Competition and coexistence of sulfate-reducing bacteria, acetogens and methanogens in a lab-scale anaerobic bioreactor as affected by changing substrate to sulfate ratio

Abstract

The microbial population structure and function of natural anaerobic communities maintained in lab-scale continuously stirred tank reactors at different lactate to sulfate ratios and in the absence of sulfate were analyzed using an integrated approach of molecular techniques and chemical analysis. The population structure, determined by denaturing gradient gel electrophoresis and by the use of oligonucleotide probes, was linked to the functional changes in the reactors. At the influent lactate to sulfate molar ratio of 0.35 mol mol(-1), i.e., electron donor limitation, lactate oxidation was mainly carried out by incompletely oxidizing sulfate-reducing bacteria, which formed 80-85% of the total bacterial population. Desulfomicrobium- and Desulfovibrio-like species were the most abundant sulfate-reducing bacteria. Acetogens and methanogenic Archaea were mostly outcompeted, although less than 2% of an acetogenic population could still be observed at this limiting concentration of lactate. In the near absence of sulfate (i.e., at very high lactate/sulfate ratio), acetogens and methanogenic Archaea were the dominant microbial communities. Acetogenic bacteria represented by Dendrosporobacter quercicolus-like species formed more than 70% of the population, while methanogenic bacteria related to uncultured Archaea comprising about 10-15% of the microbial community. At an influent lactate to sulfate molar ratio of 2 mol mol(-1), i.e., under sulfate-limiting conditions, a different metabolic route was followed by the mixed anaerobic community. Apparently, lactate was fermented to acetate and propionate, while the majority of sulfidogenesis and methanogenesis were dependent on these fermentation products. This was consistent with the presence of significant levels (40-45% of total bacteria) of D. quercicolus-like heteroacetogens and a corresponding increase of propionate-oxidizing Desulfobulbus-like sulfate-reducing bacteria (20% of the total bacteria). Methanogenic Archaea accounted for 10% of the total microbial community.

Fig. 1

Schematic representation of the experiment. Reactor R-0 was the start-up reactor. The culture in R-0 was used as an inoculum for the reactors R1 and R2a. The cultures were run at a dilution rate of 0.02 h⁻¹ for five to six volume changes with each volume change of 50 h

Fig. 2

Effluent concentrations of sulfide, sulfate, acetate, and propionate in reactors R1, R2a, and R2b. The sulfide data are minimum numbers due to N₂ stripping

Fig. 3

DGGE analysis of 16S rRNA gene fragments using DNA and RNA samples from reactors R-0, R1, R2a, and R2b. Lanes 1 (DNA) and 2 (RNA) sample from reactor R-0 (influent lactate 6.7 mM); lanes 3 (DNA) and 4 (RNA) sample from reactor R1 (influent lactate 3.5 mM); lanes 5 (DNA) and 6 (RNA) from reactor R2a (influent lactate 20 mM); lanes 7 (DNA) and 8 (RNA) from reactor R2b (in the near absence of sulfate). Bands indicated with a dot were excised and sequenced

Fig. 4

Phylogenetic tree based on 16S rRNA gene sequences obtained from the DGGE bands (see Fig. 3). Sequences determined in this study are in boldface. Black dots on the nodes indicate bootstrap values of 90% and higher (1,000 replicates). The scale bar indicates 10% sequence difference

Fig. 5

a Relative abundance of Archaea, sulfate-reducing bacteria, and acetogenic bacteria. b Relative abundance of Desulfomicrobium-, Desulfovibrio-, Desulfobulbus-, and Dendrosporobacter-like bacteria. The percentage abundance is relative to the signal obtained with probe EUBmix

Fig. 6

Whole-cell hybridization of samples from reactors R-0, R1, R2a, and R2b probe EUBmix labeled with Cy5 (blue), probe SRB385 and 385Db labeled with Fluos (green), and probe SPS770 labeled with Cy3 (red). a Reactor R-0 (influent lactate 6.7 mM); b reactor R1 (influent lactate 3.5 mM); c reactor R2a (influent lactate 20 mM); d reactor R2b (in the near absence of sulfate)