Environ. Sci. Technol. 2008, 42, 1485–1491
Experimental Studies of the Influence of Grain Size, Oxygen Availability and Organic Carbon Availability on Bioclogging in Porous Media VICTORIA L. HAND, JONATHAN R. LLOYD, DAVID J. VAUGHAN, MICHAEL J. WILKINS, AND STEPHEN BOULT* School of Earth, Atmospheric and Environmental Sciences and Williamson Research Centre for Molecular Environmental Science, The University of Manchester, Manchester M13 9PL, U.K.
Received August 14, 2007. Revised manuscript received November 9, 2007. Accepted November 29, 2007.
Changes in the hydraulic properties of porous material due to bioclogging have been observed in many laboratory simulations and field studies. Because such changes in hydraulic properties influence the movement of fluids and contaminants, microbial ecology data are required for improved transport modeling. Here we investigate the effects of environmental variables previously shown to influence bioclogging, specifically oxygen availability, sediment grain size, and organic carbon (nutrient) concentration on the hydraulic properties of simulated subsurface environments. Our study provides evidence of a different clogging mechanism for aerobic and anaerobic microbial communities under high organic carbon concentrations (400 mg L-1). This work also suggests that the clogging mechanism operating in anaerobic microbial communities is more sensitive to carbon availability than that in the aerobic microbial communities. We found that grain size does have an effect on clogging, but it appears that there is a threshold carbon concentration, and therefore biomass, below which these effects are insignificant. Differences between the microbial communities that developed under different oxygenation conditions were detected using 16s rRNA analysis.
Introduction It is widely recognized that microbial colonies are surrounded by extensive amounts of extracellular polymeric substances (EPS), and many previous studies have documented the importance of the EPS in microcolony formation, cohesion, and adhesion. Reductions in permeability and, hence, hydraulic conductivity (Ks) in porous media due to “bioclogging” caused by such attached microorganisms and their associated EPS have been observed in many laboratory studies (1–5). As the rate of fluid flow through porous media is directly proportional to the Ks of the media, such effects will directly influence mass-transfer rates of natural fluids and contaminant plumes in the subsurface (3, 6, 7). More recent studies have looked in further detail at the polysaccharide, protein, and other constituents that make up * Corresponding author phone: (+44) 161-275-3867; fax (+44) 161275-3947; e-mail:
[email protected]. 10.1021/es072022s CCC: $40.75
Published on Web 01/23/2008
2008 American Chemical Society
microbial EPS using specific probes, such as lectins (8–10). However, the specific relationships between biofilm cohesiveness and properties such as EPS composition and distribution, although important for understanding, predicting, and controlling biofilm adhesion and sloughing, remain unknown (11). Variables such as oxygen (12) and nitrogen levels (13), temperature (14), pH (15, 16), and nutrient availability (17) have been shown to affect the content and production of EPS. Variations in the relative amounts of microbes and microbial extracellular products will have a variable influence on hydraulic properties. The production of EPS from both aerobic and anaerobic bacteria has been shown to be the dominant mechanism associated with clogging in columns of sintered glass beads (18, 19). Research has shown that oxygen saturation in a system influences the degree and the mechanism of bioclogging, and numerous experimental observations have shown large decreases in Ks concomitant with the establishment of an anaerobic environment (4). Reductions of Ks in soils and aquifers due to the activities of bacteria have often been suggested to be a principally anaerobic process. However, in one study, oxygen-saturated and oxygen-free nutrient solutions were both passed through nonsterile sandstone cores, and a higher rate of clogging was observed in the resulting aerobic rather than the anaerobic systems (20). Aerobic bacteria caused clogging near the inlet surface, and anaerobic bacteria caused clogging throughout the core. Much of the reduction of Ks was observed in the inlet area in the aerobic soil columns, which suggests that the correlation between bioclogging and anaerobic conditions may not always be due to a causal relationship. Other investigations (4, 7) have found that, under growth conditions limited by the availability of oxygen, decreases in Ks do not appear to be caused by EPS; instead, the decreases were seen to be due to the presence of large aggregates of cells forming local plugs within the pores. In all cases, these authors saw large decreases in Ks without any evidence for the formation of continuous biofilms. Contrary to the established model of biofilm formation, they found that coverage of the solid surfaces by the bacterial cells is sparse and heterogeneous. One study observed that continuous biofilms could be sustained with high nutrient loadings but that patchy growth is more likely at lower loadings (21). This researcher also noted that the distinction between continuous and patchy biofilms is not crucial for modeling substrate removal but may be important for modeling clogging (21). In many previous studies, the experimental nutrient availability was designed to ensure microbial growth, whereas under the nutrient-limited conditions of a more typical subsurface environment, microbial communities are significantly different in terms of the quantity and structure of the biomass (22). It is possible that many previous simulations of bioclogging are not representative of widespread processes in the geosphere (1, 2, 23), and field observations of bioclogging have been confined to observations around wells where nutrient loadings have been relatively high because of high flow rates or nutrient additions. Consequently, there is a need to quantify the influence of microbial growth and their associated EPS under low nutrient conditions on the hydraulic properties of porous media such as unconsolidated sediment, or porous rocks such as many types of sandstone. Such studies also need to involve manipulation of the environmental variables previously shown to influence bioclogging, particularly oxygenation conditions and grain size (24, 25). Scanning electron microscope (SEM) imaging has shown that the mechanism by which clogging occurs, VOL. 42, NO. 5, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1485
TABLE 1. Anion, Cation, and Organic Acid Concentrations in the Simulated Groundwater Made Using Analytical Grade Chemicals (BDH) and 15 MΩ DIWa anion
concn, mol L-1
cation
concn, mol L-1
organic acids
concn, mol L-1
CO3 NO3 SO4 Cl
4.555 × 10-3 7.480 × 10-5 3.262 × 10-4 1.340 × 10-3
Na K Si Ca Mg
3.576 × 10-3 8.802 × 10-5 1.608 × 10-4 1.670 × 10-3 7.942 × 10-4
acetic propionic butyric isobutyric valeric isovaleric hexanoic octanoic
1.283 × 10-4 1.280 × 10-5 4.545 × 10-6 2.273 × 10-6 6.863 × 10-7 3.431 × 10-6 6.034 × 10-7 1.389 × 10-6
a Organic acid concentrations in the organic carbon-enriched experiments were 10 times greater than the concentrations shown here for the low organic carbon experiments.
FIGURE 1. Pressure changes in columns containing different grain size fractions of quartz and inoculated with a natural microbial consortium, and through which organic carbon enriched synthetic groundwater was passed for 40 days. The experiment under anaerobic conditions using the 500–701 µm fraction is shown by the dashed gray line. The other results are all for experiments under aerobic conditions (black line, 250–355 µm; gray line, 355–500 µm; dashed black line, 500–710 µm grain size fractions, respectively). and its impact on the hydraulic properties of the porous media, are also a function of grain size. Variations in the type of microbial growth and its location at the sub-millimeter scale can result in changes in porosity, permeability, dispersivity, or all of these variables. Changes in the geometries of the pore space can lead to pore-throat clogging and hence to reductions in Ks (2, 26) but may also result in streamlining of flow, reduction in dispersivity, and an increase in Ks. In the present work, microbially induced changes in the hydraulic properties of porous media of different grain sizes were studied under different nutrient and oxygenation conditions. Experimental systems were first examined to show the presence of microbes and microbial activity, and then the bulk hydraulic properties were compared between systems. Our experimental design allowed the following hypotheses concerning the influence of oxygen concentration and grain size on bioclogging to be tested: Does bioclogging occur under environmentally realistic low nutrient loading conditions? Do the factors that are known to affect bioclogging under high nutrient loadings continue to be important at low loadings? Is the mechanism by which bioclogging occurs the same under high and low nutrient conditions? 1486
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 5, 2008
Experimental Section Preparation of Experimental Columns. Quartz sand, to which microbial adhesion is relatively limited (27, 28), was chosen as the porous medium because of its inert nature. Three distinct size fractions (255–355, 355–500, and 500–710 µm) were sieved from a natural pure quartz sand (supplied by David Ball Group, Sandbach, U.K.) using standard test sieves, for investigation of grain size effects. All three fractions were used in the aerobic organic carbon-enriched experiments and the largest and smallest in the low organic carbon experiments. In both cases, under anaerobic conditions only the largest size fraction was used. All column experiments were run in triplicate unless otherwise stated. The sand fractions used in both sets of experiments were cleaned by repeated sonication and rinsing with 15 MΩ deionized water (DIW) until the supernatant ran clear and were then dried and packed into acetal columns of 20 mm diameter and 25 mm length, which were continuously tapped during filling. All columns were kept in darkness at 10 °C. Prior to microbial inoculation, all components of the system were sterilized by autoclaving at 121 °C for 20 min. Pressure transducers and some heat-sensitive tubing were sterilized by soaking overnight in Virkon and rinsing in sterile DIW.
FIGURE 2. ESEM images of quartz grains with microbial materials extracted from columns (500–700 µm) inoculated with a natural microbial consortium, and through which organic carbon enriched synthetic groundwater was passed for 40 days under aerobic (left-hand images) and anaerobic (right-hand images) conditions. Note the different scale bars. Under aerobic conditions there is an abundance of EPS stretching in a sheetlike fashion between grains. Individual microbes are visible embedded within these structures. Under anaerobic conditions, this material is visible as much thinner and weblike, filling space between grains on a smaller scale than under aerobic conditions, which is more consistent with the microbial growth seen previously under low nutrient conditions (17). Preparation of Simulated Groundwater. A simulated groundwater, representative of that from a much-studied site in Cumbria, Northwest of England, was prepared using Analar grade reagents (BDH, U.K.) (see Table 1). The simulated groundwater used in the organic carbon-enriched experiments contained organic acid concentrations ten times greater than those used in the low organic carbon experiments. All simulated groundwater was adjusted to pH 7.3 by dropwise addition of 1 M HCl and sterilized by autoclaving at 121 °C for 20 min. An inoculum containing a natural microbial consortium from a soil sample taken from the same site mentioned above was prepared by adding 20 mL of a soil infusion to 5 L of the organic carbon-enriched simulated groundwater. This inoculum was allowed to incubate for 10 days at 25 °C before being circulated through the sand columns for 48 h at 0.1 mL min-1. After this inoculation, the simulated groundwater was pumped at a constant flow rate of 0.014 mL min-1 through all columns. Monitoring of Experimental Columns. Pressure, pH, and dissolved oxygen concentrations were continuously monitored at each column inflow using pressure transducers (only available for two of the three replicates in the aerobic low carbon experiments) and chemical sensors (Intelisys Ltd., U.K.). In the low organic carbon experiment, the influent and effluent nitrate, nitrite, and sulfate concentrations were monitored every seven days over a 35 day period by highPerformance liquid chromatography (HPLC) using a Dionex 4000I isocratic ion chromatography system with a Dionex ASII-HC column. The measured concentrations of NO3- and NO2- are presented here as cumulative nitrogen storage. A minimum of one sample per week was collected during the course of the column experiments.
Environmental Scanning Electron Microscopy. Samples of porous media from the top (outlet) and bottom (inlet) of the column were taken after the columns were dismantled at the end of the experiments. This was done with great care in order not to disturb the biofilm structure within the grains. The specimens were placed directly onto an aluminum stub and viewed in an environmental scanning electron microscope (ESEM). This instrument was a Philips XL30 ESEM-FG operated between 10 and 20 kV. Microbial Community Analysis. The bacterial communities that were established in the microbially coated porous media were further investigated by extraction of DNA and comparison of 16s rRNA gene sequences. This was undertaken to assess the nature of the microbial community that had developed and the relationships between bacterial community structure and the physical and chemical changes within the columns. Experimental details of the DNA extraction and sequencing, clone library construction, RFLP grouping, and phylogenetic analysis can be found in the Supporting Information.
Results and Discussion Organic Carbon-Enriched Experiments. Pressure increases occurred in all columns over the experimental period (see Figure 1), and by the end of the experiments there was little difference in pressure between the columns. However, the rate at which the pressure increased was dependent on grain size and oxygen availability. The rate of pressure increase was slowest in the anaerobic column and within the aerobic columns increased from the smallest to the largest grain size. Given that pressure increase is indicative of bioclogging, there appears to have been initial microbial activity to a relatively VOL. 42, NO. 5, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1487
FIGURE 3. Cumulative storage of nitrogen, indicative of microbial activity, in inoculated quartz sand columns during 40 days of passing synthetic groundwater containing low organic carbon through them. Storage was calculated as the difference between influent nitrate and nitrite and effluent nitrate and nitrite flux. Total nitrogen storage was greatest under anaerobic conditions (C). There is little difference in nitrogen storage between different grain sizes (250–355 µm A; 500–710 µm B) under aerobic conditions. Between actual sample data points, values were estimated by linear interpolation. Error bars represent 95% confidence limits and are shown only for N storage for clarity.
TABLE 2. Bacteria Identified in the >500 < 710 µm Anaerobic and the >500 < 710 µm Aerobic Columns Differentiated by RFLP Analysis and Supposed Identification Assigned on the Basis of the Closest Matched 16s rRNA Gene Sequence in the NCBI Database
anaerobic, >500 < 710 µm
aerobic, >500 < 710 µm
RFLP pattern
highest Blastn matcha
no. identical/total (% identity)b
occurrence of clones analyzed (%)
1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
Pseudomonas gessardii Pseudomonas gessardii Sphingobium yanoikuyae Caulobacter crescentus Delftia acidovorans Comamonas acidovorans Brevundimonas vesicularis Methylobacterium extorquens Pseudomonas fulgida Sphingomonas paucimobilis Caulobacter crescentus Ralstonia sp. Flavobacterium heparinum Caulobacter crescentus Pseudomonas mephitica Pseudomonas trivializ Caulobacter intermedius Ralstonia basilensis
511/512 (99%) 514/517 (99%) 442/449 (98%) 377/393 (95%) 513/519 (98%) 425/506 (83%) 456/461 (98%) 459/463 (99%) 515/520 (99%) 384/394 (97%) 453/463 (97%) 500/501 (99%) 507/517 (98%) 421/463 (90%) 500/516 (96%) 368/384 (95%) 445/448 (99%) 501/503 (99%)
71.4 2.4 4.8 4.8 4.8 2.4 4.8 2.4 2.4 51.5 12.1 9.09 6.06 6.06 6.06 3.03 3.03 3.03
a Refers to the sequence in the Blastn database that possessed the highest percent similarity. × (number of identical bases)/(total number of positions compared).
steady state, which was subject to further smaller oscillations perhaps as a result of periodic sloughing. The rate at which this equilibrium becomes established is affected by grain 1488
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 5, 2008
b
Percent similarity ) 100
size and oxygen availability, but, at least on the time scale of this experiment, the magnitude of the pressure increase is not.
FIGURE 4. Pressure changes in the columns inoculated with a natural microbial consortium and through which synthetic groundwater containing low organic carbon was passed for 40 days. Pressure increases were much greater under aerobic conditions (grey region bounds the range of readings from four aerobic columns of two grain sizes of 250–355 and 500–710 µm) than under anaerobic conditions (black region bounds the range of readings from three columns of grain size of 500–710 µm). The data gap between days 13 and 18 was due to an equipment malfunction. ESEM observation of the porous material at the end of the experiment shows evidence of abundant microbial activity under both aerobic and anaerobic conditions. Much more EPS are visible in the aerobic columns in comparison to the anaerobic columns (see Figure 2). The EPS in the aerobic columns could be seen spanning gaps of 100 µm or more between grains and completely filling pore spaces of 50 µm or more. However, the morphology of the EPS in the anaerobic columns is quite different; there are less EPS than in the anaerobic columns and it appears to be more “stringy” than the thin sheets seen in the aerobic columns. Evidence of more abundant EPS together with a corresponding initial pressure increase under aerobic, and not anaerobic conditions, indicate that greater reductions of Ks occur under aerobic conditions due to microbial EPS. However, once the microbial communities are established, these differences in the amount and morphology of EPS do not have any impact on hydraulic conductivity. It is important to bear in mind the possibility of artifacts due to dehydration of the sample in the ESEM. Low Organic Carbon Experiments. The consumption of available electron acceptors such as nitrate (Figure 3) and sulfate demonstrates microbial activity in all columns. There was an active nitrate-reducing population in all of the columns. The nitrogen imbalance between removal of nitrate and the release of nitrite is shown in Figure 3, and, although gaseous nitrogen flux was not monitored, this does give a measure of nitrogen storage. In the aerobic columns, the cumulative storage of nitrogen, which provides evidence of microbial activity, was similar in both the small grain size columns and the large grain size columns. This suggests that grain size does not influence the size of the microbial population under these conditions. The cumulative storage of nitrogen from nitrate was much greater under anaerobic than under aerobic conditions. The nitrite production in the anaerobic columns was sufficiently high as to suggest a functional difference between the aerobic and anaerobic microbial communities, which was to be expected and subsequently confirmed by the microbial community analysis. The ARISA banding patterns produced using the DNA extracted from an anaerobic column and an aerobic column for the low organic carbon experiments indicate that there were at least nine dominant bacterial species in each sample among the clones investigated. Nine clones were sequenced from each of the aerobic and anaerobic columns (Table 2). The sequence data produced were analyzed phylogenetically
using the ClustalW alignment program to align the sequences, and a tree was then constructed using the TREECON program (Figure SI-S1 in the Supporting Information). The results show that the microbial communities at the end of the experiments were very different for the two different column treatments, even though the microbial consortium was the same for all columns at the beginning of the experiment. The sequences of clones 1 and 2 in the anaerobic column were 99% similar to Pseudomonas gessardii (P. gessardii). Phylogenetic analysis placed these sequences and the sequence from clone 9, which was 99% similar to Pseudomonas fulgida (P. fulgida), in the Pseudomonas group of the γ-proteobacteria. Pseudomonas species are facultative anaerobes that are able to respire using various electron acceptors, including nitrate (29). P. gessardii was first isolated from spring water in France (30) and is also capable of nitrate reduction. A sequence closely related to that of a Methylobacterium species was also found. Methylobacillus species are methanol oxidizers. Methanol is one of the byproducts of methane oxidation, indicating that methanotrophs may also be present in this microbial community. Several of the clones, e.g. P. gessardii, Caulobacter crescentus, and Brevundimonas vesicularis, are known to be biofilm-producing bacteria (31–33). Clone 1 of the aerobic column, the sequence of which is 97% similar to Sphingomonas paucimobilis (a strictly aerobic known polysaccharide producer), dominates the microbial community within the aerobic column. The presence of Caulobacter and Sphingomonadaceae (both R-proteobacteria) has previously been linked to environments with low available DOC (34). An increase in pressure is clearly seen in the aerobic columns, though to a lesser extent than under the organic carbon-enriched conditions, but is barely recognizable in the anaerobic columns (Figure 4). Like the higher nutrient condition experiments, there appears to be initial microbial activity in the aerobic columns, which in this case, reaches a relatively steady state after the first 5 days. After this there are further smaller oscillations, perhaps as a result of periodic detachment of microbial matter. In contrast to the situation in the carbon-enriched system, the rate at which this equilibrium becomes established is not influenced by grain size. Whether the anaerobic conditions caused the rate of establishment to be slower or the eventual steady-state magnitude of the pressure increase to be less than under aerobic conditions cannot be determined at the time scale of this experiment. Microbial activity was detected in all of the low organic carbon experiments. The images taken from the porous media removed from the aerobic columns (Figure 5) show clear examples of structures formed by microbial EPS both on and between the quartz grains. Some of these structures easily bridge gaps between grains of 20 µm diameter. In particular, the EPS structures visible under aerobic conditions appear to be sticking the individual quartz sand grains together, a process likely to contribute to the greater reductions in Ks evident under these conditions. EPS structures similar to these, also spanning gaps between grains of 20 µm or more, are seen in the images taken from the anaerobic columns. The microbes in these structures are very closely associated with the EPS structures, which are more delicate than those seen under aerobic conditions. As seen in the carbonenriched experiments, these observations also indicate greater reductions in Ks due to microbial EPS under aerobic conditions. The EPS structures formed under aerobic conditions appear to be more substantial and “bulkier” than those in the anaerobic columns. We conclude from this that more substantial microbial activity with the production of significant EPS occurs under aerobic conditions, which is to be expected given that oxygen is a more energetically favorable electron acceptor (35) when compared to alternatives such VOL. 42, NO. 5, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1489
FIGURE 5. ESEM images of quartz grains with microbial materials extracted from columns (500–700 µm) inoculated with a natural microbial consortium and through which synthetic groundwater with a low organic carbon concentration was passed for 40 days under aerobic (left-hand images) and anaerobic (right-hand images) conditions. The images suggest that more substantial EPS structures are produced under aerobic conditions. as nitrate. No conclusions can be drawn from the ESEM results as to the effects of grain size on microbial activity. In conclusion, this study illustrates the nature and extent of microbial activity and its influence on hydraulic conductivity under a variety of conditions relevant to simulating a subsurface environment. Even with low available organic carbon, the microbial communities present were shown to actively respire and to differ in their consumption of electron acceptors, showing differences in community structure that were further investigated and supported by 16s rRNA analysis. Under high organic carbon concentrations (400 mg L-1), although the eventual magnitude of the pressure increase was similar in aerobic and anaerobic systems, the clogging mechanisms were different. Microbial clogging, likely due to a high initial rate of microbial activity and associated EPS production, was observed at the end of the experiment under aerobic conditions. Nonetheless, the distribution and morphology of the anaerobic microbial community and their associated clogging effects are comparable with the clogging effects seen under aerobic conditions. However, the mechanism by which the anaerobic microbial community causes bioclogging seems to be more sensitive to available carbon concentration than the mechanism employed by the aerobic microbial community. Our observations indicate that the bulk hydraulic properties of the media are determined by the micrometer-scale location of the bioclogging, rather than just by the amount of microbial activity. Microbial gas production could not be measured, or imaged and, therefore, its role could not be assessed. We also found that, although grain size does have an affect on bioclogging, there appears to be a threshold carbon concentration (and therefore microbial activity) below which the effects of microbial activity (clogging) are no longer dependent on grain size. The 16s rRNA analyses have the potential to be very useful in determining the clogging 1490
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 5, 2008
mechanisms; however, there is insufficient information about the types of activity shown by the species identified here to fully realize this potential. Similar experiments to those described above but starting with specific cultures may be useful to this end.
Acknowledgments This work was supported by British Nuclear Fuels Ltd. (BNFL). Drs. Paul Humphreys and Ian Beadle of BNFL and Drs. R. Wogelius and C. Merrifield of The University of Manchester provided help and advice. The support of the Natural Environment Research Council in providing infrastructure for the Williamson Research Centre and funding through its “Micro to Macro” programme is also acknowledged. A. Bewsher is thanked for help with HPLC analysis.
Supporting Information Available Text containing experimental details of the DNA extraction and sequencing, clone library construction, RFLP grouping, and phylogenetic analysis, and Figure SI-S1 showing phylogenetic trees constructed for the gene sequences of the microbes listed in Table 2. This information is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Baveye, P.; Vandevivere, P.; Hoyle, B. L.; DeLeo, P. C.; de Lozada, D. S. Environmental impact and mechanisms of the biological clogging of saturated soils and aquifer materials. Crit. Rev. Environ. Sci. Technol. 1998, 28, 123–191. (2) Taylor, S. W.; Jaffe, P. R. Biofilm growth and the related changes in the physical properties of a porous medium. 1. Experimental investigation. 2. Permeability. 3. Dispersivity and model verification. Water Resour. Res. 1990, 26, 2153–2180. (3) Cunningham, A. B.; Characklis, W. G.; Abedeen, F.; Crawford, D. Influence of biofilm accumulation on porous media hydrodynamics. Environ. Sci. Technol. 1991, 25, 1305–1311.
(4) Vandevivre, P.; Baveye, P. Saturated hydraulic conductivity reduction caused by aerobic bacteria in sand columns. Soil Sci. Soc. Am. J. 1992, 56, 1–13. (5) Lappan, R. E.; Fogler, H. S. Reduction of porous media permeability from in situ Leuconostoc mesenteroides growth and dextran production. Biotechnol. Bioeng. 1996, 50, 6–15. (6) Taylor, S. W.; Jaffe, P. R. Substrate and biomass transport in a porous medium. Water Resour. Res. 1990, 26, 2181–2194. (7) Baveye, P.; Vandevivere, P.; De Lozoda, D.; Jaffe, P. J.; Taylor, S. W. Biofilm growth and related changes in the physical properties of a porous medium; 1. Experimental investigation, discussion and reply. Water Resources Res. 1992, 28, 1481–1484. (8) Neu, T.; Swerhone, D. G.; Lawrence, J. R. Assessment of lectinbinding-analysis for in situ detection of glycoconjugates in biofilm systems. Microbiology 2001, 147, 299–313. (9) Neu, T. R.; Swerhone, D. G. W.; Bockelmann, U.; Lawrence, J. R. Effect of carbon, nitrogen and phosphorous on the nature and development of lectin-specific glycoconjugates in lotic biofilms. Aquat. Microb. Ecol. 2005, 38, 283–294. (10) Branda, S. S.; Vik, A.; Friedman, L.; Kolter, R. Biofilms: The matrix revisited. Trends Microbiol. 2005, 13, 20–26. (11) Ahimou, F.; Semmons, M. J.; Haugstad, G.; Novak, P. G. Effect of protein, polysaccharide, and oxygen concentration profiles on biofilm cohesiveness. Appl. Environ. Microbiol. 2007, 73 (9), 2905–2910. (12) Bayer, A. S.; Eftekhar, F.; Tu, J.; Nast, C. C.; Speert, D. P. Oxygen dependent up-regulation of mucoid exopolysaccharide (alginate) production in Pseudomonas aeruginosa. Infect. Immunol. 1990, 58, 1344–1349. (13) Jarman, T. R.; Deavin, L.; Slocombe, S.; Righelato, R. C. Investigation of the effect of environmental conditions on the rate of exopolysaccharide synthesis in Azotobacter vinelandii. J. Gen. Microbiol. 1978, 107, 59–64. (14) Junkins, A. D.; Doyle, M. P. Demonstration of exopolysaccharide production by enterohemorrhagic Escherichia coli. Curr. Microbiol. 1992, 25, 9–17. (15) Ryu, J. H.; Beuchat, L. R. Factors affecting production of extracellular carbohydrate complexes by Escherichia coli O157: H7. Int. J. Food Microbiol. 2004, 95, 189–204. (16) Mao, Y.; Doyle, M. P.; Chen, J. Insertion mutagenesis of wca reduces acid and heat tolerance of enterohemorrhagic Escherichia coli O157:H7. J. Bacteriol. 2001, 183, 3811–3815. (17) Boult, S.; Hand, V. L.; Vaughan, D. J. Microbial controls on metal mobility under the low nutrient fluxes fund throughout the subsurface. Sci. Total Environ. 2006, 372, 299–305. (18) Shaw, J. C.; Bramhill, B.; Wardlaw, N. C.; Costerton, J. W. Bacterial fouling in a model core system. Appl. Environ. Microbiol. 1985, 49 (3), 693–701. (19) Stewart, T. L.; Scott Fogler, H. Biomass plug development and propagation in porous media. Biotechnol. Bioeng. 2001, 72 (3), 353–363.
(20) Raiders, R. A.; McInerney, M. J.; Revus, D. E.; Torbati, H. M.; Knapp, R. M.; Jenneman, G. E. Selectivity and depth of microbial plugging in berea sandstone cores. J. Ind. Microbiol. 1986, 1, 195–203. (21) Rittman, B. E. The significance of biofilms in porous media. Water Resour. Res. 1993, 29, 2195–2202. (22) Allison, D. G. Exopolysaccharide production in bacterial biofilms, Biofilm J. 1998, 3, 1–19. (23) Stroes-Gascoyne, S.; West, J. M. Microbial studies in the canadian nuclear fuel waste management programme. FEMS Microbiol. Rev. 1997, 20, 573–590. (24) Thullner, M.; Zeyer, J.; Kinzelbach, W. Influence of microbial growth in hydraulic properties of pore networks. Transp. Porous Media 2002, 49, 99–122. (25) Fuchs, S.; Hahn, H. H.; Roddewig, J.; Scwarzh, M.; Turkovic´, R. Biodegradation and bioclogging in the unsaturated porous soil beneath sewer leaks. Acta Hydrochim. Hydrobiol. 2004, 32, 277– 286. (26) Holm, J.; Engesgaard, P.; Jensen-Kartsten, H.; Henze, M. Biological clogging experiments and model. Geol. Soc. Am. 1996, 28 (7), 1–130. (27) Knapp, E. P.; Herman, J. S.; Hornberger, G. M.; Mills, A. L. The effect of distribution of iron-oxyhydroxide grain coatings on the transport of bacterial cells in porous media. Environ. Geol. 1998, 33, 243–248. (28) Yee, N.; Fein, J. B. Cd adsorption onto bacterial surfaces: A universal adsorption edge. Geochim. Cosmochim. Acta 2001, 65, 2037–2042. (29) Lloyd, J. R. Microbial reduction of metals and radionuclides. FEMS Microbiol. Rev. 2003, 27 (2–3), 411–425. (30) Verhille, S.; Baida, N.; Daboussi, F.; Harnze, M.; Izard, D.; Leclerc, H. Pseudomonas gessiardii sp. Nov. and Pseudomonas migulae sp. Nov., two new species isolated from natural mineral water. Int. J. Syst. Bacteriol. 1999, 49, 1559–1572. (31) Ude, S.; Arnold, D. L.; Moon, C. D.; Timms-Wilson, T.; Spiers, A. J. Biofilm formation and cellulose expression among diverse environmental Pseudomonas isolates. Environ. Microbiol. 2006, 8 (11), 1997–2011. (32) Entcheva-Dimitrov, P.; Spormann, A. M. Dynamics and control of biofilms of the oligotrophic bacterium Caulobacter crescentus. J. Bacteriol. 2004, 8254, 8266. (33) Ratto, M.; Suihko, M-L.; Siika-aho, M. Polysaccharide-producing bacteria isolated from paper machine slime deposits. J. Ind. Microbiol. Biotechnol. 2005, 32 (3), 109–114. (34) Eiler, A.; Langenheder, S.; Bertilsson, S.; Tranvik, L. J. Heterotrophic bacterial growth efficiency and community structure at different natural organic carbon concentrations. Appl. Environ. Microbiol. 2003, 69 (7), 3701–3709. (35) Madigan, M. T. Brock Biology of Microorganisms, 11th ed.; Prentice-Hall: Upper Saddler River, Pearson, NJ, 2006.
ES072022S
VOL. 42, NO. 5, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1491
