Spatial distribution of some microbial trophic groups in a plug-flow-type anaerobic bioreactor treating swine manure.

Anaerobic digestion of swine manure is carried out by a consortium of microbial species, including volatile fatty acid (VFA) producers, VFA-degraders and methanogens. The distribution of five phylogenetic groups within a plug-flow-type anaerobic bioreactor consisting of eight serially-connected tanks was examined through the sequential digestion of swine manure. Quantification was carried out using reverse transcription real-time PCR (RT-Q-PCR) assays targeting the 16S rRNA of Clostridium (cluster XIVa), Peptostreptococcus, Syntrophomonas, Methanosaeta, and Methanosarcina spp. The VFA producers Peptostreptococcus spp. and Clostridium spp. were found predominantly in compartments where hydrolysis/acidogenesis took place. The spatial distribution of the aceticlastic methanogens, Methanosaeta and Methanosarcina, within the bioreactor was not correlated with methanogenic activity. In contrast the VFA-degrading genus Syntrophomonas spp. was more abundant in compartments with elevated methanogenic activity. Multivariate statistical analyses of the RT-Q-PCR data have provided new insights into our understanding of how the various trophic groups were distributed within this bioreactor system. While the distribution of clostridia, peptostreptococci and Syntrophomonas corresponded to their known metabolic functions, aceticlastic methanogens were not apparently linked to the methanogenesis stage occurring in latter compartments, suggesting that hydrogenotrophic methanogens were the primary methane generators in this bioreactor. However, aceticlastic methanogens could be involved in compartments related to the hydrolysis/acidogenesis stage.

Multivariate statistical analyses of rDNA and rRNA fingerprint data to differentiate microbial communities in swine manure.

Fingerprint data from swine manure microbial community rRNAs and rRNA genes were treated by multivariate statistical and diversity analyses to differentiate swine manures. Microbial communities from finishing pig manure and from a mixture of manure slurries from maternity confinement and finishing pigs were characterized using a combination of amplicon length heterogeneity PCR (LH-PCR) and terminal restriction fragment length polymorphism (T-RFLP), using PCR primers targeting Bacteria and Archaea, respectively. Unweighted pair group method with arithmetic mean clustering, principal components analysis (PCA), indicator species analysis (ISA), and diversity analyses showed that rRNA-based fingerprinting methods [reverse transcription (RT)-LH-PCR and RT-T-RFLP] were more effective than rDNA-based fingerprinting methods for distinguishing the manure samples. Multiresponse permutation procedure from fingerprint data showed that all manure samples had distinct microbial communities. PCA and ISA showed that the major phylotypes differentiating the LH-PCR or the RT-LH-PCR profiles were distributed differently between manures, suggesting that the bacterial community structure was different from the metabolically active bacterial community. The detection of minor archaeal populations was greater using RT-T-RFLP instead of T-RFLP. The findings indicated that the analysis of microbial community rRNAs could differentiate each manure sample from the others and would be appropriate for the monitoring of metabolically active populations.

Bacterial community dynamics in an anaerobic plug-flow type bioreactor treating swine manure.

A plug-flow type anaerobic reactor consisting of eight sequential compartments was used to study shifts in a bacterial community adapted to degrade swine manure at 25 degrees C. The investigation was carried out during the first 6 months of reactor operation. The reactor successfully separated the hydrolysis/acidogenesis stage from the methanogenesis stage. Bacterial 16S rDNA- and rRNA-based fingerprints obtained through amplicon length heterogeneity PCR (LH-PCR) were analyzed with a view to characterizing the bacterial community structure and the metabolically active community, respectively. Multivariate statistical tools showed that the rDNA-based fingerprints described a more temporal than compartmentalized distribution of similar bacterial communities. By contrast, the rRNA-based multivariate analyses described a distribution that was linked more to reactor performance parameters, especially during short time periods. Diversity indices calculated from fingerprint data were used to assess overall diversity shifts. The decrease in rRNA-based diversity observed through the reactor compartments was greater than the decrease in rDNA-based diversity. This finding indicates that the analysis of metabolically active bacteria diversity was more discriminative than the analysis based on the mere presence of bacteria. The observed decrease in diversity suggests that the bacterial community became specialized in degrading less diversified substrates through the compartments. All these findings suggest that rRNA-based analyses are more appropriate for monitoring reactor performance.