Influence of biofilm composition on the resistance to detachment.

Bacillus cereus and Pseudomonas fluorescens were used to develop monoculture biofilms in a bioreactor rotating system using a stainless steel cylinder for biofilm formation. The biofilms were allowed to grow for 7 days, exposed continuously to a Reynolds number of agitation (ReA) of 2,400. Afterwards, the biofilms were characterised in terms of respiratory activity, amount of biomass, cellular density, cellular size and total and extracellular proteins and polysaccharides. The biofilm mechanical stability was assessed by sequential submission of the biofilms to increasing ReA, respectively, 4,000, 8,100, 12,100 and 16,100. The results showed that P. fluorescens biofilms were five times more active, had a higher amount of biomass, cellular density, a reduced cellular size and a four-fold higher amount of extracellular proteins and polysaccharides than B. cereus biofilms. The application of shear stress forces higher than the one under which the biofilm was formed (ReA = 2,400) caused biomass removal. The high percentage of removal occurred with the implementation of a ReA of 8,100 for both B. cereus and P. fluorescens biofilms. The total series of ReA did not give rise to total biofilm removal, as only about 76% of P. fluorescens biofilm mass and 53% of B. cereus biofilm mass were detached from the cylinders. This latter result evidences that B. cereus had a higher mechanical stability than P. fluorescens biofilms. The overall results demonstrate that P. fluorescens and B. cereus formed physiologically distinct biofilms, B. cereus biofilms mainly being constituted by cells and P. fluorescens biofilms largely constituted by extracellular proteins and polysaccharides. B. cereus biofilms had a substantially higher mechanical stability than P. fluorescens biofilms.     

Two-dimensional cellular automaton model for mixed-culture biofilm.

Structural and microbial heterogeneity occurs in almost any type of biofilm system. General approaches for the design of biofilm systems consider biofilms as homogeneous and of constant thickness. In order to improve the design of biofilms systems, models need to incorporate structural heterogeneity and the effect of inert microbial mass. We have improved a 2D biofilm model based on cellular automata (CA) and used it to simulate multidimensional biofilms with active and inert biomass including a self-organizing development. Results indicate that the presence of inert biomass within biofilm structures does not change considerably the substrate flux into the biofilm because the active biomass is located at the surface of the biofilm. Long-term simulations revealed that although the biofilm system is highly heterogeneous and the microstructure is continuously changing, the biofilm reaches a dynamic steady-state with prediction of biofilm thickness and substrate flux stabilizing on a delimited range.     

Biofilm formation by a biotechnologically important tropical marine yeast isolate, Yarrowia lipolytica NCIM 3589

Biofilm formation by Yarrowia lipolytica, a biotechnologically important fungus in microtitre plates, on glass slide surfaces and in flow cell was investigated. In microtitre plates, there was a short lag phase of adhesion followed by a period of rapid biofilm growth. The fungus formed extensive biofilms on glass slides, whereas in flow-cells a multicellular, three-dimensional microcolony structure was observed. The isolate formed biofilms in seawater and in fresh water media at neutral pH when grown in microtitre plates. The carbon sources differentially affected formation of biofilms in microtitre plates. Lactic acid, erythritol, glycerol, glucose and edible oils supported the formation of biofilms, while alkanes resulted in sub-optimal biofilm development. A variation in the morphology of the fungus was observed with different carbon sources. The results point to the possible existence of highly structured biofilms in varied ecological niches from where the yeast is isolated.     

Characterizing temporal development of biofilm porosity using artificial neural networks

We used artificial neural networks (ANN) to compute parameters characterising biofilm structure from biofilm images and to interpolate a limited number of experimental data characterising the effects of nutrient concentration and flow velocity on the areal porosity of biofilms. ANN were trained using a set of experimental data characterising structural parameters of biofilms of Pseudomonas aeruginosa (ATCC #700829), Pseudomonas fluorescens (ATCC #700830) and Klebsiella pneumoniae (ATCC #700831) for various flow velocities and glucose concentrations. We used 80% of the data to train ANN and 10% of the data to validate the results, which is routinely carried out as a countermeasure against overtraining. Trained ANN were used to interpolate into the data set and evaluate the missing 10% of the data. To compare ANN accuracy in evaluating the missing data with the accuracies achieved using other interpolation algorithms, we used spline, cubic, linear and nearest-neighbour interpolation algorithms to evaluate the missing data. ANN estimates were consistently closer to the experimental data than the estimates made using the other methods.     

Use of a dynamic gassing-out method for activity and oxygen diffusion coefficient estimation in biofilms

p-toluenesulphonic acid degradation by Comamonas testosteroni T-2 in multi-species biofilms was studied in a fixed bed biofilm reactor. The polypropylene static mixer elements (Sulzer Chemtech Ltd., Switzerland) were used as a support matrix for biofilm formation. Biofilm respiration was estimated using the dynamic gassing-out oxygen uptake method. A strong relation between oxygen uptake and reactor degradation efficiency was observed, because p-toluenesulphonate degradation is a strictly aerobic process. This technique also allowed us to estimate the thickness of the active layer in the studied system. The mean active thickness was in order of 200 μm, which is close to maximum oxygen penetration depth in biofilms. A transient mathematical model was established to evaluate oxygen diffusitivity in non-steady-state biofilms. Based on the DO concentration profiles, the oxygen diffusion coefficient and the maximum respiration activity were calculated. The oxygen diffusion coefficient obtained (2 10^-10-1.2 10^-9 m² s^-1) is in good agreement with published values. The DO diffusion coefficient varied with biofilm development. This may be, most likely, due to the biofilm density changes during the experiments. The knowledge of diffusivity changes in biofilms is particularly important for removal capacity estimation and appropriate reactor design.     

Microsensor measurement of oxygen concentration in biofilms: from one dimension to three dimensions.

In this study, we measured oxygen concentration in biofilms in one dimension in field conditions and in three dimensions in laboratory conditions by using a robust oxygen microsensor in combination with an automation and data acquisition system. The biofilms were on the discs of rotating biological contactors treating domestic wastewater. The results of this study provide experimental evidence on oxygen distribution in wastewater biofilms and on biofilm structure. (1) The three dimensional measurements of oxygen concentration in biofilms revealed "pockets" of oxygen in deep sections of biofilms. In these isolated "pockets," located 600-760 microm from the biofilm surface, dissolved oxygen concentration was as high as 1 mg/L. This depth of oxygen diffusion is deeper than what was determined based on one dimensional measurements. (2) The heterogeneity of oxygen distribution was related to the surface structure of biofilms. The structure of the biofilm surface affected the diffusion boundary layer over the surface and, in turn, the oxygen diffusion and distribution inside biofilms. (3) Oxygen concentration in biofilms changed generally from a high degree of heterogeneity near the biofilm surface to a low degree of heterogeneity in deep sections of biofilms, indicating a cell-clusters-like structure near the surface and a more compact base layer close to the substratum.     

The fate of legionellae within distribution pipe biofilms: measurement of their persistence, inactivation and detachment.

Distribution pipe biofilms present a currently unquantified public health risk to consumers receiving water for domestic potable and non-potable use. The aim of this study was to quantify the numbers of legionellae, used here as model bacterial pathogens, that may accumulate, persist within and detach from distribution pipe biofilms. L. pneumophila recovered by standard culture from an 8 week-old biofilm formed within a novel pilot-scale water distribution system represented 1% of those present in the adjacent bulk water. A combined chlorine concentration exceeding 0.2 mg x L(-1) eliminated culturable sessile legionellae altogether, though the reduction in FISH-positive cells represented just 75+/-25% of the original amount, compared to a 5-log reduction in culturable cells during the same period. Where there was < 0.1 mg x L(-1) combined chlorine, an exponential decay/loss of sessile L. pneumophila was observed (k = 0.37 - 0.41) over the course of a 38-day experimental period. The inoculation of the system with 1 microm fluorescent microspheres and legionellae demonstrated that removal of the latter was dominated by chemical disinfection, with erosion and biological grazing playing lesser roles. Under turbulent (Re approximately 5000) conditions, larger clusters of biofilm become detached from substrata, with more than 90% of sessile legionellae mobilised into the bulk water phase. Interaction with both biofilms and a thermophilic Acanthamoeba isolate reduced the susceptibility of legionellae to thermal inactivation by between one and two orders of magnitude, though it increased their sensitivity to chemical (free and combined chlorine) disinfection.     

Introduction to the IWA task group on biofilm modeling.

An International Water Association (IWA) Task Group on Biofilm Modeling was created with the purpose of comparatively evaluating different biofilm modeling approaches. The task group developed three benchmark problems for this comparison, and used a diversity of modeling techniques that included analytical, pseudo-analytical, and numerical solutions to the biofilm problems. Models in one, two, and three dimensional domains were also compared. The first benchmark problem (BM1) described a monospecies biofilm growing in a completely mixed reactor environment and had the purpose of comparing the ability of the models to predict substrate fluxes and concentrations for a biofilm system of fixed total biomass and fixed biomass density. The second problem (BM2) represented a situation in which substrate mass transport by convection was influenced by the hydrodynamic conditions of the liquid in contact with the biofilm. The third problem (BM3) was designed to compare the ability of the models to simulate multispecies and multisubstrate biofilms. These three benchmark problems allowed identification of the specific advantages and disadvantages of each modeling approach. A detailed presentation of the comparative analyses for each problem is provided elsewhere in these proceedings.     

Reproducibility of biofilm processes and the meaning of steady state in biofilm reactors.

The need for reproducing biofilm processes is undisputable - the quality of biofilm research depends on this reproducibility. However, as many biofilm researchers know, long-term biofilm processes are notoriously difficult to reproduce. To avoid problems related to biofilm reproducibility two strategies are used: (1) to study very young biofilms that have accumulated for a few hours to a few days only, and (2) to run biofilm experiments only once. The first approach trades reproducibility for relevance because natural biofilms are usually older, often much older than a few days. This approach can be applied to answer questions relevant to initial events of biofilm formation but not questions relevant to long-term biofilm accumulation. The second approach conceals the problem of biofilm reproducibility. To assure reproducibility of biofilm processes, we methodically followed a procedure for growing biofilms in terms of microbial makeup, media composition, temperature, surface preparation, etc. Despite all this effort the reproducibility of our results for long term growth is unimpressive. Consequently, the question had to be asked: Are biofilm processes reproducible? The experiments described in this paper address this question. Biofilms grown in two identical and identically operated biofilm reactors had comparable structure only until the first sloughing event. After that, biofilms had different patterns of accumulation.     

Behaviour of biofilm systems under varying hydrodynamic conditions.

Heterotrophic biofilms were cultivated in long-term experiments in biofilm tube reactors. During the biofilm cultivation the substrate loading of glucose was kept constant while the hydrodynamic conditions were changed stepwise. To describe the behaviour of the biofilm structure under these varying flow conditions the mass transfer and transport at the bulk/biofilm interface and inside the biofilm was investigated with oxygen microelectrodes. Furthermore, the biofilm density was used to describe the biofilm compactness before and after the change of the hydrodynamic condition. The obtained results show that the biofilm density and also the substrate flux decreased with decreasing flow velocity in the bulk phase. Additionally the slope of the oxygen concentration profiles decreased and the thickness of the concentration boundary layer increased. On the other hand, increasing the flow velocity in the bulk phase led both to a higher biofilm density and a higher maximum substrate flux. The biofilm surface became more homogenous and the thickness of the concentration boundary layer decreased. The time for adaptation of the biofilm structure after changing the hydrodynamic conditions ranged between 1 and 3 weeks.     

The rate of iron corrosion for different organic carbon sources during biofilm formation.

The effects of total organic carbon and biofilm on microbial corrosion were quantified using serum bottles in a 2 x 2 factorial design. Both organic carbon and biofilm bacteria had a significant effect on the iron corrosion rate, irrespective of the levels of the other variable (p = 0.05). There was no evidence of interaction between organic carbon and biofilm bacteria. Within the tested levels, the addition of exogenous organic carbon increased the corrosion rate by an average of 3.838 mg dm(-2) day(-1) (mdd), but the presence of biofilm bacteria decreased the rate by an average of 2.305 mdd. More iron was released from the coupon in response to organic carbon. Powder x-ray diffractometry indicated that the scales deposited on the corroded iron surface consisted primarily of lepidocrocite (gamma-FeOOH), magnetite (Fe3O4) and hematite (alpha-Fe203). Corrosion rates by different organic carbon sources, i.e. acetate, glucose and humic substances, were compared using an annular biofilm reactor. One-way ANOVA suggested that the effect of each carbon source on corrosion was not the same, with the iron corrosion rate highest for glucose, followed by acetate, humic substances and the control. Magnetite was a major constituent of the corrosion products scraped from iron slides. Examination of community-level physiological profile patterns on the biofilms indicated that acetate was a carbon source that could promote the metabolic and functional potentials of biofilm communities.     

A model of two-dimensional biofilm morphology

A mathematical model of biofilm development was proposed. The model, which can be classified as a cellular automaton, is based on simple local rules for growth and detachment of individual cells. Results of several numerical simulations suggest that the thickness of concentration and hydrodynamic boundary layers may have an important effect on the developing biofilm structure. When external mass transfer limitations are significant, model biofilms develop an open structure. When the concentration boundary layer is reduced and external mass transfer is enhanced, a dense layer of a biofilm develops. Biofilm strength to withstand erosion has a smaller but also significant effect on its structure.     

Chemical stress induced by copper: examination of a biofilm system.

Biofilm systems have been widely used in wastewater treatment plants. However, little information is available on the impact of toxic chemicals on the performance of fixed film systems. This study was aimed at evaluating the impact of copper on a biofilm system by examining a variety of parameters, including reactor pH, DO, substrate concentrations, secretion of extracellular polymeric substances (EPS), and copper removal and accumulation. The microbial communities in the biofilms were also examined using automated ribosomal intergenic spacer analysis (ARISA). Four rotating drum biofilm reactors were used to produce biofilms. One reactor was used to produce biofilms under copper free conditions; while the others were used to produce biofilms grown under three different copper contamination levels, namely 100 ppb, 200 ppb, and 500 ppb, for a prolonged period. The following results were obtained: (1) biofilm reactor performance was not significantly impacted as demonstrated by the pH, DO, substrate removal, and total solids in the effluent; (2) however, copper contamination inhibited EPS production in the biofilms; (3) copper removal efficiencies of 25-31% were obtained for the three copper contamination levels studied; (4) fixed films functionalized as a reservoir to accumulate more copper over time; and (5) copper contamination selected for specific species that were able to tolerate this stress and that may contribute to its remediation.     

Contamination potential of drinking water distribution network biofilms.

Drinking water distribution system biofilms were investigated for the presence of hygienically relevant microorganisms. Early biofilm formation was evaluated in biofilm reactors on stainless steel, copper, polyvinyl chloride (PVC) and polyethylene coupons exposed to unchlorinated drinking water. After 12 to 18 months, a plateau phase of biofilm development was reached. Surface colonization on the materials ranged between 4 x 10(6) and 3 x 10(7) cells/cm2, with heterotrophic plate count (HPC) bacteria between 9 x 10(3) and 7 x 10(5) colony-forming units (cfu)/cm2. Established biofilms were investigated in 18 pipe sections (2 to 99 years old) cut out from distribution pipelines. Materials included cast iron, galvanized steel, cement and PVC. Colonization ranged from 4 x 10(5) to 2 x 10(8) cells/cm2, HPC levels varied between 1 and 2 x 10(5) cfu/cm2. No correlation was found between extent of colonization and age of the pipes. Using cultural detection methods, coliform bacteria were rarely found, while Escherichia coli, Pseudomonas aeruginosa and Legionella spp. were not detected in the biofilms. In regular operation, distribution system biofilms do not seem to be common habitats for pathogens. However, nutrient-leaching materials like rubber-coated valves were observed with massive biofilms which harboured coliform bacteria contaminating drinking water.     

Influence of biofilm on removal of surrogate faecal microbes in a constructed wetland and maturation pond.

The effect of biofilm on the attenuation of pathogen-sized particles from wastewater was compared for biofilms cultivated in a surface flow constructed wetland (SFW) and maturation pond (MP) The fate of fluorescently labelled microspheres (FLM) as surrogates for viruses (0.1 microm), bacteria (1 microm) and parasitic protozoa (4.5 microm dia) was investigated in microcosms in the presence or absence of biofilms. Rates of FLM removal from suspension were higher in the presence of biofilms for all particle sizes (kd 0.02-0.11 h(-1)) in MP and SFW microcosms with removal efficiency related to particle size and biofilm thickness and structure. Greater removal of 0.1 microm (79-81%), 1 microm FLM (92-96%) and 4.5 microm FLM (up to 98%) from suspension were found for microcosms containing thicker (autotrophic) biofilms grown in the MP or open water zone of the SFW. Lower removal of 43% (0.1 microm), 59% (1 microm) and 84% (4.5 microm) occurred in microcosms containing thinner heterotrophic biofilms from SFW vegetated zones. Providing surfaces for attachment of photosynthetic biofilms offers potential to enhance pathogen removal in open water systems. In vegetated systems, linkage to more oxic openwater zones may allow thicker and 'stickier' epiphytic biofilms to develop, improving pathogen interception and removal.     

Characterization of natural biofilms in temperate inland waters

Community dynamics of microalgae in natural biofilms grown on 10 × 3 cm glass slides were studied in three inland water systems in Central Ontario, Canada. The periphyton communities were analyzed for species composition, diversity, density and biofilm thickness. The usefulness of periphyton community dynamics and species diversity in water quality monitoring was tested. The density of microalgae varied from 2.4 × 10⁷/cm² (Lake Couchiching) to 18 × 10⁷/cm² (Lake Simcoe) with highest species diversity at Lake Couchiching. Lake Simcoe with its moderately high phosphorus and low organic carbon showed the highest density of microalgae while Lake Couchiching with lowest total phosphorus and highest organic carbon showed the lowest density of microalgae in biofilms. The results of analysis of variance showed significant variation in the number of genera, density, biofilm thickness and diversity of microalgae in the three sampling locations. The Mill Creek site with minimum anthropogenic disturbance, minimum light availability, lower water temperature and slow but steady flowing conditions recorded the lowest species diversity and number of genera. The dominant genera of diatoms were significantly different in the three sampling locations. This study thus showed the usefulness of periphyton community dynamics in the assessment of water quality in the inland water systems.     

Biofilms in an urban water distribution system: measurement of biofilm biomass, pathogens and pathogen persistence within the Greater Stockholm Area, Sweden.

Distribution pipe biofilms can provide sites for the concentration of a wide range of microbial pathogens, thereby acting as a potential source of continual microbial exposure and furthermore can affect the aesthetic quality of water. In a joint project between Stockholm Water, the MISTRA "Sustainable Urban Water" program, the Swedish Institute for Infectious Disease Control and the Royal Technical University, Stockholm, the aim of the current study was to investigate biofilms formed in an urban water distribution system, and quantify the impact of such biofilms on potential pathogen accumulation and persistence within the Greater Stockholm Area, Sweden. When used for primary disinfection, ultra-violet (UV) treatment had no measurable influence on biofilm formation within the distribution system when compared to conventional chlorination. Biofilms produced within a model pilot-plant were found to be representative to those that had formed within the larger municipal water distribution system, demonstrating the applicability of the novel pilot-plant for future studies. Polystyrene microspheres (1.0 microm) and Salmonella bacteriophages demonstrated their ability to accumulate and persist within the model pilot-plant system, where the means of primary disinfection (UV-treatment, chlorination) had no influence on such phenomena. With the exception of aeromonads, potential pathogens and faecal indicators could not be detected within biofilms from the Stockholm water distribution system. Results from this investigation may provide information for water treatment and distribution management strategies, and fill key data gaps that presently hinder the refinement of microbial risk models.     

Growth, structure and oxygen penetration in particle supported autotrophic biofilms.

Particle supported autotrophic biofilms were cultivated in external-loop airlift reactors at two different pumice concentrations. Oxygen microelectrodes were used to investigate substrate transport and conversion. A special flow cell was designed for the measurement of oxygen concentration profiles in the particle supported biofilms under defined hydrodynamic conditions. The oxygen concentration profiles inside the biofilms were found to be steeper at high flow velocities in the bulk phase of the flow cell compared to those at low flow velocities. Furthermore, the oxygen flux increased and the thickness of the concentration boundary layer decreased with increasing flow velocity. This dependence was found to be more pronounced in less dense biofilms out of airlift reactors with lower pumice concentrations. In addition confocal laser scanning microscopy (CLSM) was used to visualize the biofilm structure. The volume fractions of bacteria and extracellular polymeric substances (lectin-specific EPS-glycoconjugates) were measured in living fully hydrated biofilms. Both the microelectrode and CLSM measurement showed the influence of shear stress on particle supported biofilms. A higher particle concentration led to dense biofilms with a homogeneous surface, lower thickness of the concentration boundary layer and steeper oxygen concentration profiles. The combination of both techniques allows a detailed and quantitative characterisation of particle associated biofilm structure and function.     

The effect of detachment on biofilm structure and activity: the oscillating pattern of biofilm accumulation.

In our previous papers we have demonstrated that biofilm structure never reaches a steady state in biofilm reactors; in this paper we link this fact to biofilm detachment and to the oscillating pattern of biofilm accumulation. In one respect reactors supporting suspended microbial growth and reactors supporting attached microbial growth (biofilms) are similar: in both the biomass accumulates in the reactor and is disposed of with the effluent. However, while in reactors with suspended microbial growth biomass accumulation and disposal occur simultaneously, in biofilm reactors these two processes are separated in time. Biomass accumulation in biofilm reactors shows a distinct pattern composed of three phases: (1) growth, (2) detachment, (3) regrowth. Despite this distinct pattern of biofilm accumulation observed at the microscale, biofilm reactors do reach a steady state of substrate removal.     

A comparison of biofilms from macrophytes and rocks for taste and odour producers in the St. Lawrence river.

Given their widespread and prolific annual development in the St. Lawrence River (SLR), macrophytes (i.e. submerged aquatic plants) represent large surface areas for biofilm growth and potentially important sites for associated production of taste and odour (T&O) compounds. We therefore evaluated the importance of submerged macrophytes and their associated biofilms for production of T&O compounds, 2-methylisoborneol (MIB) and geosmin (GM), compared with biofilms from adjacent rocks. We also tested the hypothesis that production of these compounds would differ between macrophyte species, based on the premise that they are not inert substrates but directly influence the communities that colonise their surfaces. Samples collected from transects across the SLR between Kingston and Cornwall, ON were dominated by the flat-bladed Vallisneria spp., and the leafed Myriophyllum spicatum, Elodea canadensis, Chara spp., Potamgeton spp., and Ceratophyllum spp. Overall, MIB and GM levels in biofilms ranged widely between samples. Expressed per g dry weight of biofilm, median levels from macrophyte were 50 (range 1-5000) ng MIB g(-1) and 10 (<1 to 580) ng GM g(-1) compared with 50 (range 5-970) ng MIB g(-1) and 160 (1-1600) ng GM g(-1) from rocks. Based on non-parametric statistical analysis, levels of GM were higher on a g dry weight basis in biofilms from rocks than macrophytes (P = 0.02), but MIB levels were similar (P = 0.94). However, when normalised for differences in substrate surface area (i.e. ng cm(-2)), levels of both MIB and GM were higher in biofilms from rocks than from macrophytes (P < 0.01). There were no discernable differences in MIB and GM concentrations from biofilms of different macrophytes based on either g dry weight sample or surface area (P > 0.05). Overlying water (OLW) concentrations ranged between 2-45 ng L(-1) for MIB and 5-30 ng L(-1) for GM and were not correlated with levels in adjacent biofilms. However, OLW concentrations peaked in shallow, low energy embayments consistent with enhanced production and release of MIB and GM in nearshore areas. The results support our previous work showing the importance of biofilms on various surfaces (rocks, macrophytes and zebra mussels) for MIB and GM production in the SLR, but suggest that inert surfaces like rocks are more productive sites per unit surface area than macrophytes.