of these compounds is not well understood. Here we show that phenazines and other redox-active antibiotics can promote microbial mineral reduction. Pseudomonas chlororaphis PCL1391, a root isolate that produces phenazine-1-carboxamide (PCN), is able to reductively dissolve poorly crystalline iron and manganese oxides, whereas a strain carrying a mutation in one of the phenazine-biosynthetic genes (phzB) is not; the addition of purified PCN restores this ability to the mutant strain. The small amount of PCN produced relative to the large amount of ferric iron reduced in cultures of P. chlororaphis implies that PCN is recycled multiple times; moreover, poorly crystalline iron (hydr)oxide can be reduced abiotically by reduced PCN. This ability suggests that PCN functions as an electron shuttle rather than an iron chelator, a finding that is consistent with the observation that dissolved ferric iron is undetectable in culture fluids. Multiple phenazines and the glycopeptidic antibiotic bleomycin can also stimulate mineral reduction by the dissimilatory iron-reducing bacterium Shewanella oneidensis MR1. Because diverse bacterial strains that cannot grow on iron can reduce phenazines, and because thermodynamic calculations suggest that phenazines have lower redox potentials than those of poorly crystalline iron (hydr)oxides in a range of relevant environmental pH (5 to 9), we suggest that natural products like phenazines may promote microbial mineral reduction in the environment.

FIG. 1.

Shown are the structures of a representative humic substance molecule, drawn after those described previously (49) (a); AQDS (b); bleomycin (c); and a generic phenazine (d). If x is CONH₂, the compound is PCN; if x is COOH, the compound is PCA.

FIG. 2.

Total Fe(II) produced in culture (a) and CFU (b) of P. chlororaphis PCL1391 (WT) (⧫) and PCL1119 (phzB mutant) (○) grown in the presence of poorly crystalline iron(III) hydr(oxides). Data are representative of at least three independent experiments.

FIG. 3.

(a) Total phenazine production (▴) in a PCL1391 culture representing the sum of PCN (○) and PCA (▪). Data represent the average of two independent extractions of each time point in the same experiment; bars show the data range. (b) Total Fe(II) production by PCL1391 (WT) (⧫), PCL1119 (phzB mutant) (○) and PCL1119 with PCN added (40 μM, added at time point indicated by arrow) (▵). Data are representative of at least two independent experiments.

FIG. 4.

Total Fe(II) produced (a) and cell growth achieved (b) by the reduction of poorly crystalline iron (hydr)oxide by S. oneidensis MR1 in the absence (▴) or presence of 10 μM AQDS (▪) or 10 μM PCN (○). (c) Total Fe(II) produced by the reduction of poorly crystalline iron (hydr)oxide by S. oneidensis MR1 in the absence (▴) or presence of a final 1:10 dilution of filtered culture supernatant from the WT strain PCL1391 (▪) or the mutant PCL1119 (○). (d) Total Fe(II) produced by the reduction of poorly crystalline iron (hydr)oxide by S. oneidensis MR1 in the absence (▴) or presence (○) of 15 to 20 μM bleomycin. Data represent the average value of duplicate experiments, with bars showing the data range.

FIG. 5.

E_h/pH diagram showing the thermodynamic stability region of iron species and PCN as a function of pH and E_h. The dashed lines represent the redox potential for the PCN redox couple at PCN_red/PCN_oxid ratios of 1 to 1 and 100 to 1, respectively. To draw the Fe(OH)₃/Fe²⁺ boundary line, we assumed Fe²⁺ concentrations of 10 μM, 1 mM, and 10 mM. H₂ and O₂ lines frame the stability field for water.