|Home | About | Journals | Submit | Contact Us | Français|
We assessed the potential of mixed microbial consortia, in the form of granular biofilms, to reduce chromate and remove it from synthetic minimal medium. In batch experiments, acetate-fed granular biofilms incubated aerobically reduced 0.2 mM Cr(VI) from a minimal medium at 0.15 mM day−1 g−1, with reduction of 0.17 mM day−1 g−1 under anaerobic conditions. There was negligible removal of Cr(VI) (i) without granular biofilms, (ii) with lyophilized granular biofilms, and (iii) with granules in the absence of an electron donor. Analyses by X-ray absorption near edge spectroscopy (XANES) of the granular biofilms revealed the conversion of soluble Cr(VI) to Cr(III). Extended X-ray absorption fine-structure (EXAFS) analysis of the Cr-laden granular biofilms demonstrated similarity to Cr(III) phosphate, indicating that Cr(III) was immobilized with phosphate on the biomass subsequent to microbial reduction. The sustained reduction of Cr(VI) by granular biofilms was confirmed in fed-batch experiments. Our study demonstrates the promise of granular-biofilm-based systems in treating Cr(VI)-containing effluents and wastewater.
Chromium is a common industrial chemical used in tanning leather, plating chrome, and manufacturing steel. The two stable environmental forms are hexavalent chromium [Cr(VI)] and trivalent chromium [Cr(III)] (20). The former is highly soluble and toxic to microorganisms, plants, and animals, entailing mutagenic and carcinogenic effects (6, 22, 33), while the latter is considered to be less soluble and less toxic. Therefore, the reduction of Cr(VI) to Cr(III) constitutes a potential detoxification process that might be achieved chemically or biologically. Microbial reduction of Cr(VI) seemingly is ubiquitous; Cr(VI)-reducing bacteria have been isolated from both Cr(VI)-contaminated and -uncontaminated environments (6, 7, 23, 38, 39). Many archaeal/eubacterial genera, common to different environments, reduce a wide range of metals, including Cr(VI) (6, 16, 21). Some bacterial enzymes generate Cr(V) by mediating one-electron transfer to Cr(VI) (1, 4), while many other chromate reductases convert Cr(VI) to Cr(III) in a single step.
Biological treatment of Cr(VI)-contaminated wastewater may be difficult because the metal's toxicity potentially can kill the bacteria. Accordingly, to protect the cells, cell immobilization techniques were employed (31). Cells in a biofilm exhibit enhanced resistance and tolerance to toxic metals compared with free-living ones (15). Therefore, biofilm-based reduction of Cr(VI) and its subsequent immobilization might be a satisfactory method of bioremediation because (i) the biofilm-bound cells can tolerate higher concentrations of Cr(VI) than planktonic cells, and (ii) they allow easy separation of the treated liquid from the biomass. Ferris et al. (11) described microbial biofilms as natural metal-immobilizing matrices in aqueous environments. Bioflocs, the active biomass of activated sludge-process systems are transformed into dense granular biofilms in sequencing batch reactors (SBRs). As granular biofilms settle extremely well, the treated effluent is separated quickly from the granular biomass by sedimentation (9, 24). Previous work demonstrated that aerobic granular biofilms possess tremendous ability for biosorption, removing zinc, copper, nickel, cadmium, and uranium (19, 26, 31, 32, 40). However, no study has investigated the role of cellular metabolism of aerobically grown granular biofilms in metal removal experiments. Despite vast knowledge about biotransformation by pure cultures, very little is known about reduction and immobilization by mixed bacterial consortia (8, 12, 13, 16, 20, 31, 36). Our research explored, for the first time, the metabolically driven removal of Cr(VI) by microbial granules.
The main aim of this study was to investigate Cr(VI) reduction and immobilization by mixed bacterial consortia, viz., aerobically grown granular biofilms. Such biofilm-based systems are promising for developing compact bioreactors for the rapid biodegradation of environmental contaminants (17, 24, 29). Accordingly, we investigated the microbial reduction of Cr(VI) by aerobically grown biofilms in batch and fed-batch experiments and analyzed the oxidation state and association of the chromium immobilized on the biofilms by X-ray absorption near edge spectroscopy (XANES) and extended X-ray absorption fine structure (EXAFS).
Aerobic granular biofilms were grown in a 3-liter working-volume laboratory-scale sequencing batch reactor (SBR). SBR setup and operation details have been described previously (26, 27). The SBR was inoculated with seed sludge collected from the outlet of an aeration tank of an operating domestic wastewater treatment plant at Kalpakkam, India. The reactor was operated at room temperature (30 ± 2°C) at a volumetric exchange ratio of 66% and a 6-h cycle, comprising 60 min of anaerobic static fill, 282 min of aeration, 3 min of settling, 10 min of effluent decantation, and 5 min of being idle. The SBR was fed with acetate-containing synthetic wastewater as discussed by Nancharaiah et al. (27). Granules, collected 2 months after the reactor's start-up, were washed twice with ultrapure water, and used in the bioreduction experiments. The morphology of the granular biofilms was documented with a DP70 digital camera (Olympus, Japan) connected to a stereo zoom microscope SMZ1000 (Nikon, Japan). The particle size and circularity of the granular biofilms were determined using the image analysis software Image J 1.99 (26). The settling velocity and dry weight of the aerobically grown granular biofilms were determined according to standard methods (3). The biofilm density was evaluated following the method of Beun et al. (5). Individual granular biofilms were fixed in 2.5% glutaraldehyde and dehydrated successively in 3-min steps with 50%, 80%, and 95% ethanol. Then, the biofilms were sputter-coated and imaged using scanning electron microscopy (SEM; Philips ESEM).
Chromate reduction experiments were carried out in acetate minimal medium (AMM) consisting of 1.0 g liter−1 NH4Cl, 0.2 g liter−1 MgSO4·7H2O, 3.02 g liter−1 CH3COONa, 0.5 g liter−1 KH2PO4, and 0.5 g liter−1 yeast extract. The pH of medium was adjusted to 7.0 with 0.1 N HCl before autoclaving it. To avoid precipitation, the MgSO4 stock solution (50×) was autoclaved separately and added to the autoclaved media. A stock solution of potassium dichromate [1,000 mg Cr(VI) per liter] was prepared in ultrapure water, filter sterilized, and used as required. Cr(VI) reduction was carried out under aerobic and anaerobic conditions. For the aerobic reduction of Cr(VI), 250-ml Erlenmeyer flasks containing 100 ml of AMM with chromate and the granular biofilms (wet weight, 2.5, 5, 10, and 15 g) were incubated on a rotary shaker at 100 rpm at 30°C. For anaerobic reduction, 100-ml glass bottles containing 100 ml of AMM with chromate and microbial granules were bubbled with nitrogen gas and sealed with rubber stoppers incubated at 30°C without shaking. Liquid samples were collected at regular intervals and analyzed for Cr(VI). Batch experiments also were carried out to determine the effect on Cr(VI) reduction, of chromate loading (0.2, 0.4, 0.6, 0.8, 1.0, 1.5, 2.0, and 3.0 mM) and granular biomass content (wet weight, 5, 10, 15, and 20 g 100 ml−1).
For the fed-batch experiments, Schott Duran bottles containing 100 ml of AMM amended with 0.25 mM Cr(VI) and inoculated with granules (wet weight, 10 g) were used. The bottles were incubated at 30°C without aeration. Samples were collected periodically and monitored for Cr(VI). When almost all of the Cr(VI) was removed from the medium, it was replaced with fresh sterile AMM (100% exchange) and amended with Cr(VI). This procedure was repeated up to four times. The Cr(VI) content of the liquid samples collected at different times during each batch was determined.
Before analysis, the samples were centrifuged at 10,000 rpm for 5 min to remove suspended cells. Chromate-reducing ability was estimated as the decrease in the Cr(VI) concentration in medium detected via the Cr(VI)-specific colorimetric reagent S-diphenylcarbazide (DPC) (3). A 0.25% (wt/vol) solution of DPC was prepared in acetone-H2SO4 to minimize its deterioration and stored at 4°C. When added to samples containing Cr(VI), a pink color developed; absorbance was measured immediately at 540 nm using a UV-visible spectrophotometer (Shimazdu, Japan).
Granular biofilms exposed to Cr(VI) and incubated aerobically or anaerobically were freeze-dried and the Cr oxidation state was determined by X-ray absorption near edge spectroscopy (XANES). XANES analyses was performed at the Cr K edge in the fluorescence mode employing a 13-element Ge detector on beamline X10C at the National Synchrotron Light Source (NSLS), Brookhaven Laboratory, Upton, NY. The samples were placed on an Al sample holder with a cutout geometry of 2 mm (height) × 20 mm (length) × 1.5 mm (thickness) and sealed with Kapton tape. Standards included Cr6+ (potassium chromate; K2CrO4) and Cr3+ [chromium hydroxide; Cr(OH)3]. Six spectra per sample were collected, from 200 eV below the absorption edge to 300 eV above it. Data were acquired in the XANES region at an energy step of 0.5 eV at 2.0 s per interval. A chromium metal foil sited in the reference channel was run simultaneously with each sample to monitor shifts in the beamline's energy due to possible reduction by X rays. The software programs ATHENA and AUTOBACK (34) were used to analyze the XANES data, which included background subtraction and normalization of the signal to the edge jump. The oxidation state of Cr was derived from the position of the first derivative of the absorption edge energy.
EXAFS analyses were performed to determine the association of Cr with its nearest neighbor in the granular biofilms. Samples and standards were prepared as for the XANES analysis. A Cr(III) phosphate sample was included to obtain fitting parameters for the analysis. The individual scans were averaged followed by linear pre-edge subtraction, background removal, normalization to the step edge, isolation of the χ(k) function with a cubic spline function, followed by k2 weighting. Theoretical EXAFS amplitude and phase functions for Cr-O, and Cr-P single scattering paths were then generated by FEFF 6.0 (34). Fitted parameters such as amplitude reduction factor (S02), (ΔE0), interatomic distance (R), and Debye-Waller factor (σ2) were fitted in R-space. The amplitude reduction value was as 1.0 for all fits. Errors in the overall fits were determined using a goodness-of-fit parameter (34). The spectra were combined and normalized to the edge jump, using programs from the University of Washington. The software programs ATHENA and ARTEMIS were used to process the spectra through a multistep data analysis procedure that included background subtraction and Fourier transformation.
Figure Figure11 shows the stereo zoom microscopic images of the morphology of the granular biofilms formed in the SBR and the scanning electron microscopic images of the bacteria comprising them. The granular biofilms were dense and nearly circular (Fig. (Fig.1A),1A), comparable to those reported in other studies on aerobic granulation (5, 24). The average size of the biofilms was 1.8 mm, and the average circularity was 0.89 mm. The settling velocity and density of aerobic granular biofilms were 55 m h−1 and 40 g liter−1, respectively. Visualization by scanning electron microscopy revealed that rod/coccus-shaped bacteria dominate the acetate-fed granular biofilms (Fig. (Fig.1B1B).
To determine the mechanism of Cr(VI) removal, the changes in Cr(VI) was monitored in four batch experiments: (i) without granular sludge; (ii) with lyophilized granular sludge; (iii) with intact granular sludge but no electron donor; and, (iv) with intact granular sludge with the electron donor, acetate. There was no decrease in the concentration of Cr(VI) in the cell-free control. Similarly, we found no significant change in the Cr(VI) concentration in the media in the absence of an electron donor or with lyophilized granular sludge (Fig. (Fig.2).2). These results confirm the role of bacterial cell metabolism in the process. The ability of the microbial granules to reduce Cr(VI) in the presence of the electron donor was revealed by a clear decolorization (yellow to colorless) of the media and verified in Cr(VI)-specific DPC absorbance measurements. Cr(VI) reduction was observed under both aerobic and anaerobic conditions. Although Cr(VI) reduction is similar in both (0.15 mM day−1 g−1 aerobically versus 0.17 mM day−1 g−1 anaerobically), the Cr(VI) concentration did not fall to zero under aerobic conditions (Fig. (Fig.2).2). In contrast, anaerobic Cr(VI) reduction was almost complete (Fig. (Fig.3).3). Our subsequent work was focused on anaerobic removal of Cr(VI) by granular biofilms.
Figure Figure33 illustrates the extent of anaerobic reduction of Cr(VI) by granular biofilms at different initial biomass concentrations: reduction was almost complete at all concentrations, but its rate varied with the content of the granular biofilm in the medium. Complete removal of 0.2 mM Cr(VI) took approximately 2 to 6 days, depending on the available biomass (Fig. (Fig.3).3). Figure Figure44 depicts the effect of Cr(VI) loading on its reduction by the granular biofilms. Cr(VI) content fell at all initial Cr(VI) concentrations tested and was nearly complete, irrespective of the initial concentration. However, the time taken for complete removal rose with an increase in the initial Cr(VI) concentration. The specific rates of Cr(VI) removal were, respectively, 0.17 mM day−1 g−1, 0.65 mM day−1 g−1, and 1.86 mM day−1 g−1 granular biomass at 0.2, 1.0, and 3.0 mM initial Cr(VI) concentration.
Figure Figure55 shows the removal of Cr(VI) by microbial granules in fed-batch experiments; removal approached completion in each batch and was sustained in subsequent batches. The average time for complete removal of 0.2 mM Cr(VI) was approximately 2 to 4 days for each batch. Undoubtedly, the microbial granules repeatedly can sustain the removal of Cr(VI) in fed-batch experiments.
Figure Figure66 displays the XANES spectra for the standards and samples. The first derivatives of the absorption edge energy for Cr(VI) and Cr(III) standards, respectively, are at 6005.6 eV and 6003.3 eV. In addition, the Cr(VI) standard exhibited a pre-edge peak at 5993.3 eV due to the tetrahedral coordination (1s > 3d transition) of the chromate-oxygen atoms. Under anaerobic conditions, the absorption edge energy for Cr associated with the granules shifted to 6003.8 eV, while under aerobic conditions, it moved to 6003.4 eV. This shift to lower energies compared to the Cr(VI) standard and the absence of a pre-edge peak indicate the Cr(VI) predominantly was reduced to Cr(III).
Figure Figure77 and Table Table11 give the results of fitting for Cr(III) phosphate- and Cr-containing granules (1 mM). The best-fit parameters for Cr(III) phosphate indicate that 5.7 ± 0.7 O atoms surround the octahedral Cr at 1.97 ± 0.01 Å. A fitting of the second shell shows the presence of 4.0 ± 1.3 phosphorus atoms at a distance of 3.11 ± 0.05 Å, indicating its presence as Cr(III) phosphate. Fitting of the Cr-laden granular biofilm is similar to that of the Cr(III) phosphate standard and shows the presence of 6.3 ± 1.5 O atoms in its inner sphere. The second shell has 4.0 ± 1.3 P atoms at a distance of 3.11 ± 0.03 Å from the chromium. These fitting results confirm that Cr associated with the granular biofilms is present primarily as Cr(III) phosphate; all concentrations of Cr (1.0, 1.5, and 2.0 mM) showed a similar association. EXAFS data fitted using C, N, and P in the second shell showed a much better fit using P.
Bacterial reduction of Cr(VI) was demonstrated in pure cultures and mixed species activated sludge in flowthrough columns (37). However, there are no reports on the bioreduction of chromium using microbial granules that offer very specific advantages over activated sludge. Confocal microscopic imaging after staining with nucleic-acid-binding fluorophores disclosed that granular biofilms consist of distinct cell clusters separated by voids (25-27). Microscopic imaging of these clusters revealed that rod/coccus-shaped bacteria dominated them (Fig. (Fig.1B).1B). Earlier work showed that bacteria in the form of biofilms may be more suitable for chromium bioreduction than to freely suspended cells. Thus, Morales et al. (23) reported that Streptomyces sp. strain CG252 cells grown as biofilms better removed Cr(VI) than did free cells. Similarly, Ganguli and Tripathy (14) reported that Pseudomonas aeruginosa A2Chr cells immobilized in an agarose-alginate film in a rotating biological contactor exhibited significantly higher rates of chromate reduction than did planktonic cells.
Cr(VI) removal by the granular biofilms was observed both under aerobic and anaerobic conditions. However, the incomplete removal of Cr(VI) under aerobic conditions might reflect competition between Cr(VI) and oxygen for electrons and electron transfer from intermediate Cr(V) to oxygen, resulting in Cr(VI). We observed that Cr(VI) reduction was dependent on the initial content of biomass, as have others (16). Furthermore, others found that using pregrown cellular biomass [cultivated without Cr(VI)] greatly lowers the time required for complete chromate reduction (35). Apparently, employing pregrown aerobic granular biofilms to reduce Cr(VI) avoids the metal's negative impact on the growth of the microbial biomass. The microbial species composition of the granular sludge was not identified in the present study; nonetheless, the commencement of reduction of chromium immediately after exposure to Cr(VI) suggests that bacteria able to reduce chromium already were present in the granules (without prior enrichment); the lack of a delay demonstrates that the necessary enzymes are constitutively expressed. Seemingly, previous exposure to chromium and subsequent microbial enrichment are not prerequisites for successful bioreduction. This could be mainly due to the involvement of constitutive chromate reductases, thus corroborating the earlier observation of the rapid reduction of Cr(VI) by Pseudomonas putida unsaturated biofilms (32).
Aerobic granular sludge cultivated in an SBR using acetate and lacking prior exposure to chromium efficiently reduced Cr(VI) from minimal media. Passive biosorption by the granular biomass was ruled out because Cr(VI) removal was negligible in the absence of a carbon source and by lyophilized granules. Analysis of chromium speciation by XANES further confirmed the bioreduction of Cr(VI) to Cr(III), thereby pointing to the involvement of cell metabolism. Nonmetabolic reduction of Cr(VI) to Cr(III) by bacterial surfaces under nonnutrient conditions has been reported by Fein et al. (10). In this study, no such reduction of Cr(VI) to Cr(III) was observed under nonnutrient conditions. EXAFS analyses revealed that the granular biofilm-bound Cr(III) occurs as Cr(III) phosphate. Earlier, Neal et al. (28) reported that only Cr(III) was bound to live Shewanella oneidensis cells. XANES and EXAFS analyses of a Cr(III)-laden biomass of nonliving seaweed, Ecklonia, were very similar to spectra from Cr(III) acetate (30). Kemner et al. (18) reported that the speciation of chromium associated with Pseudomonas fluorescens cells was consistent with association of Cr(III) with a phosphoryl functional group. A recent study showed reduction of Cr(VI) to Cr(III) by methane-oxidizing bacteria, a ubiquitous group of environmental bacteria (2). EXAFS analysis showed that Methylococcus capsulatus-associated chromium predominantly existed as Cr(III) and most likely associated with phosphate groups. EXAFS spectra of our Cr(III)-laden granular biomass revealed the presence of Cr(III)-phosphate after Cr(VI) reduction. Overall, our findings suggest the potential use of mixed microbial granules to bioremediate Cr(VI)-containing wastewater or industrial effluents.
This research was supported by the Department of Atomic Energy, Government of India, and in part by the Environmental Remediation Sciences Division, Office of Biological and Environmental Research, Office of Science, U.S. Department of Energy under contract no. DE-AC02-98CH10886. Y.V.N. gratefully acknowledges the American Society for Microbiology for the Indo-US Visiting Research Professorship Award.
We thank Avril D. Woodhead for editorial help.
Published ahead of print on 19 February 2010.