Pelagic photoferrotrophy and iron cycling in a modern ferruginous basin.

Iron-rich (ferruginous) ocean chemistry prevailed throughout most of Earth's early history. Before the evolution and proliferation of oxygenic photosynthesis, biological production in the ferruginous oceans was likely driven by photoferrotrophic bacteria that oxidize ferrous iron {Fe(II)} to harness energy from sunlight, and fix inorganic carbon into biomass. Photoferrotrophs may thus have fuelled Earth's early biosphere providing energy to drive microbial growth and evolution over billions of years. Yet, photoferrotrophic activity has remained largely elusive on the modern Earth, leaving models for early biological production untested and imperative ecological context for the evolution of life missing. Here, we show that an active community of pelagic photoferrotrophs comprises up to 30% of the total microbial community in illuminated ferruginous waters of Kabuno Bay (KB), East Africa (DR Congo). These photoferrotrophs produce oxidized iron {Fe(III)} and biomass, and support a diverse pelagic microbial community including heterotrophic Fe(III)-reducers, sulfate reducers, fermenters and methanogens. At modest light levels, rates of photoferrotrophy in KB exceed those predicted for early Earth primary production, and are sufficient to generate Earth's largest sedimentary iron ore deposits. Fe cycling, however, is efficient, and complex microbial community interactions likely regulate Fe(III) and organic matter export from the photic zone.

Vertical Distribution of Functional Potential and Active Microbial Communities in Meromictic Lake Kivu.

The microbial community composition in meromictic Lake Kivu, with one of the largest CH4 reservoirs, was studied using 16S rDNA and ribosomal RNA (rRNA) pyrosequencing during the dry and rainy seasons. Highly abundant taxa were shared in a high percentage between bulk (DNA-based) and active (RNA-based) bacterial communities, whereas a high proportion of rare species was detected only in either an active or bulk community, indicating the existence of a potentially active rare biosphere and the possible underestimation of diversity detected when using only one nucleic acid pool. Most taxa identified as generalists were abundant, and those identified as specialists were more likely to be rare in the bulk community. The overall number of environmental parameters that could explain the variation was higher for abundant taxa in comparison to rare taxa. Clustering analysis based on operational taxonomic units (OTUs at 0.03 cutoff) level revealed significant and systematic microbial community composition shifts with depth. In the oxic zone, Actinobacteria were found highly dominant in the bulk community but not in the metabolically active community. In the oxic-anoxic transition zone, highly abundant potentially active Nitrospira and Methylococcales were observed. The co-occurrence of potentially active sulfur-oxidizing and sulfate-reducing bacteria in the anoxic zone may suggest the presence of an active yet cryptic sulfur cycle.

Bacterial community composition in three freshwater reservoirs of different alkalinity and trophic status.

In order to investigate the factors controlling the bacterial community composition (BCC) in reservoirs, we sampled three freshwater reservoirs with contrasted physical and chemical characteristics and trophic status. The BCC was analysed by 16S rRNA gene amplicon 454 pyrosequencing. In parallel, a complete dataset of environmental parameters and phytoplankton community composition was also collected. BCC in the analysed reservoirs resembled that of epilimnetic waters of natural freshwater lakes with presence of Actinobacteria, Alpha- and Betaproteobacteria, Cytophaga-Flavobacteria-Bacteroidetes (CFB) and Verrucomicrobia groups. Our results evidenced that the retrieved BCC in the analysed reservoirs was strongly influenced by pH, alkalinity and organic carbon content, whereas comparatively little change was observed among layers in stratified conditions.

Seasonal variations and resilience of bacterial communities in a sewage polluted urban river.

The Zenne River in Brussels (Belgium) and effluents of the two wastewater treatment plants (WWTPs) of Brussels were chosen to assess the impact of disturbance on bacterial community composition (BCC) of an urban river. Organic matters, nutrients load and oxygen concentration fluctuated highly along the river and over time because of WWTPs discharge. Tag pyrosequencing of bacterial 16S rRNA genes revealed the significant effect of seasonality on the richness, the bacterial diversity (Shannon index) and BCC. The major grouping: -winter/fall samples versus spring/summer samples- could be associated with fluctuations of in situ bacterial activities (dissolved and particulate organic carbon biodegradation associated with oxygen consumption and N transformation). BCC of the samples collected upstream from the WWTPs discharge were significantly different from BCC of downstream samples and WWTPs effluents, while no significant difference was found between BCC of WWTPs effluents and the downstream samples as revealed by ANOSIM. Analysis per season showed that allochthonous bacteria brought by WWTPs effluents triggered the changes in community composition, eventually followed by rapid post-disturbance return to the original composition as observed in April (resilience), whereas community composition remained altered after the perturbation by WWTPs effluents in the other seasons.

Identification of antimicrobial resistant bacteria in rivers: insights into the cultivation bias.

In the present study, the antimicrobial resistant (AR) bacteria were quantified and identified in different river samples using in parallel a culture-based approach and a culture-independent one. The objective was to evaluate the importance of the cultivation bias when studying antimicrobial resistance among environmental bacteria. Three different river samples covering a gradient of anthropic influence were tested and three different antimicrobial compounds were used as selective agents: amoxicillin, tetracycline and sulfamethoxazole. From a quantitative point of view, our results highlight the importance of the culture media used, as for the same sample and the same selective agent significant differences were observed in the counts of culturable AR bacteria depending on the culture media used. The identification of AR bacteria through culture or culture-independent methods put on evidence AR bacterial communities that differ dramatically: γ-proteobacteria and more specifically Aeromonadaceae dominated among the isolates while ß-proteobacteria (Comamonadaceae), dominated among the sequences obtained without culture. Altogether these results highlight the necessity to develop a methodological consensus preferably without culture, to approach this important topic in the coming years.

Identification of a metagenomic gene cluster containing a new class A beta-lactamase and toxin-antitoxin systems.

Several reports mention the presence of antibiotic resistance genes in natural and polluted environments, but many studies are based on their detection via polymerase chain reaction (PCR amplification of known genes and not on an activity screening. We constructed a metagenomic fosmid bank from DNA isolated from a polluted river in Brussels, Belgium, the Zenne. A total of 120,000 clones were pooled and plated directly on solid media containing different antibiotics. Several clones were isolated which could grow in the presence of ampicillin. The DNA from several clones was extracted and subjected to restriction analysis and, based on their restriction pattern, two different clones were found. One of the clones was selected for further study as it showed a higher level of resistance to different ß-lactams antibiotics (ticarcilline and ceftazidime). To find out which gene is responsible for the resistance, an in vitro transposon mutagenesis was performed and clones having lost the resistance phenotype were analyzed via inverse PCR amplification. Several clones had an insert in a gene encoding a new type of ß-lactamase. The amplified fosmid DNA was fully sequenced revealing an insert of 41 kb containing 39 open reading frames (ORFs). Transposon insertions inactivating the resistance to ß-lactams were also found in the ORF upstream of the blaA gene, encoding an aminotransferase, suggesting a polar effect on the transcription of the gene downstream. In addition, other genes were found such as histidine biosynthesis genes, which were found to be scattered on the insert, a relA/spoT gene, and genes belonging to type II toxin-antitoxin system. This predicted system was experimentally validated in Escherichia coli using an inducible expression system.

Diversity of Microbial Communities in Production and Injection Waters of Algerian Oilfields Revealed by 16S rRNA Gene Amplicon 454 Pyrosequencing.

The microorganisms inhabiting many petroleum reservoirs are multi-extremophiles capable of surviving in environments with high temperature, pressure and salinity. Their activity influences oil quality and they are an important reservoir of enzymes of industrial interest. To study these microbial assemblages and to assess any modifications that may be caused by industrial practices, the bacterial and archaeal communities in waters from four Algerian oilfields were described and compared. Three different types of samples were analyzed: production waters from flooded wells, production waters from non-flooded wells and injection waters used for flooding (water-bearing formations). Microbial communities of production and injection waters appeared to be significantly different. From a quantitative point of view, injection waters harbored roughly ten times more microbial cells than production waters. Bacteria dominated in injection waters, while Archaea dominated in production waters. Statistical analysis based on the relative abundance and bacterial community composition (BCC) revealed significant differences between production and injection waters at both OTUs0.03 and phylum level. However, no significant difference was found between production waters from flooded and non-flooded wells, suggesting that most of the microorganisms introduced by the injection waters were unable to survive in the production waters. Furthermore, a Venn diagram generated to compare the BCC of production and injection waters of one flooded well revealed only 4% of shared bacterial OTUs. Phylogenetic analysis of bacterial sequences indicated that Alpha-, Beta- and Gammaproteobacteria were the main classes in most of the water samples. Archaeal sequences were only obtained from production wells and each well had a unique archaeal community composition, mainly belonging to Methanobacteria, Methanomicrobia, Thermoprotei and Halobacteria classes. Many of the bacterial genera retrieved had already been reported as degraders of complex organic molecules and pollutants. Nevertheless, a large number of unclassified bacterial and archaeal sequences were found in the analyzed samples, indicating that subsurface waters in oilfields could harbor new and still-non-described microbial species.

Identification of a five-oxidoreductase-gene cluster from Acetobacter pasteurianus conferring ethanol-dependent acidification in Escherichia coli.

Acetobacter pasteurianus, a Gram-negative bacterium belonging to the α-divison of Proteobacteria, produces acetic acid through ethanol oxidation. A genomic bank of A. pasteurianus 386B DNA was cloned in the low-copy cosmid pRG930Cm vector and the resulting clones were screened for the production of protease using the skimmed-milk agar assay whereby a clearing zone around the inoculated spots indicates casein degradation. Several positive clones were selected and restriction analysis revealed that many contained the same inserts. One clone was further analyzed and the cosmid DNA subjected to in vitro transposon insertion. After electroporation, several clones having lost the capacity to cause casein degradation were isolated and the sequence of the transposon-flanking regions analyzed. The majority of insertions mapped to one gene encoding an NAD(P)(+)-dependent aldehyde dehydrogenase (ALDH) of the PNTB superfamily, whereas one insert was found upstream in a gene encoding an ethanol dehydrogenase. Addition of phenol red to the medium confirmed the ethanol-dependent acidification around the inoculated spots of the clones without transposon insertion, suggesting that casein degradation is due to the production of acetic acid as a result of the combined activities of the alcohol dehydrogenase and ALDH. Quantitative data and pH measurements confirmed a significant acidification, and the presence of acetic acid.

Antimicrobial resistance of heterotrophic bacteria in sewage-contaminated rivers.

Sewage-contaminated rivers are ecosystems deeply disturbed by human activity due to the release of heavy metals, organic pollutants and pharmaceuticals as well as faecal and pathogenic micro-organisms, which coexist with the autochthonous microbial population. In this study, we compared the percentage of resistance in faecal and heterotrophic bacteria in rivers with different degrees of sewage pollution. As a matter of fact, no correlation was found neither between the degree of sewage pollution and the percentage of antimicrobial resistant heterotrophic bacteria nor between the number of resistant faecal bacteria and that of resistant heterotrophic bacteria. Most of the resistant isolates from the Zenne river downstream Brussels were multi-resistant and the resistance patterns were similar among the strains of each phylogenetic group. The total microbial community in this polluted river (as evaluated through a 16S rRNA gene clone library analysis) appeared to be dominated by the phyla Proteobacteria and Bacteroidetes while the phylum TM7 was the third most represented.

Diversity of the total bacterial community associated with Ghanaian and Brazilian cocoa bean fermentation samples as revealed by a 16 S rRNA gene clone library.

Cocoa bean fermentation is a spontaneous process involving a succession of microbial activities, starting with yeasts, followed by lactic acid bacteria and acetic acid bacteria. So far, all microbiological studies about cocoa bean fermentation were based on culture-dependent (isolation, cultivation, and identification), or, more recently, culture-independent (PCR-DGGE, or polymerase chain reaction denaturing gradient gel electrophoresis) methods. Using a metagenomic approach, total DNA was extracted from heap and box fermentations at different time points and from different locations (Ghana and Brazil, respectively) to generate a 16 S rDNA clone library that was sequenced. The sequencing data revealed a low bacterial diversity in the fermentation samples and were in accordance with the results obtained through culture-dependent and a second, culture-independent analysis (PCR-DGGE), suggesting that almost all bacteria involved in the fermentation process are cultivable. One exception was the identification by 16 S rDNA library sequencing of Gluconacetobacter species of acetic acid bacteria that were not detected by the two other approaches. The presence of Enterobacteriaceae related to Erwinia/Pantoea/Tatumella, as revealed by 16 S rDNA library sequencing, suggests an impact of these bacteria on fermentation.

Abundance of culturable versus viable Escherichia coli in freshwater.

Approved methods traditionally used for Escherichia coli enumeration in waters are culture-based. However, these methods can underestimate the E. coli abundance in aquatic systems because they do not take into account cells that remain viable but have lost the ability to grow in or on culture media. We investigated, in freshwater samples, the abundance of (i) culturable E. coli, enumerated by the most probable number microplate method and (ii) viable E. coli, estimated using a procedure called DVC-FISH, which couples fluorescent in situ hybridization (FISH) and a viability testing technique (direct viable count (DVC)). The ratio of culturable to viable E. coli was close to 1 in highly contaminated waters (samples with a high concentration of culturable E. coli), but decreased drastically for weakly contaminated samples. This indicates a large fraction of viable but nonculturable (VBNC) E. coli in the latter samples. Microcosm experiments showed that some environmental factors, such as nutrient scarcity and solar irradiation, could lead to the presence of a high proportion of VBNC E. coli.

Partitioning and fate of particle-associated E. coli in river waters.

Attachment of fecal bacteria to suspended matter in the water column has important implications for its fate in rivers. We examined the part of Escherichia coli (E. coli) associated with suspended matter in natural river water samples, using a combination of 5-microm filtration and beta-D-glucuronidase (GLUase) assay to estimate the E. coli abundance and attachment. We observed that the fraction of particle-associated E. coli was positively correlated with suspended matter concentration. The settling rate of particle-associated E. coli was found to be positively correlated with suspended matter concentration for samples with suspended matter content lower than 50 mg/L. For samples with higher suspended matter concentration, the settling rate was quite constant (0.066 m/h, on average). In batch experiments using river waters, we observed that free E. coli had a decay rate approximately 2 times higher than particle-associated E. coli. This information can be used to improve the models on the fate of E. coli in rivers.

Faecal bacterial indicators removal in various wastewater treatment plants located in Almendares River watershed (Cuba).

The Almendares River, located in Havana city, receives the wastewaters of more than 200,000 inhabitants. The high abundance of faecal bacterial indicators (FBIs) in the downstream stretch of the river reflects the very poor microbiological water quality. In this zone, the Almendares water is used for irrigation of urban agriculture and recreational activities although the microbiological standards for these uses are not met. Improvement of wastewater treatment is absolutely required to protect the population against health risk. This paper compares the removal of FBIs in three wastewater treatment plants (WWTPs) located in this watershed: a conventional facility using trickling filters, a constructed wetland (CW) and a solar aquatic system (SAS). The results indicate better removal efficiency in the two natural systems (CW and SAS) for all the measured parameters (suspended matters, biological oxygen demand, total coliforms, E. coli and enterococci). Removals of the FBIs were around two log units higher in both natural systems than in the conventional one. A longitudinal profile of the microbiological quality of the river illustrates the negative impact of the large conventional WWTP. This case study confirms the usefulness of small and natural WWTPs for tropical developing countries, even in urban and periurban areas.

Comparison of culturable fecal coliforms and Escherichia coli enumeration in freshwaters.

Fecal coliforms (FC) counts were compared with Escherichia coli counts in differently contaminated freshwater samples (n = 166). FC were enumerated by plate count on triphenyl 2,3,5-tetrazolium chloride Tergitol medium. Escherichia coli were enumerated by the most probable number microplate method based on the detection of glucuronidase activity. FC and E. coli counts were highly correlated; an average E. coli/FC ratio equal to 0.77 was found, meaning that on average, 77% of FC were E. coli. Knowing the E. coli/FC ratio allows us to convert the historical microbiological quality data expressed in FC counts into E. coli abundance and thus to compare with present and future monitoring data that are (or will be) based on E. coli enumeration.

Fecal bacteria in the rivers of the Seine drainage network (France): sources, fate and modelling.

The Seine river watershed (France) is a deeply anthropogenically impacted area, due to the high population density, intense industrial activities and intensive agriculture. The water quality and ecological functioning of the different rivers of the Seine drainage network have been extensively studied during the last fifteen years within the framework of a large French multidisciplinary scientific program (PIREN Seine program). This paper presents a synthesis of the main data gained in the scope of this program concerning the microbiological water contamination of the rivers of the Seine drainage network. The more common indicator of fecal contamination (fecal coliforms) was mainly used; some complementary works used E. coli and intestinal enterococci as alternative fecal indicators. Point sources (outfall of wastewater treatment plants) and non point sources (surface runoff and soil leaching) of fecal pollution to the rivers of the watershed were quantified. Results showed that, at the scale of a large urbanised watershed as the Seine basin, the input of fecal micro-organisms by non-point sources is much lower than the inputs by point sources. However, the local impact of diffuse non-human sources (especially surface runoff of pastured fields) can be of major importance on the microbiological quality of small headwater rivers. Fecal contamination of the main rivers of the Seine watershed (Seine, Marne, Oise rivers) was studied showing high level of microbiological pollution when compared to European guidelines for bathing waters. The strong negative impact of treated wastewater effluents outfall on the microbiological quality of receiving rivers was observed in different areas of the watershed. Once released in rivers, culturable fecal bacteria disappeared relatively rapidly due to mortality (protozoan grazing, lysis) or loss of culturability induced by stress conditions (sunlight effect, nutrient concentration, temperature). Mortality rates of E. coli were studied in different types of rivers within the watershed showing, in summer conditions, no major difference in the mortality rates in small and large rivers. As a result of these studies, a module describing the dynamics of fecal bacteria has been developed and embedded within a hydro-ecological model describing the functioning of the rivers of the whole watershed (the SENEQUE model). Once validated, such a model can be used for testing predictive scenarios and thus can be a very useful tool for the management of microbiological water quality at the scale of the whole basin.

Enumeration of viable E. coli in rivers and wastewaters by fluorescent in situ hybridization.

A combination of direct viable count (DVC) and fluorescent in situ hybridization (FISH) procedures was used to enumerate viable Escherichia coli in river waters and wastewaters. A probe specific for the 16S rRNA of E. coli labeled with the CY3 dye was used; enumeration of hybridized cells was performed by epifluorescence microscopy. Data showed that the method was able to accurately enumerate a minimum of 3000 viable E. coli among a large number of non-fecal bacteria. When applied to river water and wastewater samples, the DVC-FISH method gave systematically higher E. coli counts than a reference culture-based method (miniaturized MPN method). The ratio between both counts (DVC-FISH/MPN) increased with decreasing abundance of culturable E. coli indicating that the proportion of viable but non-culturable (VBNC) E. coli (detectable by the DVC-FISH procedure and not by a culture-based method) was higher in low contaminated environments. We hypothesized that the more stressing conditions, i.e. nutritional stress and sunlight effect, met in low contaminated environments were responsible for the larger fraction of VBNC E. coli. A survival experiment, in which sterile mineral water was inoculated with a pure E. coli strain and incubated, confirmed that stressing conditions induced the apparition of non-culturable E. coli detectable by the DVC-FISH procedure. The analysis of the E. coli concentration along a Seine river longitudinal profile downstream a large input of fecal bacteria by a WWTP outfall showed an increasing fraction of VBNC E. coli with increasing residence time of the E. coli in the river after release. These data suggest that the DVC-FISH method is useful tool to analyze the dynamics of fecal bacteria in river water.