Removal of cyanobacterial metabolites through wastewater treatment plant filters.

Wastewaters have the potential to proliferate excessive numbers of cyanobacteria due to high nutrient levels. This could translate to the production of metabolites, such as the saxitoxins, geosmin and 2-methylisoborneol (MIB), which can impair the quality of wastewater destined for re-use. Biological sand filtration was assessed for its ability to remove these metabolites from a wastewater. Results indicated that the sand filter was incapable of effectively removing the saxitoxins and in some instances, the effluent of the sand filter displayed greater toxicity than the influent. Conversely, the sand filter was able to effectively remove geosmin and MIB, with removal attributed to biodegradation. Granular activated carbon was employed as an alternative filter medium to remove the saxitoxins. Results showed similar removals to previous drinking water studies, where efficient removals were initially observed, followed by a decrease in the removal; a consequence of the presence of competing organics which reduced adsorption of the saxitoxins.

Release and oxidation of cell-bound saxitoxins during chlorination of Anabaena circinalis cells.

Surface water sources are increasingly subject to proliferation of toxic cyanobacteria. Direct chlorination of source water containing toxic cyanobacterial cells for different treatment purposes might cause cell damage and toxin release. There is limited information available on chlorination of saxitoxins (STXs: saxitoxin, C-toxins, and gonyautoxins) produced by Anabaena circinalis. This work: (1) investigated the impact of chlorination on cell lysis and toxin/odor compound release in natural waters; (2) assessed the rates of chlorination of total STXs, and (3) estimated apparent rate constants for STX oxidation in ultrapure and natural waters. With a chlorine exposure (CT) value of 7.0 mg x min/L all cells lost viability causing toxin release. Cell-membrane damage occurred faster than released STXs oxidation. All saxitoxin and more than 95% of other STX analogues were subsequently oxidized. Kinetic analysis of the oxidation of STX analogues revealed significant differences in the susceptibility to chlorine, saxitoxin being the easiest to oxidize. Also, concentrations of trihalomethanes, haloacetic acids, and N-nitrosodimethylamine as chlorination byproducts were respectively <50 µg/L and 11 ng/L even at the highest CT value (50.3 mg x min/L).

Determining the fate of Microcystis aeruginosa cells and microcystin toxins following chloramination.

The cyanobacterium Microcystis aeruginosa can produce potent toxins known as microcystins. While many studies have focussed on the chlorination of microcystin toxins, little work has been conducted with respect to the chloramination of the microcystins. In addition, no studies have been reported on the effect of chloramination on intact Microcystis cells. This study was conducted to determine the fate of M. aeruginosa cells and microcystin toxins following chloramination of a drinking water source. Results indicate that monochloramine could effectively oxidise dissolved microcystin-LR (MCLR) provided high CT values were employed, typically greater than 30,000 mg min L(-1). The decay of MCLR was demonstrated to be a pseudo first-order reaction with rate constants ranging from 9.3x10(-7) to 1.1x10(-5) s(-1) at pH 8.5. However, in the presence of Microcystis cells, monochloramine was ineffective in oxidising microcystin toxins due to the cells exerting a demand on the oxidant. The doses of monochloramine applied (2.8 and 3.5 mg L(-1)) were shown to rapidly release intracellular microcystins into the dissolved state. Flow cytometric analysis of the cells determined that the lower monochloramine dose did not compromise the cell membrane integrity, even though microcystins were rapidly released from the cells. In contrast the higher monochloramine dose resulted in cell membrane disruption with up to 90% of the cells shown to be non-viable after the high dose was applied.

Removal of microcystins from a waste stabilisation lagoon: Evaluation of a packed-bed continuous flow TiO2 reactor.

Photocatalysis has been shown to successfully remove microcystins (MC) in laboratory experiments. Most research to date has been performed under ideal conditions in pure or ultrapure water. In this investigation the efficiency of photocatalysis using titanium dioxide was examined in a complex matrix (waste stabilisation lagoon water). A flow-through photocatalytic reactor was used for the photocatalytic removal of four commonly occurring microcystin analogues (MC-YR, MC-RR, MC-LR, and MC-LA). Up to 51% removal for single MC analogues in waste lagoon water was observed. Similar removal rates were observed when a mixture of all four MC analogues was treated. Although treatment of MC-containing cyanobacterial cells of Microcystis aeruginosa resulted in no decline in cell numbers or viability with the current reactor design and treatment regime, the photocatalytic treatment did improve the overall quality of waste lagoon water. This study demonstrates that despite the presence of natural organic matter the microcystins could be successfully degraded in a complex environmental matrix.

Development of an mlrA gene-directed TaqMan PCR assay for quantitative assessment of microcystin-degrading bacteria within water treatment plant sand filter biofilms.

We report for the first time a quantitative mlrA gene-directed TaqMan PCR assay for the rapid detection of microcystin-degrading bacteria. This was applied, in combination with 16S ribosomal DNA-directed quantitative PCR and denaturing gradient gel electrophoresis, to study virgin sand filter column biofilm development and to correlate mlrA gene abundance with microcystin removal efficiency.

Enhancing the biofiltration of geosmin by seeding sand filter columns with a consortium of geosmin-degrading bacteria.

Geosmin is a secondary metabolite that can be produced by many species of cyanobacteria and Actinomycetes. It imparts a musty/earthy taste and odour to drinking water which can result in consumer complaints and a general perception that there is a problem with the water quality. As geosmin is recalcitrant to conventional water treatment, processes are sought to ensure effective removal of this compound from potable water. Biological filtration (biofiltration) is an attractive option for geosmin removal as this compound has been shown to be biodegradable. However, effective biofiltration of geosmin can be site specific as it is highly dependent upon the types of organism present and there is often an extended acclimation period before efficient removals are achieved. We report here, a novel approach to enhance the biofiltration of geosmin by seeding sand filter columns with a bacterial consortium previously shown to be capable of effectively degrading geosmin. Geosmin removals of up to 75% were evident through sand columns which had been inoculated with the geosmin-degrading bacteria, when compared with non-inoculated sand columns where geosmin removals were as low as 25%. These low geosmin removals through the non-inoculated sand columns are consistent with previous studies and were attributed to physical/abiotic losses. The presence of an existing biofilm was shown to influence geosmin removal, as the biofilm allowed for greater attachment of the geosmin-degrading consortium (as determined by an ATP assay), and enhanced removals of geosmin. Minimal difference in geosmin removal was observed when the geosmin-degrading bacteria were inoculated into the sand columns containing either an active or inactive biofilm.

Biodegradation of geosmin by a novel Gram-negative bacterium; isolation, phylogenetic characterisation and degradation rate determination.

Biologically active sand filters within water treatment plants (WTPs) are now recognised as an effective barrier for the removal of geosmin. However, little is known regarding the actual microbiological processes occurring or the bacteria capable of degrading geosmin. This study reports the enrichment and isolation of a Gram-negative bacterium, Geo48, from the biofilm of a WTP sand filter where the isolate was shown to effectively degrade geosmin individually. Experiments revealed that Geo48 degraded geosmin in a planktonic state by a pseudo-first-order mechanism. Initial geosmin concentrations ranging from 100 to 1000ng/l were shown to directly influence geosmin degradation in reservoir water by Geo48, with rate constants increasing from 0.010h(-1) (R(2)=0.93) to 0.029h(-1) (R(2)=0.97) respectively. Water temperature also influenced degradation of geosmin by Geo48 where temperatures of 11, 22 and 30 degrees C resulted in rate constants of 0.017h(-1) (R(2)=0.98), 0.023h(-1) (R(2)=0.91) and 0.019h(-1) (R(2)=0.85) respectively. Phylogenetic analysis using the 16S rRNA gene of Geo48 revealed it was a member of the Alphaproteobacteria and clustered with 99% bootstrap support with an isolate designated Geo24, a Sphingopyxis sp. previously described as degrading geosmin but only as a member of a bacterial consortium. Of the previously described bacteria, Geo48 was most similar to Sphingopyxis alaskensis (97.2% sequence similarity to a 1454bp fragment of the 16S rRNA gene). To date, this is the only study to report the isolation and characterisation of a Gram-negative bacterium from a biologically active sand filter capable of the sole degradation of geosmin.

Optimising water treatment practices for the removal of Anabaena circinalis and its associated metabolites, geosmin and saxitoxins.

The cyanobacterium Anabaena circinalis has the ability to co-produce geosmin and saxitoxins, compounds which can compromise the quality of drinking water. This study provides pertinent information in optimising water treatment practices for the removal of geosmin and saxitoxins. In particular, it demonstrates that pre-oxidation using potassium permanganate could be applied at the head of water treatment plants without releasing intracellular geosmin and saxitoxins from A. circinalis. Furthermore, powdered activated carbon (PAC) was shown to be an effective treatment barrier for the removal of extracellular (dissolved) geosmin and saxitoxins, with similar adsorption trends of both compounds. The relative removal of the saxitoxins compared with geosmin was determined to be 0.84 +/- 0.27, which implies that saxitoxin removal with PAC can be estimated to be approximately 60 to 100% of the removal of geosmin under equivalent conditions. Chlorine was shown to be effective for the oxidation of the saxitoxins with CT values of approximately 30 mg min l(-1) required for greater than 90% destruction of the saxitoxins.

Diagnosing water treatment critical control points for cyanobacterial removal: Exploring benefits of combined microscopy, next-generation sequencing, and cell integrity methods.

A wide range of cyanobacterial species and their harmful metabolites are increasingly detected in water bodies worldwide, exacerbated by climate change and human activities. The resulting bloom conditions represent significant challenges to production of safe drinking water and cost effective water reuse, therefore their removal is a priority to ensure public safety. While current microscopic taxonomy identification methods provide valuable information about cell numbers during treatment, these methods are incapable of providing information about the fate of cells during treatment. The objectives of this study were to (1) identify the critical control points for breakthrough and accumulation of cells by investigating the fate of cells during treatment processes using a combination of taxonomy, cell integrity and next-generation sequencing (NGS), and (2) assess the impact of pre-treatment processes on breakthrough prevention at critical control points, and the benefits of cell integrity and NGS analysis for improved management purposes. This paper presents the results of an unprecedented cyanobacterial monitoring program conducted in four full scale water treatment plants located in three different climate zones. Cyanobacterial cell integrity and accumulation during operation process were assessed for the first time using next generation of gene sequencing methods. NGS analysis led to detection of cyanobacterial and melainabacteria orders in water samples that were not identified by microscopy. 80 ±â€¯5% of cells were completely lysed post pre-oxidation (for both ozone and potassium permanganate). However unlike pre-ozonation, the remaining cells were undamaged cells with the potential to accumulate and grow within the plants post-KMnO4 treatment, particularly in clarifier sludge. To effectively monitor water quality, this study presents a synergistic approach coupling new and traditional analytical methods and demonstrates the importance of identifying critical points for managing accumulation of cyanobacteria within plants.

Investigating the fate of saxitoxins in biologically active water treatment plant filters.

The saxitoxins are potent neurotoxins, which can be produced by freshwater cyanobacteria. This study assessed the fate of five saxitoxins variants through biologically active laboratory filters containing media sourced from the filters beds of two water treatment plants (WTPs). Decreases in the concentration of the less toxic variants coincided with increases in the concentrations of the more toxic variants through the filters containing anthracite sourced from two different WTPs. No changes in toxin concentrations were evident through parallel filters containing sand. The results strongly suggest that organisms within the biofilm of the anthracite filters possessed the ability to biotransform the saxitoxins variants, which has important implications for drinking water treatment, particularly since this has the potential to increase the toxicity of the filtered water.

Bacterial degradation of microcystin toxins in drinking water eliminates their toxicity.

Microcystin-LR and -LA were readily biodegraded by a bacterium, Sphingpoyxis sp. LH21, in a treated reservoir water. Detection of the microcystins was conducted using high-performance liquid chromatography (HPLC), protein phosphatase 2A (PP2A) inhibition assay and a cell-based cytotoxicity assay. The HPLC results correlated well with the two assays. The decrease in cytotoxicity, coupled with the associated decrease in microcystin concentrations, indicated that no cytotoxic by-products were being generated, highlighting the applicability of biodegradation as a feasible treatment option for effective microcystin removal.

Isolation and identification of a novel microcystin-degrading bacterium from a biological sand filter.

A novel bacterium capable of degrading two microcystin analogues, microcystin-LR and -LA (MCLR and MCLA), was isolated from a biological sand filter which was previously shown to effectively remove these toxins from source waters. Based on phylogenetic analysis of the 16S rRNA gene sequence, the isolated organism, LH21, most likely belonged to the genus Sphingopyxis and of the previously cultured species clustered with Sphingopyxis witflariensis. Using polymerase chain reaction (PCR), isolate LH21 was shown to contain homologues to each of the four genes, mlrA, mlrB, mlrC and mlrD previously associated with the degradation of MCLR by Sphingomonas sp. ACM-3962. Isolate LH21 was able to effectively degrade MCLR and MCLA in batch experiments under environmentally relevant conditions, with complete removal observed within 5h after re-exposure of the toxins.

Discriminating and assessing adsorption and biodegradation removal mechanisms during granular activated carbon filtration of microcystin toxins.

Microcystins are cyanobacterial toxins that are problematic for water authorities due to their resistance to conventional water treatment. Granular activated carbon (GAC) filtration has been shown to be effective in removing microcystin from water using both adsorption and biodegradation removal mechanisms; however, little is known regarding which removal mechanism predominates and to what extent. In this study, microcystin removal due to adsorption and biodegradation in GAC filtration were discriminated and assessed by commissioning three parallel laboratory columns, including a sterile GAC column, a conventional GAC column and a sand column. The results demonstrate that biodegradation is an efficient removal mechanism once it commences and that the rate of biodegradation was dependent upon temperature and initial bacterial concentration. Adsorption of microcystins was prevalent during the initial stages of the GAC columns and was modelled using the homogeneous surface diffusion model (HSDM). The HSDM provided evidence that an active biofilm present on the surface of the conventional GAC hindered adsorption of microcystin compared with the sterile GAC with no active biofilm. Up to 70% removal of microcystin-LR was still observed after 6 months of operation of the sterile GAC column, indicating that adsorption still played a vital role in the removal of this toxin.

Biodegradation rates of 2-methylisoborneol (MIB) and geosmin through sand filters and in bioreactors.

Taste and odour (T&O) causing compounds, in particular, 2-methylisoborneol (MIB) and geosmin, are a problem for water authorities as they are recalcitrant to conventional water treatment. In this study, biological sand filtration was shown to be an effective process for the complete removal of MIB and geosmin, with removal shown to be predominantly through biodegradation. In addition, MIB and geosmin were also effectively degraded in batch bioreactor experiments using biofilm sourced from one of the sand filters as the microbial inoculum. The biodegradation of MIB and geosmin was determined to be a pseudo-first-order reaction with rate constants ranging between 0.10 and 0.58 d(-1) in the bioreactor experiments. Rate constants were shown to be dependent upon the initial concentration of the microbial inoculum but not the initial concentration of MIB and geosmin when target concentrations of 200 and 50 ng l(-1) were used. Furthermore, rate constants were shown to increase upon re-exposure of the biofilm to both T&O compounds. Enrichment cultures with subsequent community profile analysis using 16S rRNA-directed PCR-DGGE identified four bacteria most likely involved in the biodegradation of geosmin within the sand filters and bioreactors. These included a Pseudomonas sp., Alphaproteobacterium, Sphingomonas sp. and an Acidobacteriaceae member.

Treatment challenge of a cyanobacterium Romeria elegans bloom in a South Australian wastewater treatment plant - a case study.

A bloom of the non-toxic cyanobacterium Romeria elegans in waste stabilisation ponds (WSPs) within Angaston waste water treatment plant (WWTP) has posed an unprecedented treatment challenge for the local water utility. The water from the WSPs is chlorinated for safety prior to reuse on nearby farmland. Cyanobacteria concentrations of approximately 1.2 × 106 cells mL-1 increased the chlorine demand dramatically. Operators continuously increased the disinfectant dose up to 50 mg L-1 to achieve operational guideline values for combined chlorine (0.5-1.0 mg L-1) prior to reuse. Despite this, attempts to achieve targeted combined chlorine residual (CCR) failed. In this study, samples from the waste stabilisation pond at Angaston WWTP were chlorinated over a range of doses. Combined chlorine, disinfection by-product formation, cyanobacteria cell concentration, Escherichia coli inactivation, as well as dissolved organic carbon and free ammonia were monitored. This study shows that, in the occurrence of cyanobacterial blooms, CCR does not directly suggest pathogen removal efficiency and is therefore not an ideal parameter to evaluate the effectiveness of disinfection process in WWTP. Instead, E. coli removal is a more direct and practical parameter for the determination of the efficiency of the disinfection process.

Decrease in toxicity of microcystins LA and LR in drinking water by ozonation.

Unchlorinated treated waters from two Australian reservoirs were spiked with microcystin-LA and -LR extracted from a toxic scum of Microcystis aeruginosa. The two waters had considerably different water quality and therefore ozone demands. The spiked sample waters were ozonated using the batch method of ozonation at a range of doses and the samples were analysed for toxins using high-performance liquid chromatography (HPLC). The toxin content of the samples was also determined using a protein phosphatase type 2A inhibition assay (PP2A) and toxicity via the standard mouse bioassay. The HPLC results correlated well with the PP2A results and toxicity tests for both waters. A loss of both toxins and toxicity was observed with increasing ozone dose, resulting in a complete loss of toxicity for both waters once an ozone residual had been achieved. At this ozone residual no toxin was detected using HPLC. The results indicate that microcystins are not transformed into toxic by-products.

Bacterial degradation of microcystin toxins within a biologically active sand filter.

Microcystin toxins are a problem for water authorities as they are recalcitrant to conventional water treatment. In this study, biological sand filtration was assessed in laboratory column experiments for its ability to remove two microcystin analogues, microcystin-LR and microcystin-LA. A lag period of 3 days was evident prior to the commencement of degradation. Contact times were varied during the experiment; however, no microcystin was detected in the effluent after 4 days, even under conditions similar to those of a rapid sand filter. Removals of microcystin through the sand filters were shown to be primarily through biological degradation processes. Using polymerase chain reaction (PCR), biofilm, extracted from one of the sand filters that had effectively removed the microcystins, was shown to contain bacteria with the mlrA gene. Detection of this gene provided additional evidence that biological degradation of microcystin was the primary removal mechanism.

Differences in the chlorine reactivity of four microcystin analogues.

The presence of microcystin toxins in drinking water is highly undesirable as they have the potential to adversely affect human health. Consequently, effective removal of these toxins from water is a major goal for water authorities. In this study, four microcystin analogues were chlorinated in two treated waters, and two of the analogues were chlorinated in deionised water. The oxidation of the microcystins was related to the chlorine exposure (CT) of the sample waters with the ease of oxidation following the trend: microcystin-YR > microcystin-RR > microcystin-LR > or = microcystin-LA. This trend was in agreement with published data on model compounds and free amino acids. Values of CT of up to 25 mg min L(-1) were required for oxidation of all microcystin analogues to below the World Health Organization guideline value of 1.0 microg L(-1). Results from this study indicate that for some water resources it is important to determine the speciation of the microcystin analogues to optimise chlorination practices.

[The characteristic IR spectra of microcystin-LR].

The IR spectra of mincrocystin-LR (mLR) were studied by Fourier transform infrared spectroscopy in the scanning range of 4,000-600 cm(-1). The characteristic IR spectra from main function groups such as monosubstituted benzene, guanidine residues and gamma-carboxylic groups etc have been identified.

Fate of cyanobacteria in drinking water treatment plant lagoon supernatant and sludge.

In conventional water treatment processes, where the coagulation and flocculation steps are designed to remove particles from drinking water, cyanobacteria are also concentrated into the resultant sludge. As a consequence, cyanobacteria-laden sludge can act as a reservoir for metabolites such as taste and odour compounds and cyanotoxins. This can pose a significant risk to water quality where supernatant from the sludge treatment facility is returned to the inlet to the plant. In this study the complex processes that can take place in a sludge treatment lagoon were investigated. It was shown that cyanobacteria can proliferate in the conditions manifest in a sludge treatment lagoon, and that cyanobacteria can survive and produce metabolites for at least 10days in sludge. The major processes of metabolite release and degradation are very dependent on the physical, chemical and biological environment in the sludge treatment facility and it was not possible to accurately model the net effect. For the first time evidence is provided to suggest that there is a greater risk associated with recycling sludge supernatant than can be estimated from the raw water quality, as metabolite concentrations increased by up to 500% over several days after coagulation, attributed to increased metabolite production and/or cell proliferation in the sludge.