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

water research 43 (2009) 2927–2935

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/watres

Biodegradation of geosmin by a novel Gram-negative bacterium; isolation, phylogenetic characterisation and degradation rate determination Daniel Hoefela,b,*, Lionel Hoa,b, Paul T. Monisa, Gayle Newcombea, Christopher P. Sainta,b a

Australian Water Quality Centre, South Australian Water Corporation, Adelaide, South Australia, Australia School of Earth and Environmental Sciences, The University of Adelaide, Adelaide, South Australia, Australia

b

article info

abstract

Article history:

Biologically active sand filters within water treatment plants (WTPs) are now recognised as

Received 20 January 2009

an effective barrier for the removal of geosmin. However, little is known regarding the

Received in revised form

actual microbiological processes occurring or the bacteria capable of degrading geosmin.

1 April 2009

This study reports the enrichment and isolation of a Gram-negative bacterium, Geo48,

Accepted 2 April 2009

from the biofilm of a WTP sand filter where the isolate was shown to effectively degrade

Published online 17 April 2009

geosmin individually. Experiments revealed that Geo48 degraded geosmin in a planktonic state by a pseudo-first-order mechanism. Initial geosmin concentrations ranging from 100

Keywords:

to 1000 ng/l were shown to directly influence geosmin degradation in reservoir water by

Biodegradation

Geo48, with rate constants increasing from 0.010 h1 (R2 ¼ 0.93) to 0.029 h1 (R2 ¼ 0.97)

Biofiltration

respectively. Water temperature also influenced degradation of geosmin by Geo48 where

Geosmin

temperatures of 11, 22 and 30  C resulted in rate constants of 0.017 h1 (R2 ¼ 0.98), 0.023 h1

Rate constant

(R2 ¼ 0.91) and 0.019 h1 (R2 ¼ 0.85) respectively. Phylogenetic analysis using the 16S rRNA

Sand filter

gene of Geo48 revealed it was a member of the Alphaproteobacteria and clustered with 99%

Taste and odour

bootstrap support with an isolate designated Geo24, a Sphingopyxis sp. previously described

Water treatment

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 1454 bp 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. Crown Copyright ª 2009 Published by Elsevier Ltd. All rights reserved.

1.

Introduction

Climatic conditions around the world are changing such that more extreme events of flooding and drought, and a general increase in ambient temperatures, are favouring both the occurrence and intensity of blue–green algae (cyanobacteria) blooms (Paerl and Huisman, 2008). Of continued concern to

the water industry is the production of secondary-algal metabolites during such blooms, in particular the taste and odour compound geosmin (trans-1,10-dimethyl-trans-9-decalol). When present in dissolved form, geosmin is recalcitrant to conventional water treatment processes (Rittmann et al., 1995) where alternative and more advanced treatment options such as oxidation by chlorine or ozone are also not entirely

* Corresponding author. Australian Water Quality Centre, South Australian Water Corporation, 250 Victoria Square, Adelaide, South Australia, 5000, Australia. Tel.: þ61 8 7424 2142; fax: þ61 8 7003 2142. E-mail address: [email protected] (D. Hoefel). 0043-1354/$ – see front matter Crown Copyright ª 2009 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2009.04.005

2928

water research 43 (2009) 2927–2935

effective (Glaze et al., 1990; Ho et al., 2002). Also, treatment of geosmin by activated carbon is often hindered by the presence of other far more abundant natural organic matter (NOM) within the source water (Cook et al., 2001). As geosmin has an earthy taste and odour at a low threshold of approximately 2 ng/l (Young et al., 1996), penetration through water treatment and into distribution systems often results in consumer complaints and a general perception by the public that the water is unsafe to drink. Biologically active sand filters have recently been recognised as a barrier for the enhanced removal of geosmin through conventional water treatment (Ho et al., 2007a), and have been applied full-scale across the USA (Metz et al., 2006), Europe (Uhl et al., 2006) and Australia (McDowall et al., 2007), where degradation of geosmin by bacteria within biofilm is believed to be an efficient treatment process (Namkung and Rittmann, 1987). The application of biologically active sand filters is favoured by water utilities because the process is generally low technology, requires little maintenance and often has already established infrastructure. Another advantage is that it is a process free of any chemical addition. Previous studies have demonstrated that either sand (Ashitani et al., 1988; Lundgren et al., 1988) or granular activated carbon filters (Elhadi et al., 2004) are suitable solid support matrices for the attachment of biofilmassociated bacteria involved in the biofiltration of geosmin. Until recently, the biological degradation of geosmin had only been reported by Gram-positive bacteria. These included strains of Bacillus cereus (Silvey et al., 1970; Narayan and Nunez, 1974), B. subtilis (Narayan and Nunez, 1974), Arthrobacter atrocyaneus, A. globiformis, Chlorophenolicus strain N-1053 and Rhodococcus maris (Saadoun and El-Migdadi, 1998). However, demonstration of geosmin degradation in many of these studies was conducted under conditions of little significance to what may be encountered within biofilters. More recently we successfully reported, for the first time, the involvement of a consortium of three Gram-negative bacteria that were shown to cooperatively degrade geosmin at environmental significant concentrations ranging 37–131 ng/l (Hoefel et al., 2006). Those consortium members included Sphingopyxis sp. Geo24, Novosphingobium sp. Geo25 and Pseudomonas sp. Geo33, which were isolated from the biofilm of a sand filter within a South Australian Water Treatment Plant (WTP). It has been subsequently demonstrated that the seeding of the Gram-negative bacterial consortium onto laboratory-scale sand filters enhanced the removals of geosmin (McDowall et al., 2009), and this has the potential for full-scale enhancement of geosmin removal during water treatment. The process may be further enhanced by the application of a single bacterium that has the capacity to degrade geosmin individually rather than relying on a consortium, which may degrade geosmin by a more complex process. To date, there has not been an individual Gram-negative bacterium described in the literature with the capacity to degrade geosmin. In addition, there are knowledge gaps regarding the mechanism by which geosmin is biodegraded by bacteria; only one reported study has attempted to investigate the various biodegradation products formed as part of the microbiological breakdown of geosmin and this

study was conducted under ideal conditions, far removed from those that would be experienced in water storages (Saito et al., 1999). Isolation of a single Gram-negative bacterium with the capacity to degrade geosmin would allow for more simplified molecular-based approaches to elucidate the precise mechanisms that occur within bacteria during the degradation of the taste and odour compound and also possibly allow optimisation of the process. The objectives of this study are to isolate a single Gram-negative bacterium which has the capacity to degrade geosmin, and subsequently investigate the effect of geosmin concentration, water temperature and initial cell abundance upon the rates of geosmin removal.

2.

Experimental procedures

2.1.

Geosmin analysis

Geosmin was purchased as a racemic mixture (Ultrafine Chemicals, UK) and stock solutions were prepared in Milli-Q water. All samples for geosmin analysis were pre-concentrated using a solid-phase microextraction syringe fibre (Supelco, Australia). Geosmin analysis was performed in parallel against previously quantified internal standards, using a 5890 Series II Gas Chromatograph with a 5971 Series Mass Selective Detector (Hewlett Packard, Melbourne, Australia) as reported previously (Hayes and Burch, 1989).

2.2.

Bacterial enumeration

Numbers of active bacteria were determined by staining with the BacLight bacterial viability kit (Molecular Probes, Eugene, OR, USA) and subsequent enumeration by flow cytometry (FCM) (FACSCalibur, Becton Dickinson, San Jose, CA, USA) as described previously (Hoefel et al., 2003).

2.3.

Isolation of a geosmin-degrading bacterium

Sand was collected from the filter beds of Morgan WTP, South Australia. Biofilm was released from the sand medium by 15 min of periodic vortexing in Bushnell–Haas (BH) minimal medium (0.1% (w/v) NH4NO3, 0.1% (w/v) K2HPO4, 0.1% (w/v) KH2PO4, 0.01% (w/v) Mg SO4.7H2O, 0.01% (w/v) FeCl3.6H2O and 0.001% (w/v) CaCl2.2H2O). Bacteria within the supernatant were then washed twice by centrifugation at 1000  g for 15 min with re-suspension of the bacterial pellet each time in BH medium. The numbers of active bacteria within the supernatant were then determined by FCM. Enrichment of geosmin-degrading bacteria was carried out as described previously (Hoefel et al., 2006). Briefly, 1  106/ml of active bacteria (final cell concentration) were inoculated in sterile BH medium, supplemented with 28 mg/l of geosmin (final culture volume of 20 ml). Two additional control cultures included an equivalent number of active bacteria inoculated into BH minimal medium without geosmin, and an equivalent portion of inactivated bacteria (autoclaved at 121  C for 20 min) inoculated into BH medium supplemented with 28 mg/l of geosmin. Each culture was incubated at 22  C with shaking at 100 rpm for 63 days.

2929

water research 43 (2009) 2927–2935

2.4. Characterisation of geosmin degradation by Geo48 under various conditions

2.5.

Phylogenetic analysis of Geo48

The DNA sequence of a fragment of the 16S rRNA gene from Geo48 was determined following PCR-amplification with the primer set 27F/1492R as reported previously (Hoefel et al., 2005). Sequence similarity searches were conducted using the National Center for Biotechnology Information BLAST network service (blastn). Similar sequences, from previously cultured bacteria, were obtained from GenBank and aligned against the DNA sequences of Geo48 using ClustalX version 1.64b software (Thompson et al., 1997). Neighbour-joining analysis with 1000 bootstrap replicates was then performed using the software program MEGA (MEGA version 2.1; Arizona State University, Tempe, AZ, USA).

3.

Results and discussion

3.1. Enrichment of geosmin-degrading bacteria and isolation of Geo48 Geosmin-degrading bacteria within the sand filter biofilm underwent enrichment in minimal BH medium supplemented with geosmin as the sole carbon source. As shown in Fig. 1, measurable geosmin degradation appeared to begin following day 35, with the maximum amount of geosmin degradation occurring between days 42 and 56. By day 56 a total of 21.5 mg/l of geosmin had been degraded, resulting in a 1.78 log10 increase in active bacterial abundance.

6.0x107

32 28

5.0x107

24 4.0x107

20

3.0x107

16 12

2.0x107

8 1.0x107

4 0

10

20

30

40

50

60

Geosmin concentration (mg/l)

Surface water, collected from the Myponga Reservoir in South Australia, was filtered through an AcroPak 0.2 mm nominal filter cartridge (Pall Corporation, USA), then sterilized by autoclaving (121  C for 20 min). Sterility was verified by the absence of any active bacteria detected using the FCM method and by the absence of bacterial growth on R2A medium following 7 days incubation at 22  C. Characteristics of the reservoir water following autoclaving included dissolved organic carbon (DOC) 11.9 mg/l, UV254 0.362 cm1, colour 44 HU and pH 8.1. Geo48 was cultured in R2A liquid medium for 65 h at 22  C with continuous shaking. The culture was then washed by centrifugation at 3500  g for 15 min followed by re-suspension of the cell pellet in the sterile reservoir water. To study the effect of initial geosmin concentration upon its degradation by Geo48, 500 ml of sterile reservoir water was spiked separately with 100, 500 and 1000 ng/l of geosmin and inoculated with final active Geo48 cell numbers equivalent to 1  106/ml. Incubation was performed at 22  C (1  C) for 7 days with continuous stirring. A control culture consisted of 500 ng/l geosmin in 500 ml reservoir water but with 1  106/ml inactivated Geo48 cells. For experiments investigating the effect of Geo48 cell numbers upon the degradation of geosmin, 500 ml of sterile reservoir water was inoculated separately with 1  104, 1  105 and 1  106/ml of active Geo48 cells. Each culture was then spiked with 500 ng/l of geosmin and incubated at 22  C for 7 days with continuous stirring. A control culture consisted of 500 ng/l geosmin in 500 ml reservoir water but with 1  106/ml inactivated Geo48 cells. For experiments studying the effect of incubation temperature upon the rate of geosmin degradation by Geo48, three vessels of the sterile reservoir water (500 ml) were pre-acclimatised and spiked with 500 ng/l of geosmin and 1  106/ml of active Geo48 cells. Incubation was performed separately at 11, 22 and 30  C (1  C) for 10 days with continuous stirring. A control culture consisted of 500 ng/l geosmin in 500 ml reservoir water, 1  106/ml inactivated Geo48 cells, with incubation performed at 30  C (1  C). For each of the cultures described above,

samples were collected in duplicate at various time points for the assessment of geosmin concentration.

Active bacteria/ml

During enrichment, samples were periodically taken for geosmin analysis by gas chromatography-mass spectrometry (GC-MS; duplicate analyses), bacterial enumeration by FCM (triplicate analyses), and community profiling by denaturing gradient gel electrophoresis (DGGE) of the 16S rRNA gene (in duplicate) as described previously (Hoefel et al., 2006) to identify bacteria that became predominant during the enrichment culture. Subsequent isolation of the predominant bacteria identified by DGGE was achieved by inoculating tenfold dilutions of the day 63 enrichment culture onto solid R2A medium, using the spread plate technique, and selecting predominant colony types that had formed following 7 days of incubation at 22  C. All colony types were then analysed individually by DGGE, in parallel with the day 63 sample from the enrichment culture. A DGGE band from isolate Geo48 revealed co-migration with a predominant band from the enrichment culture, and DNA sequence analysis of both bands was used as confirmation of identity.

0 70

Time (d) Fig. 1 – Active bacterial numbers and geosmin concentration remaining within the enrichment culture and the associated control cultures. Open triangles (6) represent the geosmin concentration remaining within the enrichment culture; closed triangles (:) represent the geosmin concentration remaining within the control culture containing inactive bacteria; open squares (,) represent active bacterial numbers within the enrichment culture; and closed squares (-) represent active bacterial numbers within the control culture containing no carbon source. Error bars represent 95% confidence intervals.

2930

water research 43 (2009) 2927–2935

3.2. Effect of initial geosmin concentrations upon geosmin degradation

Day 0 M

Day 28

Day 42

Day 49

Day 63

Geo48

Variations in geosmin concentration upon the degradation of the taste and odour compound are shown in Fig. 3. The linear relationship between the natural log value of the fraction of geosmin remaining versus time (Fig. 3 inset) were consistent with that of a pseudo-first-order mechanism of degradation. This is in agreement with the findings of Ho et al. (2007a), who reported a pseudo-first-order mechanism for the biodegradation of geosmin by a mixed sand filter biofilm community suspended within a bioreactor. This mechanism of degradation has also been supported by Schmidt et al. (1985) and

M

Fig. 2 – Denaturing gradient gel electrophoresis (DGGE) profiles of 16S rRNA gene fragments from bacteria within the enrichment culture at various stages of incubation (left panel) and from Geo48 (right panel). M; DNA marker.

Effect of [Geosmin]0 -1

ln([Geosmin]/[Geosmin]°)

1300 1200

Geosmin concentration (ngl/)

DGGE was used to identify bacteria that became predominant within the enrichment culture, as such organisms were most likely responsible for the degradation of geosmin described in Fig. 1. DGGE analysis revealed a distinct change in community composition at day 28 compared with day 0 (Fig. 2 left panel). Following day 28, the community composition appeared to remain relatively constant despite a 1.78 log10 increase in bacterial abundance. By day 63 there appeared to be three predominant DGGE bands and these were hypothesised as representing bacteria that were involved in the degradation of geosmin within the enrichment culture (Fig. 2 left panel). The 16S rRNA gene fragments of Geo48, isolated on solid R2A medium from the enrichment culture, co-migrated with the band highlighted on the DGGE gel (Fig. 2). Band excision and subsequent DNA sequence analysis of the co-migrating bands (both the day 63 sample band and the Geo48 isolate band) confirmed that the Geo48 isolate was indeed represented by the DGGE band highlighted within the day 63 enrichment culture (Fig. 2). Interestingly, the absence of the Geo48 band on day 0 suggested that this organism was not dominant within the initial enrichment culture inoculum, and therefore may not have been a predominant member within the original sand filter biofilm. Attempts were also made to isolate the bacteria represented by the other predominant DGGE bands but these were unsuccessful.

1100 1000 900 800 700 600 500

2

100ng/l (k=0.0104h , R =0.93) -1 2 500ng/l (k=0.0241h , R =0.99) -1 2 1000ng/l (k=0.02914h , R =0.97)

0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 -3.0 -3.5 -4.0 -4.5 -5.0 -5.5 -20

0

20

40

60

80 100 120 140 160 180

Time (h)

400 300 200 100 0 0

20

40

60

80

100

120

140

160

180

Time (h) Fig. 3 – The effect of initial target geosmin concentration of 100 ng/l (-), 500 ng/l (C), and 1000 ng/l (:) upon the degradation of the taste and odour compound by Geo48 in sterilized reservoir water. Inset: Pseudo-first-order kinetic plots using data from the removal curve. Error bars represent 95% confidence intervals.

Anderozzi et al. (2006), who described the mechanism for the aerobic biodegradation of organic compounds, when present as secondary substrates, being of pseudo-first-order. In our study here, geosmin (100–1000 ng/l) was most likely degraded by Geo48 as a secondary substrate when compared with the far more abundant NOM present (DOC of 11.90 mg/l) within the reservoir water, which is most likely the primary substrate(s) (Rittmann et al., 1995). It is also anticipated that geosmin is biodegraded as a secondary substrate within WTP sand filter biofilm and within surface waters following blooms of geosmin-producing algae (Ho et al., 2007a). Rate constant data revealed that the higher initial geosmin concentrations resulted in a more rapid degradation of the compound, with rate constants of 0.010 h1 (equivalent to 0.25 d1; R2 ¼ 0.93) for 100 ng/l geosmin, 0.024 h1 (equivalent to 0.58 d1; R2 ¼ 0.99) for 500 ng/l geosmin and 0.029 h1 (equivalent to 0.67 d1; R2 ¼ 0.97) for 1000 ng/l geosmin (Fig. 3 inset). No loss of geosmin was observed within the control culture (data not shown). Ho et al. (2007a) previously reported rate constants for the bacterial degradation of geosmin but by a mixture of sand filter biofilm-associated bacteria suspended in WTP treated water. In that study, there was little increase in rate constants from 0.21 to 0.24 d1 for increases in geosmin concentrations from 50 to 200 ng/l (Ho et al., 2007a), and rate constant data was similar to that of 0.25 d1 reported here for 100 ng/l geosmin. Greater changes in rate constant data may have been observed by Ho et al. (2007a) if a broader range of geosmin concentrations were tested. However, these rate constant data were far greater than those reported for the removal of geosmin through sand filter columns, where a rate constant of 0.075 d1 (R2 ¼ 0.90) was reported previously for the biofiltration of geosmin (Ho et al., 2007a). The genetic mechanism for geosmin degradation by bacteria is currently unknown; however, it is possible that this

2931

water research 43 (2009) 2927–2935

occurs via an operon of genes in a similar way to that proposed for degradation mechanism of 2-methylisoborneol (Oikawa et al., 1995). To date, the best evidence for the pathway of geosmin degradation by bacteria has been provided by Saito et al. (1999), who identified four possible biodegradation products of geosmin. Two of the products were identified as 1,4a-dimethyl-2,3,4,4a,5,6,7,8-octahydronaphthalene and enone, both of which have been used previously in the chemical synthesis of ()-geosmin (Marshall and Hochstetler, 1968; Saito et al., 1996). It is possible that geosmin may be biodegraded by a pathway similar to that of cyclohexanol (Ho et al., 2007a), where strains of Acinetobacter and Nocardia have been shown to degrade cyclohexanol using monooxygenase enzymes, similar to the biological Baeyer– Villiger reaction (Trudgill, 1984). However, the application of degenerate PCR within our laboratory targeting the gene chnB, which encodes for the monooxygenase enzyme in cyclohexanol degradation (Cheng et al., 2000), has failed to detect the presence of homologues in Geo24 or Geo48 (unpublished data).

3.3.

Effect of initial cell numbers upon degradation

Within the first 6 h the rate of geosmin degradation appeared to coincide with the number of initial Geo48 cells (Fig. 4), and during this time there was no measurable change in bacterial abundance as measured by FCM (data not shown). However, following the first 6 h, there was little difference between the rates of geosmin degradation (Fig. 4), where the mechanism was again demonstrated to be of pseudo-first-order (Fig. 4 inset). The similarity in rate constant data between the different suspensions was most likely due to a 1.66 and 0.85 log10 fold increase in Geo48 cell numbers for the 1  104 and 1  105 cells/ml suspensions respectively over the course of the experiment, compared with only a 0.2 log10 fold increase in the number of Geo48 cells within the 1  106 cells/

ml suspension (measured by FCM, data not shown). As a result, from approximately 24 h onwards each suspension contained numbers of Geo48 cells within the same order of magnitude. The respective increases in bacterial abundance within the suspensions were most likely due to growth of Geo48 utilising the natural DOC as a primary substrate, as similar observations have been reported previously for the inoculation of individual bacterial species within sterilized reservoir water (Ho et al., 2007b). The results here are in contrast to Ho et al. (2007a), who reported a rate constant increase for geosmin degradation from 0.12 to 0.24 d1 when suspensions of mixed biofilmassociated bacteria in WTP treated water were increased from 1  103 to 1  105 cells/ml. However, the prolific increase in cell numbers during geosmin degradation was not monitored in that study.

3.4.

Effect of water temperature upon degradation

Fig. 5 shows the data regarding the effect of three different water temperatures upon the rate of geosmin degradation by Geo48. The data revealed that the most rapid rate of geosmin biodegradation by Geo48 occurred at a water temperature of 22  C, followed by 30  C and then 11  C, with a rate constants of 0.023 h1 (equivalent to 0.54 d1, R2 ¼ 0.91), 0.019 h1 (equivalent to 0.45 d1, R2 ¼ 0.85) and 0.017 h1 (equivalent to 0.41 d1, R2 ¼ 0.98) respectively. No loss of geosmin was observed within the control culture (data not shown). The biodegradation of geosmin at each temperature again occurred via a pseudo-first-order mechanism (Fig. 5 inset). These data suggest that Geo48 has the capability to degrade geosmin over the range of water temperatures that would be encountered from tropical to cool temperate climates. Such a finding is important as it suggests that water temperature may not limit the potential for the seeding of Geo48 into WTP

Effect of temperature -1 2 11°C (k=0.0169h , R =0.98) -1 2 22°C (k=0.0227h , R =0.91) -1 2 30°C (k=0.0187h , R =0.85)

0.5

500 400 300

-1

2

1E4 cells/mL (k=0.0256h , R =0.90)

-1.0 -1.5 -2.0 -2.5 -3.0 -3.5 -4.0 -4.5 -20

200

600

° -1 2 1E6 cells/mL (k=0.0247h , R =0.94) 2 1E5 cells/mL (k=0.0243h-1, R =0.97)

0.0 -0.5

0

20

40

60

80 100 120 140 160 180

Time (h)

100 0

Geosmin concentration (ng/l)

ln([Geosmin]/[Geosmin] ) °

Geosmin concentration (ng/l)

600

Effect of [Ge48]

ln([Geosmin]/[Geosmin] ) °

0.5

500 400 300

0.0 -0.5 -1.0 -1.5 -2.0 -2.5 -3.0 -3.5 -4.0 -4.5 -20

0

20

40

200

60

80 100 120 140 160 180 200

Time (h)

100 0

0

20

40

60

80

100

120

140

160

180

Time (h) Fig. 4 – The effect of initial target Geo48 cell numbers of 1 3 106/ml (-), 1 3 105/ml (C), and 1 3 104/ml (:) upon the degradation of the taste and odour compound in sterilized reservoir water. Inset: Pseudo-first-order kinetic plots using data from the removal curve. Error bars represent 95% confidence intervals.

0

50

100

150

200

250

Time (h) Fig. 5 – The effect of various incubation temperatures of 11 8C (-), 22 8C (C), and 30 8C (:) upon the degradation of geosmin by Geo48 in sterilized reservoir water. Inset: Pseudo-first-order kinetic plots using data from the removal curve. Error bars represent 95% confidence intervals.

2932

water research 43 (2009) 2927–2935

sand filters as a further approach for the enhanced removal of geosmin using biofiltration processes (McDowall et al., 2009). It should be considered that the geosmin degradation data presented in this study could not account for all environmental factors that may influence geosmin degradation by planktonic Geo48 in situ. By investigating the degradation of geosmin by Geo48 in sterile reservoir water, the organism was not influenced by any interactions with the indigenous 90 40

microbial community normally active within such waters. This may have included any extracellular signalling by other active microbial species, where such cell to cell communication can often have effects upon the regulation of genes such as those encoding degradative enzymes (Dunny et al., 2008). The absence of active indigenous microorganisms would have also eliminated any nutrient competition (Gottschal, 1993) that may normally influence the degradation of geosmin by

Novosphingobium subterraneum (AB025014) Novosphingobium aromaticivorans (AB025012)

43

Novosphingobium yangbajingensis (EU118985)

31

Novoshpingobium capsulatum (D16147) Novosphingobium taihuense (AY500142)

32

Novosphingobium stygiae (AB025013)

69

Geo25 (DQ137853)

96

26

Novosphingobium hassiacum (AJ416411)

30

Novoshingobium lentum (AJ303009)

90

Novosphingobium tardaugens (AB070237) Novosphingobium nitrogenifigens (DQ448852) Sphingopyxis panaciterrae (AB245354)

62

Sphingopyxis macrogoltabida (D84530)

37

Sphingomonas taejonensis (AF131297) 34

Sphingomonas chilensis (AF367204)

55

Sphingopyxis witflariensis (AJ416410) 41

Sphingopyxis composta (AY563034)

21 36

52

Sphingopyxis ginsengisoli (AB245343) Sphingopyxis alaskensis (AF378796) Kartchner Caverns bacterium HI-I1 (DQ205298) Phenanthrene-degrading bacterium M20 (AY177357)

97

99

Geo24 (DQ137852) 99 73 Geo48 Sphingomonas adhaesiva (D13722)

84

Sphingopyxis terrae (D13727)

81

Sphingomonas bosoensis (AB275605)

85

Novosphingobium subarcticum (X94104) Sphingopyxis flavimaris (AY554010)

86 81

81

Sphingomonas baekryungensis (AY608604) Novosphingobium indicum (EF549586)

80

Novosphingobium panipatensis (EF424402) 43

36

Novosphingobium pentaromativorans (AF502400)

93 98 23

Novosphingobium mathurensis (EF424403) Sphingomonas xenophaga (EU240401)

Sphingomonas suberifaciens (D13737) Sphingosinicella soli (DQ087403) 55

99 56

Sphingosinicella microcystinivorans (AB219940) Other Sphingomonadaceae (37 species) Caulobacter fusiformis (AJ007803)

0.02

Fig. 6 – Neighbour-joining phylogenetic tree showing the position of isolate Geo48 (bold) in relation to other closely related species, including isolates previously implicated in the degradation of geosmin (underlined). Accession numbers correspond to partial 16S rRNA gene sequences. Numerical tree values represent bootstrap support. Scale bar represents expected changes per site.

water research 43 (2009) 2927–2935

Geo48 in situ. Additional environmental factors that could not be accounted for may have included the transient changes in DOC (Judd et al., 2006) that often occur following heavy inflows into water storages, any sub-lethal effects of solar UV irradiation (Ha¨der and Sinha, 2005) and the predation of Geo48 by higher eukaryotic organisms (Ju¨rgens and Matz, 2002). Consideration should also be given to the potential differences in degradation of geosmin by Geo48 in a planktonic state compared with that of Geo48 when integrated as a member of a sand filter biofilm. It is well documented that integration of bacteria into a biofilm matrix is a complex and multifactorial process, often resulting in phenotypic alterations of biofilm-associated organisms with respect to gene transcription (Singh et al., 2006). In particular, Whiteley et al. (2001) demonstrated that certain bacterial subpopulations exhibit different patterns of gene expression when in a biofilm compared with a planktonic state and such differences may also be true for the genes associated with geosmin degradation in Geo48. Furthermore, biofilm-mediated bioremediation is often a more proficient alternative to planktonic bioremediation as it is believed that cells within a biofilm can more easily adapt to changing conditions and are more protected within the matrix (Decho, 2000). It is therefore possible that the biodegradation efficiency of geosmin by planktonic Geo48 cells reported in this study may be somewhat different to sand filter biofilm-associated Geo48 cells in situ. Further studies involving seeding of Geo48 onto model filters will yield useful information on this.

3.5.

Phylogenetic analysis of Geo48

Phylogenetic analysis revealed that isolate Geo48 belonged to Sphingomonadaceae of the sub-class Alphaproteobacteria. As seen in Fig. 6, Geo48 clustered strongly (99% bootstrap support) with Sphingopyxis sp. Geo24 (accession number DQ137852) and the two organisms were 99.7% similar over a 1412 bp fragment of their 16S rRNA genes. Sphingopyxis sp. Geo24 was previously isolated in our laboratory from Morgan WTP (South Australia) sand filter material but was shown to degrade geosmin only as a consortium member together with two other organisms, Novosphingobium sp. Geo25 and Pseudomonas sp. Geo33 (Hoefel et al., 2006). Sphingopyxis sp. Geo24 was tested in this present study in parallel with Geo48 but no measurable degradation of geosmin was achieved (data not shown). This is despite the two organisms being closely related, based on 16S rRNA gene sequence data. To date it is unknown if the apparent inability of Geo24 to degrade geosmin individually is due to the genotypic absence or mutation of one or more of the genes critical for the degradation of geosmin. It is possible that some (or all) of the genes involved in geosmin degradation reside on a catabolic plasmid, which Geo24 may not have retained. However, plasmid screening of Geo48 and Geo24 in our laboratory to date has indicated the absence of plasmid DNA in both organisms. Alternatively, Geo24 may be metabolically deficient (Hoefel et al., 2006), and in the presence of the required co-factor or signalling molecule, may display a geosmin-degrading phenotype. Nevertheless, this study is the first to report the isolation of a Gramnegative bacterium, Geo48, which has the demonstrated ability to degrade geosmin individually.

2933

Geo48 also clustered strongly (99% bootstrap support) with a phenanthrene-degrading bacterium M20 (accession number AY177357; 99.8% similarity to a 1358 bp sequence of the16S rRNA gene) and a Kartchner Caverns bacterium HI-I1 (accession number DQ205298; 99.6% similarity to a 1379 bp sequence of the 16S rRNA gene). Both M20 (Bodour et al., 2003) and HI-I1 (Ikner et al., 2007) were isolated from North America and therefore have significantly different geographical locations to that of Geo48 and Geo24. Whilst many metabolic traits between microorganisms are congruent with the evolution of so called housekeeping genes such as the 16S rRNA (Ma and Zeng, 2004), in the absence of further studies it can not be assumed that M20 and HI-I1 also have the capacity to degrade geosmin. Of the named species, Geo48 was most closely related to Sphingopyxis alaskensis (97.2% sequence similarity to a 1454 bp sequence of the 16S rRNA gene). The DNA sequence of the 16S rRNA gene fragment of Geo48 was deposited to GenBank under accession number EU816422.

4.

Conclusions

This study is the first to report the isolation and characterisation of a Gram-negative bacterium capable of degrading geosmin individually. The bacterium was isolated from the biofilm of a WTP sand filter and demonstrated to degrade geosmin in a planktonic state by a pseudo-first-order mechanism of degradation. Such data provides further insights into the largely unknown microbiological processes that occur during the biological removal of geosmin through water treatment. Further studies are planned to investigate the efficiency of geosmin removals following the seeding of Geo48 into WTP filters, as we have recently demonstrated this to be a potential approach to enhancing the biological filtration of this taste and odour compound (McDowall et al., 2009). Attempts are also underway to elucidate the precise molecular mechanisms and genes responsible for the degradation of geosmin by Geo48. This will allow for the subsequent development of molecular tools to screen both natural environmental and engineered processes for the presence of geosmin-degrading organisms and also the development of biosensors for the early detection of geosmin episodes.

Acknowledgements This work was funded by a grant from the American Water Works Association Research Foundation and United Water International. We thank the Cooperative Research Centre for Water Quality and Treatment, the Australian Water Quality Centre and the South Australian Water Corporation for supporting this project.

references

Anderozzi, R., Cesaro, R., Marotta, R., Pirozzi, F., 2006. Evaluation of biodegradation kinetic constants for aromatic compounds

2934

water research 43 (2009) 2927–2935

by means of aerobic batch experiments. Chemosphere 62, 1431–1436. Ashitani, K., Hishida, Y., Fujiwara, K., 1988. Behavior of musty odourous compounds during the process of water treatment. Water Sci. Technol. 20, 261–267. Bodour, A.A., Wang, J.M., Brusseau, M.L., Maier, R.M., 2003. Temporal change in culturable phenanthrene degraders in response to long-term exposure to phenanthrene in a soil column system. Environ. Microbiol. 5, 888–895. Cheng, Q.S., Thomas, M., Kostichka, K., Valentine, J.R., Nagarajan, V., 2000. Genetic analysis of a gene cluster for cyclohexanol oxidation in Acinetobacter sp. strain SE19 by in vitro transposition. J. Bacteriol. 182, 4744–4751. Cook, D., Newcombe, G., Sztajnbok, P., 2001. The application of powdered activated carbon for MIB and geosmin removal: predicting PAC doses in four raw waters. Water Res 35, 1325–1333. Decho, A.W., 2000. Microbial biofilms in intertidal systems: an overview. Cont. Shelf Res. 20, 1257–1273. Dunny, G.M., Brickman, T.J., Dworkin, M., 2008. Multicellular behavior in bacteria: communication, cooperation, competition and cheating. Bioessays 30, 296–298. Elhadi, S.L.N., Huck, P.M., Slawson, R.M., 2004. Removal of geosmin and 2-methylisoborneol by biological filtration. Water Sci. Technol. 49, 273–280. Glaze, W.H., Schep, R., Chauncey, W., Ruth, E.C., Zarnoch, J.J., Aieta, E.M., Tate, C.H., McGuire, M.J., 1990. Evaluating oxidants for the removal of model taste and odor compounds from a municipal water supply. J. Am. Water Works Assoc. 82, 79–84. Gottschal, J.C., 1993. Growth kinetics and competition – some contemporary comments. Antonie van Leeuwenhoek 63, 299–313. Ha¨der, D.P., Sinha, R.P., 2005. Solar ultraviolet radiation-induced DNA damage in aquatic organisms: potential environmental impact. Mut. Res. 571, 221–233. Hayes, K.P., Burch, M.D., 1989. Odorous compounds associated with algal blooms in South Australian waters. Water Res 23, 115–121. Ho, L., Newcombe, G., Croue´, J.-P., 2002. Influence of the character of NOM on the ozonation of MIB and geosmin. Water Res 36, 511–518. Ho, L., Hoefel, D., Bock, F., Saint, C.P., Newcombe, G., 2007a. Biodegradation rates of 2-methylisoborneol (MIB) and geosmin through sand filters and in bioreactors. Chemosphere 66, 2210–2218. Ho, L., Hoefel, D., Saint, C.P., Newcombe, G., 2007b. Isolation and identification of a novel microcystin-degrading bacterium from a biological sand filter. Water Res 41, 4685–4695. Hoefel, D., Grooby, W.L., Monis, P.T., Andrews, S., Saint, C.P., 2003. Enumeration of water-borne bacteria using viability assays and flow cytometry: a comparison to culture-based techniques. J. Microbiol. Methods 55, 585–597. Hoefel, D., Monis, P.T., Grooby, W.L., Andrews, S., Saint, C.P., 2005. Profiling bacterial survival through a water treatment process and subsequent distribution system. J. Appl. Microbiol. 99, 175–186. Hoefel, D., Ho, L., Aunkofer, W., Monis, P.T., Keegan, A., Newcombe, G., Saint, C.P., 2006. Cooperative biodegradation of geosmin by a consortium comprising three gram-negative bacteria isolated from the biofilm of a sand filter column. J.Appl. Microbiol. 43, 417–423. Ikner, L.A., Toomey, R.S., Nolan, G., Neilson, J.W., Pryor, B.M., Maier, R.M., 2007. Culturable microbial diversity and the impact of tourism in Kartchner Caverns, Arizona. Microb. Ecol. 53, 30–42. Judd, K.E., Crump, B.C., Kling, G.W., 2006. Variation in dissolved organic matter controls bacterial production and community composition. Ecology 87, 2068–2079.

Ju¨rgens, K., Matz, C., 2002. Predation as a shaping force for the phenotypic and genotypic composition of planktonic bacteria. Antonie Van Leeuwenhoek 81, 413–434. Lundgren, B.V., Grimvall, A., Savenhed, R., 1988. Formation and removal of off-flavour compounds during ozonation and filtration through biologically active sand filters. Water Sci. Technol. 20, 245–253. Ma, H.-W., Zeng, A., 2004. Phylogenetic comparison of metabolic capacities of organisms at genome level. Mol. Phylogen. Evol. 31, 204–213. Marshall, J.A., Hochstetler, A.R., 1968. The synthesis of ()-geosmin and the other 1,10-dimethyl-9-decalol isomers. J. Org. Chem. 30, 3642–3646. McDowall, B., Ho, L., Saint, C.P., Newcombe, G., 2007. Biological removal of MIB and geosmin through rapid gravity filters. Water: J. Aust. Water Assoc. 34, 48–54. McDowall, B., Hoefel, D., Ho, L., Saint, C.P., Newcombe, G., 2009. Enhancing the biofiltration of geosmin by seeding sand filter columns with a consortium of geosmin-degrading bacteria. Water Res. 43, 433–440. Metz, D.H., Pohlman, R.C., Vogt, J., Summers, R.S., 2006. Removal of MIB and geosmin by full scale biological sand filters. In: Gimbel, R., Graham, N.J.D., Collins, M.R. (Eds.), Recent Progress in Slow Sand and Alternative Biofiltration Processes. IWA Publishing, London, UK, pp. 352–359. Namkung, E., Rittmann, B.E., 1987. Removal of taste- and odorcausing compounds by biofilms grown on humic substances. J. Am. Water Works Assoc. 66, 532–536. Narayan, L.V., Nunez, W.J., 1974. Biological control: isolation and bacterial oxidation of the taste and odor compound geosmin. J. Am. Water Works Assoc. 66, 532–536. Oikawa, E., Shimizu, A., Ishibashi, Y., 1995. 2-Methylisoborneol degradation by the CAM operon from Pseudomonas putida PpG1. Water Sci. Technol. 31, 79–86. Paerl, H.W., Huisman, J., 2008. Climate. Blooms like it hot. Science 320, 57–58. Rittmann, B.E., Gantzer, C.J., Montiel, A., 1995. Biological treatment to control taste-and-odor compounds in drinking water treatment. In: Suffet, I.H., Mallevialle, J., Kawczynski, E. (Eds.), Advances in Taste-and-odor Treatment and Control. American Water Works Association Research Foundation, Denver, USA 203–240. Saadoun, I., El-Migdadi, F., 1998. Degradation of geosmin-like compounds by selected species of Gram-positive bacteria. Lett. Appl. Microbiol. 29, 98–100. Saito, A., Tanaka, A., Oritani, T., 1996. A practical synthesis of enantiomerically pure ()-geosmin via highly diastereoselective reduction of (4aS,8S)-4,4a,5,6,7,8hexahydro-4a,8-dimethyl-2(3H)-naphthalenone. Tetrahedron: Asymmetry 7, 2923–2928. Saito, A., Tokuyama, T., Tanaka, A., Oritani, T., Fuchigami, K., 1999. Microbial degradation of ()-geosmin. Water Res 33, 3033–3036. Schmidt, S.K., Simkins, S., Alexander, M., 1985. Models for the kinetics of biodegradation of organic compounds not supporting growth. Appl. Environ. Microbiol. 50, 323–331. Silvey, J.K.G., Henley, A.W., Nunez, W.J., Cohen, R.C., 1970. Biological Control: Control of Naturally Occurring Taste and Odors by Microorganisms. Proceedings of the National Biological Congress, Detroit, USA. Singh, R., Paul, D., Jain, R.K., 2006. Biofilms: implications in bioremediation. Trends Microbiol. 14, 389–397. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., Higgins, D.G., 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25, 4876–4882.

water research 43 (2009) 2927–2935

Trudgill, P.W., 1984. Microbial degradation of the alicyclic ring: structural relationships and metabolic pathways. In: Gibson, D.T. (Ed.), Microbial Degradation of Organic Compounds. Marcel Dekker Inc., New York, pp. 131–180. Uhl, W., Persson, F., Heinicke, G., Hermansson, M., Hedberg, T., 2006. Removal of geosmin and MIB in biofilters – on the role of biodegradation and adsorption. In: Gimbel, R., Graham, N.J.D., Collins, M.R. (Eds.), Recent progress in slow sand and

2935

alternative biofiltration processes. IWA Publishing, London, UK, pp. 360–368. Whiteley, M., Bangera, M.G., Bumgarner, R.E., Parsek, M.R., Teitzel, G.M., Lory, S., Greenberg, E.P., 2001. Gene expression in Pseudomonas aeruginosa biofilms. Nature 413, 860–864. Young, W.F., Horth, H., Crane, R., Ogden, T., Arnott, M., 1996. Taste and odor threshold concentrations of potential potable water contaminants. Water Res 30, 331–340.