Influence of backwashing on the microbial community in a biofilm developed on biological activated carbon used in a drinking water treatment plant.

The influence of backwashing on the biofilm community developed on biological activated carbon (BAC) used in a drinking water treatment plant was investigated by means of bacterial cell enumeration and terminal-restriction fragment length polymorphism (T-RFLP) fingerprinting analysis of bacterial and eukaryotic ribosomal RNA genes (rDNA). After backwashing, the attached bacterial abundance in the top layer of the BAC bed decreased to 64% of that before backwashing. The community level changes caused by backwashing were examined through the T-RFLP profiles. In the bacterial 16S rDNA analysis, the relative abundances of some terminal-restriction fragments (T-RFs) including the Planctomycetes-derived fragment increased; however, the relative abundances of some T-RFs including the Betaproteobacteria-derived fragments decreased. In the eukaryotic 18S rDNA analysis, the relative abundances of some T-RFs including the protozoan Cercozoa-derived fragments increased; however, the relative abundances of some T-RFs including the metazoan Chaetonotus- and Paratripyla-derived fragments decreased. The T-RFLP analysis suggests that backwashing can cause changes in the relative compositions of microorganisms in a BAC biofilm in the top layer of the bed.     

Determination of the microbial diversity of anaerobic-aerobic activated sludge by a novel molecular biological technique

Sequential anaerobic-aerobic batch reactors were maintained on acetate/peptone and two different P/total organic carbon ratios that select for microbial communities enriched for either glycogen-accumulating organisms (GAO) or polyphosphate-accumulating organism (PAO). The community profiles of the eubacterial population and gram-positive high G-C bacteria (HGC) were characterized and compared by determining the terminal restriction fragment length polymorphisms (T-RFLP) of 16S rDNA. The Hhal+Rsal digested 5′ T-RFLP patterns of the eubacterial 16S rDNA amplified from the GAO- and PAOenriched communities were made up with 12 and 14 rank-abundant fragments (i.e., ribotypes), respectively. Among those ribotypes detected in the GAO-enriched community, only seven were found in the PAO-enriched community. The HGC group could only account for no more than 6% and 17% of the eubacterial 16S rDNA amplified from the GAO- and PAO-enriched communities, respectively. Within the HGC community, at least 16 and 10 rank-abundant ribotypes were observed m the Mspl digested T-RFLP patterns of GAO- and PAO-enriched communities, respectively. Among those HGC ribotypes observed in both communities, only five were in common. These indicate that the enrichment processes leading to the establishment of GAO- and PAO-specific communities caused the dramatic difference and complexity in the microbial population.     

Analysis of microbial community structure and diversity in the thermophilic anaerobic digestion of waste activated sludge.

The microbial community structures in the thermophilic anaerobic digestion (TAD) of waste activated sludge (WAS) and WAS were analyzed with molecular biological techniques including real-time polymerase chain reaction (PCR) and cloning. The microbial community of TAD had less diversity than that of WAS, and the sequences obtained in WAS were not present in TAD by the cloning analysis. In the TAD bacterial clone library, 97.5% of total clones were affiliated with the phylum Firmicutes and 73.1% with the genus Coprothermobacter. Real-time PCR and cloning analysis revealed that the number of Methanosarcina thermophila, which is an acetoclastic methanogen, is larger than that of Methanoculleus thermophilus, which is a hydrogenotrophic methanogen, in terms of the numbers of copies of 16S ribosomal DNA (rDNA).     

Comparison of nitrification performance and microbial community between submerged membrane bioreactor and conventional activated sludge system.

A submerged membrane bioreactor (SMBR) and a conventional activated sludge system (CAS) were compared in parallel over a period of more than 260 days on treating synthetic ammonia-bearing inorganic wastewater without sludge purge under decreased hydraulic retention times (HRTs). Conversion of NH4(+)-N to NO3(-)-N was achieved with an efficiency of over 98% at an HRT > or = 10 h in the SMBR, while similar performance was obtained at an HRT > or = 20 h in the CAS. Denaturing gradient gel electrophoresis (DGGE) of polymerase chain reaction (PCR) amplified 16S rDNA was used to monitor variations of community structures in the two systems. With the prolongation of operation, the number of DGGE bands in the SMBR gradually increased from the initial 11 bands to the final 22 bands, whereas that in the CAS varied in a range between 13 and 183 Sequence analysis indicates that Nitrosomonas sp. and Nitrospira sp. were the dominating nitrification species responsible for ammonia and nitrite oxidation, respectively. Heterotrophic bacteria like Pseudomonas sp. and Flavobacteria sp. existed in both of the systems although only inorganic wastewater was fed. Substantive accumulation of extracellular polymeric substances (EPS) in the SMBR was confirmed by scanning electron microscopy and EPS analysis.     

Distribution of ammonia-oxidizing bacteria in sewage activated sludge: analysis based on 16S rDNA sequence.

This study carried out analysis of ammonia-oxidizing bacteria (AOB) communities in 12 sewage activated sludge systems standing in eight sewage treatment plants located in Tokyo. The systems were different in the treatment process configuration: anaerobic/anoxic/aerobic (A20), anaerobic/aerobic (AO), and conventional activated sludge (AS) processes. AOB communities were analyzed by sequences of 16S rDNA amplicons, which were separated by denaturing gradient gel eletrophoresis (DGGE) after specific polymerase chain reaction (PCR) amplification. The results demonstrated that low ammonium concentrations in the influents of the 12 sewage activated sludge systems resulted in the dominance of Nitrosomonas oligotropha-like sequences. Further, Nitrosomonas europaea- and Nitrosomonas cryotolerans-like sequences were recovered from only one A20 system of which the influent contained higher ammonium and chloride concentrations than those of other systems. Nitrosomonas communis-like sequences were found in every A20 and AO system, but mostly not found in every AS system. In summary, influent characteristics and treatment process configuration affected the AOB communities in the 12 sewage activated sludge systems.     

The use of 16S rDNA clone libraries to describe the microbial diversity of activated sludge communities

Clone libraries were prepared from polymerase chain reaction amplified 16S rDNAs from activated sludge community DNAs. Eight different libraries from a range of samples were prepared. From each library, up to 100 clones were examined. In some libraries, the clone inserts were grouped into operational taxonomic units (OTUs) by restriction enzyme analysis (REA). Then, either the clones or representatives of OTUs were partially sequenced using either 27f or 530f conserved bacterial primers. The sequence data was phylogenettcally analysed to group the clones and the method currently gives the best insight into the activated sludge microbial community biodiversity. The method for clone library production is described and the pros and cons of the procedure are outlined. In summary, the use of clone libraries has resulted in the discovery of unimagined biodiversity in activated sludge. The abundance of some unpredicted bacterial groups (e.g. beta subclass Proteobactena) and the paucity of expected ones (e.g. Acinetobacter) highlights the inadequacy of traditional culture dependent methods that rely on sample dilution and spread plate inoculation.     

Dynamics of dewaterability and bacterial populations in activated sludge

Relationships of bacterial populations and extracellular polymer substances (EPS) to dewaterability of activated sludge were studied on three laboratory-scale activated sludge reactors fed with synthetic wastewater. Dewaterability of activated sludge was evaluated by a novel method developed by the authors, in which small amount of sludge was centrifugally dewatered, and its water content was measured. Bacterial populations during the reactor operation were analyzed by the polymerase chain reaction/terminal-restriction fragment length polymorphism (PCR/T-RFLP) targeted at a partial 16S rRNA gene. Extracellular polymeric substances (EPS) were extracted using cation exchange resin (CER), and polysaccharides and total protein in EPS were determined. Some of the dominant terminal-restriction fragments (T-RFs) were observed to have significant relationships with dewaterability of sludge, and it was suggested that bacterial species corresponding to those peaks significantly affected dewaterability. On the other hand, significant relationships were not found between EPS concentration and dewaterability of sludge.     

Genetic diversity and population dynamics of Hyphomicrobium spp. in a sewage treatment plant and its receiving lake

Hyphomicrobium spp. were counted and isolated for 12 months in a sewage treatment plant with a combination of simultaneous and intermittent nitrification and denitriftcation using Most-Probable-Number methods. Genomic DNA of these hyphomicrobia was investigated by Southern or dot blot hybridizations with gene probes specific for genes of dissimilatory nitrate reduction (nitrate reductase, narG; cytochrome c,d-containing nitrite reductase, nirS; Cu-containing nitrite reductase, nirK; nitrous oxide reductase, nosZ), nitrification (ammonia monooxygenase, antoA), and N₂-fixation (nitrogenase, nifH). In particular, the Hyphomicrobium DNA/DNA-hybridization group HG 27 constituted one of the dominant denitrifying Hyphomicrobium populations in the activated sludge of this sewage treatment plant. A species-specific gene probe (Hvu-1) for HG 27 was generated from a transposon Tn5-132 insertion mutant defective in methanol oxidation using the inverse polymerase chain reaction. With this probe the abundance of this group in activated sludge of the sewage treatment plant and its receiving lake was determined as a subfraction of the total cultivable hyphomicrobia. Fragments of the mxaF gene encoding for the α-subunit of the methanol dehydrogenase of Hyphomicrobium spp, were amplified by PCR and analysed by denaturing gradient gel electrophoresis (DGGE). The DGGE analysis pattern showed a substantial separation of these fragments according to their nucleic acid sequences.     

湖泊水体营养状态对沉积物好氧氨氧化菌异质性的影响

采用荧光定量PCR(real-time fluorescence quantitative polymerase chain reaction)和末端限制性片段长度多态性(terminal restriction fragment length polymorphism,T-RFLP)方法,对江苏7个湖泊沉积物中好氧氨氧化微生物进行分析,研究湖泊水体营养状态对沉积物好氧氨氧化微生物空间异质性的影响。综合营养指数分析结果表明,23个采样点中,61%的湖区为中营养状态,39%的湖区为轻度富营养状态。荧光定量PCR分析结果显示,每克底泥中氨氧化古菌(ammonia-oxidizing archaea,AOA)和氨氧化细菌(ammonia-oxidizing bacteria,AOB)amo A基因拷贝数虽然分别从中营养湖泊的3.91×10~6和3.82×10~6上升到轻度富营养湖泊的1.30×107和6.07×10~6,但湖泊水体营养状态并未显著影响沉积物AOA和AOB的丰度。T-RFLP分析结果表明,湖泊水体营养状态对AOA和AOB的优势种属及群落多样性也未产生显著影响。典范对应分析结果表明,湖泊水体营养盐浓度能解释AOA的群落结构差异的56.3%,而仅能解释27.2%的AOB的群落结构差异,TN和NO-3-N浓度是影响沉积物AOA分布异质性的主要环境因子,湖泊水体营养盐浓度比综合营养状态指数更能影响AOA和AOB的群落结构组成。     

Microbial community structure of activated sludge during aerobic granulation in an annular gap bioreactor.

A novel annular gap reactor was designed to create a controlled shear environment in which aerobic granular sludge could be developed. The bacterial and eukaryal community structures during two aerobic granular sludge experiments were tracked using denaturing gradient gel electrophoresis (DGGE). The first granule cultivation experiment, using an organic loading rate of 1.6 kg/m3d COD, resulted in biomass that was dominated by filamentous bacteria and Zoogloea ramigera colonies. A second experiment with a higher organic loading rate of 6 kg/m3d COD developed a granule-like morphology but was ultimately dominated by filamentous fungi. Species identification via DGGE band purification and DNA sequencing closely matched the observed sludge morphology and behavior.     

Evaluation of the impacts of model-based operation of SBRs on activated sludge microbial community.

Impact of model-based operation of nutrient removing SBRs on the stability of activated sludge population was studied in this contribution. The optimal operation scenario found by the systematic model-based optimisation protocol of Sin et al. (Wat. Sci. Tech., 2004, 50(10), 97-105) was applied to a pilot-scale SBR and observed to considerably improve the nutrient removal efficiency in the system. Further, the process dynamics was observed to change under the optimal operation scenario, e.g. the nitrite route prevailed and also filamentous bulking was provoked in the SBR system. At the microbial community level as monitored by DGGE, a transient shift was observed to gradually take place parallel to the shift into the optimal operation scenario. This implies that the model-based optimisation of a nutrient removing SBR causes changes at the microbial community level. This opens future perspectives to incorporate the valuable information from the molecular monitoring of activated sludge into the model-based optimisation methodologies. In this way, it is expected that model-based optimisation approaches will better cover complex and dynamic aspects of activated sludge systems.     

Long-term evaluation of a sequential batch reactor (SBR) treating dairy wastewater for carbon removal.

Many dairy industries have been using SBR wastewater treatment plants because they allow optimal working condition to be reached. However, to take advantage of SBR capabilities, strong process automation is needed. The aim of this work is to study the factors that influence SBR performance to improve modelling and control. To better understand the whole process we studied the kinetic modelling, the carbon removal mechanism and the relation between reactor performance, aerobic heterotrophic activity and bacterial population dynamics (by terminal restriction fragment length polymorphisms of 16S rDNA, T-RFLP). The heterotrophic activity values presented high variability during some periods; however, this was not reflected on the reactor performance. As sludge health indicator, the average activity in a period was better than individual values. Although all the carbon removal mechanisms are still unclear for this process, they seemed to be influenced by non-respirometric ways (storage, biosorption, accumulation, etc.). The variability of heterotrophic activity could be correlated with the bacterial population diversity over time. Despite the high variability of the activity, a simple kinetic model (pseudo ASM1) based on apparent constant parameters was developed and calibrated. Such modellisation provided a good tool for control purposes.     

Vertical microbiological variation of a coastal aquifer in southern China

Microbial communities in a coastal aquifer in the Zhuhai region of southern China were investigated by culture-independent molecular approaches. Four 16S rRNA gene libraries of three groundwater samples from varied depths and one seawater sample were constructed and analysed by the amplified ribosomal DNA restriction analysis technique (ARDRA). The phylogenetic analysis indicated that the 16S rDNA of clones presenting dominant ARDRA patterns were most similar to Proteobacteria, Bacteroidetes, Actinobacteria, Planctomycetes, and candidate divisions OPx (such as OP3, OP8, and OP11). In samples extracted from wells of 5-, 20-, and 35-m depth (i.e., D1, D5, and D6) Proteobacteria made up 32.3, 34.3, and 46.7% of the microbial communities, respectively. The same samples from D1, D5, and D6 also consisted of 5.0, 11.2, and 6.5% Bacteroidetes and 5.4, 6.6, and 7.8% Actinobacteria, respectively. In contrast, the seawater clone library had a predominant number of Proteobacteria (32.8%), while Bacteroidetes and Planctomycetes both accounted for 9.3%. Total microbial diversity remained relatively constant over the top layer to a depth of approximately 35 m, significant community vertical and horizontal (seawater-groundwater) shifts were observed for certain bacterial populations.     

Microbial communities and their interactions in biofilm systems: an overview.

Several important advances have been made in the study of biofilm microbial populations relating to their spatial structure (or architecture), their community structure, and their dependence on physicochemical parameters. With the knowledge that hydrodynamic forces influence biofilm architecture came the realization that metabolic processes may be enhanced if certain spatial structures can be forced. An example is the extent of plasmid-mediated horizontal gene transfer in biofilms. Recent in situ work in defined model systems has shown that the biofilm architecture plays a role for genetic transfer by bacterial conjugation in determining how far the donor cells can penetrate the biofilm. Open channels and pores allow for more efficient donor transport and hence more frequent cell collisions leading to rapid spread of the genes by horizontal gene transfer. Such insight into the physical environment of biofilms can be utilized for bioenhancement of catabolic processes by introduction of mobile genetic elements into an existing microbial community. If the donor organisms themselves persist, bioaugmentation can lead to successful establishment of newly introduced species and may be a more successful strategy than biostimulation (the addition of nutrients or specific carbon sources to stimulate the authochthonous population) as shown for an enrichment culture of nitrifying bacteria added to rotating disk biofilm reactors using fluorescent in situ hybridization (FISH) and microelectrode measurements of NH4+, NO2-, NO3-, and O2. However, few studies have been carried out on full-scale systems. Bioaugmentation and bioenhancement are most successful if a constant selective pressure can be maintained favoring the promulgation of the added enrichment culture. Overall, knowledge gain about microbial community interactions in biofilms continues to be driven by the availability of methods for the rapid analysis of microbial communities and their activities. Molecular tools can be grouped into those suitable for ex situ and in situ community analysis. Non-spatial community analysis, in the sense of assessing changes in microbial populations as a function of time or environmental conditions, relies on general fingerprinting methods, like DGGE and T-RFLP, performed on nucleic acids extracted from biofilm. These approaches have been most useful when combined with gene amplification, cloning and sequencing to assemble a phylogenetic inventory of microbial species. It is expected that the use of oligonucleotide microarrays will greatly facilitate the analysis of microbial communities and their activities in biofilms. Structure-activity relationships can be explored using incorporation of 13C-labeled substrates into microbial DNA and RNA to identify metabolically active community members. Finally, based on the DNA sequences in a biofilm, FISH probes can be designed to verify the abundance and spatial location of microbial community members. This in turn allows for in situ structure/function analysis when FISH is combined with microsensors, microautoradiography, and confocal laser scanning microscopy with advanced image analysis.     

Investigation of microbial communities on reverse osmosis membranes used for process water production.

In the present study, the diversity and the phylogenetic affiliation of bacteria in a biofouling layer on reverse osmosis (RO) membranes were determined. Fresh surface water was used as a feed in a membrane-based water purification process. Total DNA was extracted from attached cells from feed spacer, RO membrane and product spacer. Universal primers were used to amplify the bacterial 16S rRNA genes. The biofilm community was analysed by 16S rRNA-gene-targeted denaturing gradient gel electrophoresis (DGGE) and the phylogenetic affiliation was determined by sequence analyses of individual 16S rDNA clones. Using this approach, we found that five distinct bacterial genotypes (Sphingomonas, Beta proteobacterium, Flavobacterium, Nitrosomonas and Sphingobacterium) were dominant genera on surfaces of fouled RO membranes. Moreover, the finding that all five "key players" could be recovered from the cartridge filters of this RO system, which cartridge filters are positioned before the RO membrane, together with literature information where these bacteria are normally encountered, suggests that these microorganisms originate from the feed water rather than from the RO system itself, and represent the fresh water bacteria present in the feed water, despite the fact that the feed water passes an ultrafiltration (UF) membrane (pore size approximately 40 nm), which is able to remove microorganisms to a large extent.     

Real-time PCR quantification of nitrite reductase (nirS) genes in a nitrogen removing fluidized bed reactor.

Molecular approaches were applied to identify and enumerate denitrifying bacteria subsisting in a fluidized bed reactor (FBR). The FBR was continuously operated as a unit for the removal of nitrogen from the effluents of domestic sewage treatment plant, with an additional supply of methanol as a carbon source. By denaturing gradient gel electrophoresis (DGGE) and sequence analysis of 16S ribosomal RNA genes, Thauera group was found to be dominant among the denitrifying bacteria in the FBR sludge. Oligonucleotide probe THA155 for fluorescence in situ hybridization (FISH) was newly designed for specifically targeting the Thauera group. However, the THA155 signal obtained from the sludge was only 0.9-5.7% of the DAPI-stained total cells. The real-time polymerase chain reaction (PCR) targeting the sequences of nitrite reductase (NIR) gene, a key enzyme of denitrification processes, was performed to quantify the cells of denitrifying bacteria cells including the Thauera group in FBR sludge. An excellent correlation was obtained between the numbers of nirS genes and the activity of denitrifiers in the FBR sludge.     

Performance and microbial ecology of the hybrid membrane biofilm process for concurrent nitrification and denitrification of wastewater.

We report on a novel process for total nitrogen (TN) removal, the hybrid membrane biofilm process (HMBP). The HMBP uses air-supplying hollow-fibre membranes inside an activated sludge tank, with suppressed aeration, to allow concurrent nitrification and denitrification. We hypothesised that a nitrifying biofilm would form on the membranes, and that the low bulk-liquid BOD concentrations would encourage heterotrophic denitrifying bacteria to grow in suspension. A nitrifying biofilm was initially established by supplying an influent ammonia concentration of 20 mgN/L. Subsequently, 120 mg/L acetate was added to the influent as BOD. With a bulk-liquid SRT of only 5 days, nitrification rates were 0.85 gN/m(2) per day and the TN removal reached 75%. The biofilm thickness was approximately 500 lim. We used DGGE to obtain a microbial community fingerprint of suspended and attached growth, and prepared a clone library. The DGGE results, along with the clone library and operating data, suggest that nitrifying bacteria were primarily attached to the membranes, while heterotrophic bacteria were predominant in the bulk liquid. Our results demonstrate that the HMBP is effective for TN removal, achieving high levels of nitrification with a low bulk-liquid SRT and concurrently denitrifying with BOD as the sole electron donor.     

Activated sludge foaming: The true extent of actinomycete diversity

Isolates from activated sludge foam were provisionally assigned to the genera Gordona and Tsukamurella on the basis of colony morphology and pigmentation. Representatives of the first group were compared with marker strains of validly described species of Gordona by using Curie-point pyrolysis mass spectrometry. Most of the isolates fell into a number of taxa which were equated with groups of marker strains which corresponded to the known Gordona species. In a corresponding experiment the activated sludge isolates, which were provisionally labelled –Tsukamurella spumae– formed a group which was well separated from the marker strains of the genus. Representatives of the isolates were found to have chemical properties consistent with their classification in the genus Tsukamurella. 16S rDNA sequence data, when taken with previous DNA:DNA relatedness results, suggest that Tsukamurella paurometabola is heterogeneous. Similarly, the sequence data provide some evidence that “Tsukamurella spumae” may merit recognition as a novel species.     

Revealing microbial community structures in large- and small-scale activated sludge systems by barcoded pyrosequencing of 16S rRNA gene

The diversity of bacterial groups in activated sludge from large- and small-scale wastewater treatment plants was explored by barcoded pyrosequencing of 16S rRNA gene. Activated sludge samples (three small and 17 large scale) were collected from 12 wastewater treatment plants to clarify precise taxonomy and relative abundances. DNA was extracted, and amplified by 4 base barcoded 27f/519r primer set. The 454 Titanium (Roche) pyrosequences were obtained and analyses performed by Quantitative Insight Into Microbial Ecology (QIIME) with around 100,000 reads. Sequence statistics were computed, while constructing a phylogenetic tree and heatmap. Computed results explained total microbial diversity at phylum and class level and resolution was further extended to Operational Taxonomic Unit (OTU) based taxonomic assignment for investigating community distribution based on individual sample. Composition of sequence reads were compared and microbial community structures for large- and small-scale treatment plants were identified as major phyla (Proteobacteria and Bacteroidetes) and classes (Betaproteobacteria and Bacteroidetes). Also, family level breakdowns were explained and differences in family Nitrospiraceae and phylum Actinobacteria found at their species level were also illustrated. Thus, the pyrosequencing method provides high resolution insight into microbial community structures in activated sludge that might have been unnoticed with conventional approaches.     

Syntrophic acetate oxidation in two-phase (acid-methane) anaerobic digesters

The microbial processes involved in two-phase anaerobic digestion were investigated by operating a laboratory-scale acid-phase (AP) reactor and analyzing two full-scale, two-phase anaerobic digesters operated under mesophilic (35 °C) conditions. The digesters received a blend of primary sludge and waste activated sludge (WAS). Methane levels of 20% in the laboratory-scale reactor indicated the presence of methanogenic activity in the AP. A phylogenetic analysis of an archaeal 16S rRNA gene clone library of one of the full-scale AP digesters showed that 82% and 5% of the clones were affiliated with the orders Methanobacteriales and Methanosarcinales, respectively. These results indicate that substantial levels of aceticlastic methanogens (order Methanosarcinales) were not maintained at the low solids retention times and acidic conditions (pH 5.2-5.5) of the AP, and that methanogenesis was carried out by hydrogen-utilizing methanogens of the order Methanobacteriales. Approximately 43, 31, and 9% of the archaeal clones from the methanogenic phase (MP) digester were affiliated with the orders Methanosarcinales, Methanomicrobiales, and Methanobacteriales, respectively. A phylogenetic analysis of a bacterial 16S rRNA gene clone library suggested the presence of acetate-oxidizing bacteria (close relatives of Thermacetogenium phaeum, 'Syntrophaceticus schinkii,' and Clostridium ultunense). The high abundance of hydrogen consuming methanogens and the presence of known acetate-oxidizing bacteria suggest that acetate utilization by acetate oxidizing bacteria in syntrophic interaction with hydrogen-utilizing methanogens was an important pathway in the second-stage of the two-phase digestion, which was operated at high ammonium-N concentrations (1.0 and 1.4 g/L). A modified version of the IWA Anaerobic Digestion Model No. 1 (ADM1) with extensions for syntrophic acetate oxidation and weak-acid inhibition adequately described the dynamic profiles of volatile acid production/degradation and methane generation observed in the laboratory-scale AP reactor. The model was validated with historical data from the full-scale digesters.