Agarose hydrogels as EPS models.

For investigating the influence of extracellular polymeric substances (EPSs) on biofilm properties, artificial models of defined composition and structure can be very helpful. Different immobilised cell systems can be useful in the fitting of experimental results. Two different types of of artificial EPS matrix model were developed earlier. Homogeneous agarose beads (50-500 microns diameter) and porous beads (260 microns mean diameter) containing pores with diameters from 10 to 80 microns (28 microns on average), allowed the embedding of cells, particles and typical EPS matrix components such as proteins and polysaccharides. In this paper, some physico-chemical properties of the artificial EPS matrices were described and compared with results known from natural EPS. The stability of the artificial matrix polymer against solution in the aqueous medium was studied in batch experiments. The water binding and water retaining capabilities of the EPS models were investigated by drying and re-swelling experiments. The simulation of protective effects of the artificial EPS matrix against toxic substances like biocides in comparison to such known protective effects of the EPS of native biofilms were proved by the application of sodium hypochlorite (0.5 mg/l, 30 min) and subsequent microscopic investigation of the cell population after LIVE/DEAD staining (Molecular Probes).

A new method for extraction of extracellular polymeric substances from biofilms and activated sludge suitable for direct quantification of sorbed metals.

A method for extraction of extracellular polymeric substances (EPSs) with a dicyclohexyl-18-crown-6 ether was developed to determine levels of organic and inorganic contaminants sorbed to EPS. The crown ether selectively binds alkaline and alkaline earth metals but not heavy metals. The effectiveness of the extraction procedure was higher than that of 2 other methods tested and comparable with that of a method based on a cation exchange resin. On average it was possible to extract 20% of the TOC, 12% of the total protein content, and 4% of the total carbohydrate content of sludge or biofilm biomass. Metal sorption studies in activated sludge showed no influence of exposure time on the fractionation of metals within the biomass. Metals sorbed mostly to cellular material. In biofilms 12.2% of the cadmium and 9.1% of the zinc added was found in the EPS. In activated sludge EPS contained only 2.9% zinc. The distribution of metals within the biomass was dose dependent. The percentage of metals found in EPS decreased with increasing metal concentration. This indicates a higher affinity of metals for cellular binding sites. Time course experiments in a rotating biofilm annular reactor, which consisted of an external cylinder with removable slides and an internal solid drum, revealed a gradual change in zinc concentration associated with EPS, although the total zinc concentration in the biomass remained constant. Concurrently, the amount of extractable EPS decreased. This was a consequence of a microbial population shift, with bacterial counts decreasing and algal and fungal biomass increasing. Using confocal laser scanning microscopy and the fluorescent metal complexing agent Newport Green for in situ detection of zinc it was shown that metals were bound to algae and fungi in the latter part of the experiment. The biofilm became more and more heterogeneous coinciding with a decrease in EPS. To summarize, the observed sorption behavior of metals cannot be explained with the conventional paradigm of EPS as hydrophilic gel. Obviously, different binding mechanisms must be invoked to explain the role of EPS in the sorption and removal of toxic substances in activated sludge and biofilm systems. It is important to consider the microbial population to understand differences in sorption in different matrices.

Artificial biofilm model--a useful tool for biofilm research.

For biofilm studies, artificial models can be very helpful in studying processes in hydrogels of defined composition and structure. Two different types of artificial biofilm models were developed. Homogeneous agarose beads (50-500 microm diameter) and porous beads (260 microm mean diameter) containing pores with diameters from 10 to 80 microm (28 microm on average) allowed the embedding of cells, particles and typical biofilm matrix components such as proteins and polysaccharides. The characterisation of the matrix structures and of the distribution of microorganisms was performed by confocal laser scanning microscopy. The physiological condition of the embedded bacteria was examined by redox activity (CTC-assay) and membrane integrity (Molecular Probes LIVE/DEAD-Kit). Approximately 35% of the immobilised cells (Pseudomonas aeruginosa SG81) were damaged due to the elevated temperature required for the embedding process. It was shown that the surviving cells were able to multiply when provided with nutrients. In the case of homogeneous agarose beads, cell growth only occurred near the bead surface, while substrate limitation prevented growth of more deeply embedded cells. In the porous hydrogel, cell division was observed across the entire matrix due to better mass transport. It could be shown that embedding in the artificial gel matrix provided protection of immobilized cells against toxic substances such as sodium hypochlorite (0.5 mg/l, 30 min) in comparison to suspended cells, as observed in other immobilized systems. Thus, the model is suited to simulate important biofilm matrix properties.

Enzymatic activity in the activated-sludge floc matrix.

The enzymatic activity of activated sludge was investigated with special emphasis on the localization of the enzymes in the sludge floc matrix. Activated sludge from an advanced activated-sludge treatment plant, performing biological N and P removal, was used. An enzymatic fingerprint was established using a panel of six different enzymes. The fingerprint revealed peptidase as the most dominating specific enzyme tested. By monitoring sludge bulk enzymatic activity over a 3-month period using fluorescein diacetate as an enzyme substrate, considerable variations in activity were observed even over short periods (a few days). The variation in esterase activity was to some extent correlated to the presence of humic compounds in the sludge, but not to the sludge protein content. Comparison of full sludge enzyme activity to the activity of a batch-grown sludge culture indicated that enzymes accumulated in sludge flocs. A large proportion of the exoenzymes were immobilized in the sludge by adsorption in the extracellular polymeric substances (EPS) matrix. This was demonstrated by extraction of EPS from the activated sludge using cation exchange. Contemporary to the release of EPS a very large fraction of the exoenzymes was released into the water. This showed that the exoenzymes should be considered to be an integrated part of the EPS matrix rather than as direct indicators of the microbial activity or biomass.

Biofilm parameters influencing biocide efficacy.

The influence of biofilm areal cell density, species composition, and the presence of abiotic particles on the disinfection and removal of bacterial biofilms by monochloramine was investigated. Mono- and binary population biofilms of Pseudomonas aeruginosa and Klebsiella pneumoniae were grown on stainless-steel slides in a continuous flow annular reactor. Biofilms were treated in the reactor with a pulse/step dose of 4 mg/L monochloramine for 2 h. Biofilm samples were disaggregated and assayed for colony formation on R2A agar and for total cell numbers by acridine orange direct counts. These data were used to determine apparent first order rate coefficients for the processes of disinfection and detachment. Disinfection rate coefficients exceeded detachment rate coefficients by as much as an order of magnitude and the two coefficients were poorly correlated (r = 0.272). The overall decay rate coefficient (disinfection plus detachment) depended strongly on the initial biofilm areal cell density. It displayed a parabolic dependence on cell density with a maximum near 10(8) cfu/cm(2). This result points to multiple factors influencing biofilm susceptibility to antimicrobial challenge. Decay rates of K. pneumoniae measured in binary population biofilms were comparable with those measured in monopopulation biofilms (p = 0.61). P. aeruginosa decayed more slowly in biofilm dominated by K. pneumoniae (p = 0.028), indicating some interaction between species. The presence of kaolin and calcium carbonate particles in the biofilm reduced disinfection efficacy.

Comparison of respiratory activity and culturability during monochloramine disinfection of binary population biofilms.

Biofilm bacteria challenged with monochloramine retained significant respiratory activity, even though they could not be cultured on agar plates. Microbial colony counts on agar media declined by approximately 99.9% after 1 h of disinfection, whereas the number of bacteria stained by a fluorescent redox dye experienced a 93% reduction. Integrated measures of biofilm respiratory activity, including net oxygen and glucose utilization rates, showed only a 10 to 15% reduction. In this biofilm system, measures of microbial respiratory activity and culturability yielded widely differing estimates of biocide efficacy.

Kinetic analysis of microbial sulfate reduction by desulfovibrio desulfuricans in an anaerobic upflow porous media biofilm reactor.

An anaerobic upflow porous media biofilm reactor was designed to study the kinetics and stoichiometry of hydrogen sulfide production by the sulfate-reducing bacterium (SRB) Desulfovibrio desulfuricans (ATCC 5575) as the first step for the modeling and control of formation souring (H(2)S) in oil field porous media. The reactor was a packed bed (50 x 5.5 cm) tubular reactor. Sea sand (140 to 375 mum) was used as the porous media. The initial indication of souring was the appearance of well-separated black spots (precipitates of iron sulfide) in the sand bed. The blackened zones expanded radially and upward through the column. New spots also appeared and expanded into the cone shapes. Lactate (substrate) was depleted and hydrogen sulfide appeared in the effluent.Analysis of the pseudo-steady state column shows that there were concentration gradients for lactate and hydrogen sulfide along the column. The results indicate that most of the lactate was consumed at the front part of the column. Measurements of SRB biomass on the solid phase (sand) and in the liquid phase indicate that the maximum concentration of SRB biomass resided at the front part of the column while the maximum in the liquid phase occurred further downstream. The stoichiometry regarding lactate consumption and hydrogen sulfide production observed in the porous media reactor was different from that in a chemostat. After analyzing the radial dispersion coefficient for the SRB in porous media and kinetics of microbial growth, it was deduced that transport phenomena dominate the souring process in our porous media reactor system. (c) 1994 John Wiley & Sons, Inc.