Table 1.
Characteristics of the feeding sludge (FS) and of the outlet sludge (eco 1, 2 or 3) from the mesophilic bioreactors inoculated with the three microbial communities at steady state.
Fig 1.
Boxplot of the anaerobic pollutants removal at steady state according to the inocula.
The removals for eco1 (white), eco2 (grey) and eco3 (black) were measured at steady state. The LMW-PAH, the MMW-PAH and the HMW-PAH are defined in Table 1. The box shows 75th percentiles (top line), 50th percentile (middle line) and 25th percentile (bottom line).
Fig 2.
Percentage of PAH removal over the time (expressed as Hydraulic Retention Times) according to the inoculum.
Phenanthrene is a representant of the LMW-PAH and Dibenzo(a,h)Anthracene of the HMW-PAH. The performance of the eco1, eco2 and eco3 are respectively in white, grey and black.
Fig 3.
Correlation between DM (black square) or VS (grey triangle) removals and PAH removal.
Fig 4.
Average enumeration according to the inoculum origin (eco1 in white, eco2 in grey, eco3 in black) of Bacteria (left) and Archaea (right) by q-PCR.
The x axis represents the hydraulic retention time. Standard errors are displayed. HRT4 and HRT 5 represent the steady state based on the reactors performances.
Fig 5.
Archaea / Bacteria ratio depending on the methane production at steady state.
The ratio is either calculated with the q-PCR or the 454-pyrosequencing data.
Fig 6.
(A) Influence of the inoculum origin (eco1, eco2, eco3) and time (HRT0, HRT3, HRT4 and HRT5) on the difference in genetic structure of bacterial (on the left) and archaeal (on the right) communities through a PCA. The percentage variation explained by each principal component and the principal component scores of the sample are plotted on their respective axes. Changes over time are graphically represented by arrows.(B) Principal Component Analysis (PCA) biplot of bacterial (on the left) and archaeal (on the right) communities. PCA displayed 65.4% of variance for bacterial CE-SSCP fingerprints and 74.3% of variance for archaeal CE-SSCP fingerprints. Only significant correlations with operational characteristics were presented as arrows.
Fig 7.
Principal Component Analysis (PCA) biplot at steady state of microbial communities (eco1, eco2, eco3) obtained with the CE-SSCP data (on the left) and with the sequencing data (on the right).
PCA displayed 92% and 92.3% of variance for microbial fingerprints and sequencing respectively. Most discriminant CE-SSCP peaks (on the left) and most discriminant species (on the right) were highlighted as arrows that were directed according to their explanatory outputs. 1, 2, 3: Clostridium sp. (88%), 4: Parabacteroides sp. (87%), 5: Dysgonomas sp. (90%), 6: Parabacteroides sp. (86%), 7: Clostridium sp (93%), 8: Anaerobaculum mobile (99%), 9: Thiohalomonas sp. (82%), 10: Op9 (99%), 11: Pseudomonas stutzeri (99%).