Development and application of a molecular assay to detect and monitor geosmin-producing cyanobacteria and actinomycetes in the Great Lakes

Journal of Great Lakes Research 40 (2014) 404–414

Contents lists available at ScienceDirect

Journal of Great Lakes Research journal homepage: www.elsevier.com/locate/jglr

Development and application of a molecular assay to detect and monitor geosmin-producing cyanobacteria and actinomycetes in the Great Lakes Olga A. Kutovaya 1, Sue B. Watson ⁎ Watershed Hydrology and Ecology Research Division (WHERD), Environment Canada, National Water Research Institute, 867 Lakeshore Rd, Burlington, ON L7R 4A6, Canada

a r t i c l e

i n f o

Article history: Received 4 December 2013 Accepted 14 March 2014 Available online 1 May 2014 Communicated by R. Michael McKay Index words: Taste–odor Geosmin Geosmin synthase Cyanobacteria Actinomycetes Great Lakes

a b s t r a c t Geosmin is a potent earthy-smelling sesquiterpene responsible for the majority of biologically induced taste-andodor (T/O) episodes in drinking and recreational water, and major economic losses to commercial, farmed and sports fisheries. Geosmin is produced by a range of microorganisms, notably by cyanobacteria and actinomycetes. However, the effective management of geosmin T/O episodes has been hindered by our inability to identify the major biological sources, and absence of sensitive methods for early detection of T/O events. The main goal of this study was to develop taxon-specific PCR and RT-PCR assays that facilitate early detection of potential geosmin releases and identification of the likely biological sources. We developed a molecular assay for the detection and expression of geosmin synthase (geoA), a gene involved in geosmin biosynthesis in cyanobacteria and actinomycetes. Using the molecular probes, we detected geoA in a wide range of geosmin-producing taxa in both laboratory cultures and environmental samples which provided an important new database of geoA sequences from a diversity of GSM producers, allowing an evaluation of their phylogenetic relationships based on this gene. This study indicated that potential geosmin producers in the Laurentian Great Lakes water and sediment samples are predominantly represented by members of the Nostocaceae, and in particular Anabaena spp. No correlations were apparent between gene expression and ambient temperature, nutrient concentrations and total phytoplankton biomass. Rather, evaluation of the geoA expression pattern in the environmental samples tested from various watersheds and seasons suggested constitutive gene expression. Crown Copyright © 2014 Published by Elsevier B.V. on behalf of International Association for Great Lakes Research. All rights reserved.

Introduction Geosmin (trans-1, 10-dimethyl-trans-9-decalol) and 2-methylisoborneol (2-MIB) are terpenoid secondary metabolites belonging to the sesquiterpenes and monoterpenes, respectively. These compounds have been studied extensively since their identification in the 1960s as two of the most common sources of biologically induced taste-andodor (T/O) (Gerber and Lechevalier, 1965) and the confirmation of cyanobacteria and actinomycetes as their major producers in sourceand drinking-water and aquaculture systems worldwide (Jüttner, 1995; Jüttner and Watson, 2007; Medsker et al., 1968; Persson, 1996; Zaitlin and Watson, 2006). Cyanobacteria are considered to be the dominant source of geosmin in aquatic environments, particularly in eutrophic systems (Jüttner et al., 1986; Jüttner and Watson, 2007; Matsumoto and Tsuchiya, 1988; Persson, 1988), while actinomycetes make a significant contribution to geosmin and 2-MIB related T/O

⁎ Corresponding author. Tel.: +1 905 336 4759. E-mail addresses: [email protected] (O.A. Kutovaya), [email protected] (S.B. Watson). 1 Tel.: +1 905 336 4759.

during runoff events or where there is a significant water–sediment interaction (Jensen et al., 1994; Zaitlin et al., 2003). Cyanobacteria and actinomycete species of both planktonic and benthic origin produce these compounds (Jüttner, 1984; Medsker et al., 1968; Schrader and Dennis, 2005) and while many studies have focused on planktonic producers, benthic taxa that cause sometimes significant T/O episodes are often overlooked (Izaguirre and Taylor, 2004; Jüttner and Watson, 2007). This study focused on geosmin-producing cyanobacteria and actinomycetes from both planktonic and benthic habitats. Geosmin is relatively stable to chemical and biological degradation and resists conventional water treatment. It is one of the most common causes of off-flavor in drinking water and farmed fish, and incurs substantial costs to treatment and lost revenue (Cook et al., 2001; Hanson, 2003; Howgate, 2004; Smith et al., 2002). Successful prediction and control of geosmin events require knowledge of the producers and the factors affecting geosmin biosynthesis. Environmental factors such as light, temperature, nutrients, chlorophyll concentration and bacterial interaction are suggested to correlate to geosmin production (Aoyama et al., 1995; Bowmer et al., 1992; Naes and Post, 1988; Rosen et al., 1992; Saadoun et al., 2001; Watson and Ridal, 2004). Despite this knowledge, taste and odor outbreaks are still unpredictable and continue to occur in freshwater systems worldwide, including the lower Great

http://dx.doi.org/10.1016/j.jglr.2014.03.016 0380-1330/Crown Copyright © 2014 Published by Elsevier B.V. on behalf of International Association for Great Lakes Research. All rights reserved.

O.A. Kutovaya, S.B. Watson / Journal of Great Lakes Research 40 (2014) 404–414

Lakes (Lakes Erie, Michigan and Ontario) (Taylor et al., 2006; Watson et al., 2008). There is a growing need for a sensitive, reliable and rapid diagnostic method for detecting, monitoring and managing geosmin events. Current methods used to identify geosmin producers are often inconclusive. Microscopy is widely used to monitor source waters for potential algal producers but is not applicable to actinomycetes and at best, yields tentative diagnostics, particularly with new case studies. It requires knowledge of likely producers based on the source-water (if recorded) or other studies, yet few species have been tested and confirmed as geosmin producers (e.g. Watson, 2003). Taxonomic expertise has traditionally relied on morphological traits which can be highly variable; current systematics are now increasingly integrating this with molecular and biochemical characteristics (e.g. Komárek, 2002; Komárek and Zapomělová, 2007, 2008). Per capita geosmin yield often differs among species and strains of the same species (Jüttner and Watson, 2007), and is further influenced by factors such as nutrients, light, temperature, and bacterial interactions (e.g. Aoyama et al., 1995; Bowmer et al., 1992; Naes and Post, 1988; Naes et al., 1989; Rosen et al., 1992; Saadoun et al., 2001). More recent efforts to use fluorescence in situ hybridization (FISH) and immunofluorescence techniques are limited by the non-specificity of these methods, which detect both geosmin producers and non-producers (Klausen et al., 2005; Nielsen et al., 2006). Finally, even though advanced analytical techniques such as gas chromatography–mass spectrometry (GC–MS) is useful for quantifying geosmin production, it cannot identify the biological source(s). In this study, we describe a molecular approach that can address many the limitations of the current techniques and provide an affordable and more effective method to detect and manage potential odor outbreaks. Geosmin is synthesized from the universal sesquiterpene precursor, farnesyl diphosphate (FPP), in a two-step process catalyzed by a bifunctional sesquiterpene cyclase (also known as terpene synthase or geosmin synthase geoA) in the presence of Mg2+ (Giglio et al., 2008; Jiang et al., 2007; Ludwig et al., 2007). In this study, the geosmin synthase gene (geoA) was employed as a proxy to detect potential geosmin producers (using PCR) and evaluate the expression pattern (using RTPCR). Molecular studies using the geosmin synthase have been done previously to better comprehend the nature of geosmin-producers in both cyanobacteria and actinomycetes (Agger et al., 2008; Cane et al., 2006; Giglio et al., 2011; Jiang and Cane, 2008; Ludwig et al., 2007). In recent work, Giglio et al. (2008, 2011) isolated and characterized this gene and demonstrated its universality for both cyanobacteria and actinomycetes. However, there is still little known about geosmin synthase distribution and expression in environmental samples. We developed a molecular assay for an early detection and monitoring of geosmin producers, and evaluated the geoA expression pattern in both planktonic and benthic cyanobacteria and actinomycetes in the Laurentian Great Lakes, notably Lakes Ontario and Erie, the most severely affected by taste–odor events (Watson et al., 2007, 2008). Using cyanobacterial and actinomycete strains isolated from the Great Lakes and other waterbodies we first developed molecular probes to evaluate their phylogenetic relationships based on their potential for geosmin production. We then applied these probes to water samples and sediments collected from across Lakes Erie and Ontario to evaluate the genetic sources and potential for geosmin production in these lakes. Materials and methods Cultures and cultivation The actinomycetes and cyanobacteria strains used in the study are listed in Table 1 and include a mix of strains from different waterbodies. The cyanobacterial strains from the Laurentian Great Lakes, Lake Winnipeg and Lake of the Woods (S. Watson; EC strains) were initially isolated into sterile filtered lake water by micropipetting followed by repeated washings, and later transferred to growth media once established

405

(Andersen, 2005). With the exception of the Maumee Lyngbya wollei mats, the cyanobacteria strains were monoalgal but not axenic. Stock cultures were maintained at 20 °C in Z8 media (Kotai, 1972) under a light regime of 16:8 h light:dark at 50 μmol photons m−2 s−1, and transferred on a monthly basis using sterile technique. For the purposes of this study, inoculates were grown under the same conditions in 250 mL batch culture to late exponential phase (2–3 weeks). Attempts to isolate L. wollei BH-LE were not successful; however, we were able to maintain viable, nonaxenic mats of this cyanobacterium under low light in algal medium for extended periods. The actinomycete strains were isolated from a Maumee River (Lake Erie, USA) Lyngbya mat onto BD Difco™ Actinomycete Isolation Agar (product # 212168, BD New Jersey USA) and cultured at 20 °C in TSB liquid media (T8907, Sigma Aldrich Canada) until stationary growth had been achieved. Environmental samples Information on water and sediment sampling station locations and dates can be found in Electronic Supplementary Material (ESM) Tables S1 and S2. At each site, water temperature was measured using a YSI profiler. Surface water was collected from a depth of 1 m using a Van Dorn (or equivalent) sampler and processed for the analysis of water chemistry, major nutrients (total phosphorus [TP], soluble reac− tive P [SRP], nitrate/nitrite [NO− 3 + NO2 ], ammonia [NH3]) and chlorophyll-a (chla) by the National Laboratory for Environmental Testing in Burlington, Ontario using standard methods (Environment Canada, 1994). Total P was preserved with 1% (v/v) H2SO4 in 120 mL glass bottles and analyzed following persulfate digestion. Dissolved nutrients were analyzed from 0.45-μm membrane filtrates (Sartorius) prepared on site; biomass for chla analysis was collected onto 45 mm GF/C filters (Whatman Corporation) and frozen until extraction. Subsamples were preserved in Lugol's iodine for later identification and enumeration of major algal and cyanobacterial taxa using the Utermöhl method (Lund et al., 1958). Water was filtered using a peristaltic pump onto 0.22-μm Sterivex cartridges (Millipore) and immediately frozen at − 80 °C for nucleic acid extraction as described below. Surficial (top ~ 5 cm) sediment samples were collected using a Ponar sampler into 50 mL screw-capped polyethylene tubes and immediately frozen at −80 °C until analyzed. Primers design and evaluation All primers used in this study are shown in Table 2. Degenerate primers for cyanobacterial geosmin synthase, geo_cya543F and geo_cya728R, were developed based on the alignment of thirteen sequences (Fig S1). Nine of the sequences were obtained through testing the primers from Giglio et al. (2008, 2011) against genomic DNA extracted from cyanobacterial isolates and sequencing the positive samples. The other four geosmin synthase sequences were acquired from GenBank (Anabaena ucrainica CHAB1432 — HQ404996.1; A. ucrainica CHAB2155 — HQ404997.1; Phormidium sp. P2r — EF619621.1; Nostoc punctiforme PCC 73102 NPUNMOD — FJ010203.1). Degenerate primers for actinomycete geosmin synthase, geo_act532F and geo_act698R were designed taking into consideration an alignment from Auffret et al. (2011). The primers evaluation and validation was performed in two steps. First, the specificities of the new primers were tested using BLASTn and Primer-BLAST. Forward and reverse primers for both cyanobacterial geoA and actinomycete geoA assays indicated relatively high E-values with the geosmin synthase sequences in the GenBank and were the top hits with the intended targets (data not shown). Second, the primer efficacy was evaluated with template DNA from the isolates that have been previously confirmed as geosmin producing organisms by GC–MS (Watson, unpublished data). The isolates were the following: Anabaena lemmermannii GI CA799, Aphanizomenon gracile SAG 31–79, Geitlerinema splendida NIVA 244, Phormidium GI LM788 and Actinomyces sp. EC 012-1, 2, 3 (three isolates) (Table 1).

406

O.A. Kutovaya, S.B. Watson / Journal of Great Lakes Research 40 (2014) 404–414

Table 1 Actinomycete and cyanobacterial strains used in this study and results of 16S rRNA gene sequences (PCR), geosmin synthase (PCR) and GC–MS analyses. Uppercase letters after species indicates the source of the cultures: SAG (Sammlung von Algenkulturen der Universität Göttingen), PCC (Pasteur Culture Collection of Cyanobacteria), NIES (National Institute for Environmental Studies), NIVA (Norwegian Institute for Water Research), GI (isolated by George Izaguirre, 2004), EC (isolated by S. Watson, Environment Canada strain). The underlined and bold-type strains are known geosmin producers and strains that were selected for gene expression analysis, respectively. Note that planktic Anabaena species have been reclassified under the genus Dolichospermum (Wacklin et al. 2009). Geosmin production as indicated by 16S, geoA_cya and GC–MS (geosmin) is given on a qualitative scale from not detected (ND), trace "(+)", moderate “+”, strong, “++”, high “+++”. N/A: sample not analyzed. The geoA_cya primers correspond to nucleotide sequences of the geosmin synthase in Nostoc. Strain

Origin

16S

geoA_cya

GC–MS (geosmin)

Actinomyces sp. EC 012-1 WS Maumee River Actinomyces sp. EC 012-2 WS Maumee River Actinomyces sp. EC 012-3 WS Maumee River Anabaena flos-aquaea CPCC 64 Anabaena lemmermannii GI CA799 A. lemmermannii EC LE011-02 A. lemmermannii LE 011-03 A. lemmermannii LE 011-04 A. lemmermannii LO 006-02 A. lemmermanniia LO 006-07

Maumee River Ohio (Lake Erie) 2012 Maumee River Ohio (Lake Erie) 2012 Maumee River Ohio (Lake Erie) 2012 Lake Ontario 1984 Castaic Lake, CA, USA 1991 Lake Erie, St.938 2011 Lake Erie, East Basin (St.938) 2011 Lake Erie, East Basin (St.938) 2011 Lake Ontario, West basin 2006 Lake Ontario, West basin 2006

+ + + + + + + + + +

+ + + + + weak + N/D + +

A. lemmermannii LO 008-01 A. lemmermannii LW 011-02 Anabaena planktonica Anabaena ucrainica NIES 825 A. ucrainica NIES 826 Anabaena variabilis NIVA 19 Anabaena sp. SAG 28.79 Aphanizomenon flos-aquae EC HH06-01 Aphanizomenon gracile SAG 31-79 Calothrix sp. PCC7507 Geitlerinema splendida NIVA 244 Lyngbya wollei LSL121b Lyngbya wollei BH-LEb Oscillatoria limosa GI LBD 305b Phormidium GI LM788 Planktothrix EC LE 011-05 Planktothrix rubescens NIES 1266 Planktothrix suspensa EC SP011-01 Pseudanabaena EC LE 011-01 Pseudanabaena limnetica NIVA 111

Lake Ontario, West basin 2008 Lake Winnipeg, North Basin (St. W3) 2011 Unknown; from F Jüttner, University of Zurich Lake Sagami Kanagawa, Japan 1991 Lake Sagami Kanagawa Japan 1991 Lake Mendota, Madison, WIS, USA 1948-09 Soil, Pakistan; before 1978 Hamilton Harbour, St.1001 2006 Schleswig–Holstein, Plußsee 1982 Sphagnum bog, Vierwaldstättersee, Switzerland Lake Steinsfjorden, Buskerud, Norway 1963 Lac St Louis, St Lawrence River 2007 Maumee River Ohio USA 2007 Lake Bard, CA, USA Lake Mathews, CA, USA 1998 Sandusky Bay, Lake Erie 2011 Neuglobsow, Germany 2000 Smart Park Pond, MB, Canada 2011 Lake Erie 2011 Lake Biwa, Japan 1952

+ + + + + + + + + + + + + + + + + + + +

N/D + weak + + + + N/D + + + + + N/D + + + + N/D N/D

Pseudanabaena sp. GI ECR 601 Synechococcus rhodobactron NIVA 8 Synechococcus sp. GI CL 792

Eagle Creek Res., Indiana USA 2001 Lake Steinsfjorden, Norway 1963 Canyon Lake, CA, USA

+ + +

+ N/D weak

+ + + (+) +++ ND N/A ND + ND (previously weak producer) ND N/A ND + + ND + ND ++ + + + + ND + + N/A N/A ND ND (MIB producer) N/A ND ND (MIB producer)

a b

Anabaena flos-aquae CPCC 64 and Anabaena lemmermannii LO 006-02 were re-identified as Trichormus cf variabilis CPCC 64 and Trichormus cf. doliolum LO 006-02, respectively. Lyngbya wollei are not pure strains but represent mat samples collected and washed in the field and held in liquid media (may include epiphytes).

Analysis by GC–MS Isolates were examined for geosmin production using a headspace solid-phase microextraction (HSPME) GC–MS protocol (Watson et al., 2000). A volume of 25 mL of culture material was transferred to 40 mL pre-cleaned septa capped vials with 6 g pre-baked NaCl and 100 ng/L

biphenyl d-10 as an internal standard. Samples were heated to 60 °C, stirred with a pre-cleaned Teflon stir bar and extracted with a 65-μm polydimethylsiloxane–divinyl benzene fiber (Supelco) for 30 min. The fiber was desorbed for 2 min at an inlet temperature of 250 °C (splitless mode) on a Shimadzu QP2010 GC–MS, operated in full scan mode and using an HP-5MS column. Retention times and mass spectra were

Table 2 Primers used in PCR and RT-PCR reactions. Gene

Primer code

Primer sequence

Annealing temperature (˚C)

geoA_cyaa Geosmin synthase in cyanobacteria geoA_act Geosmin synthase in actinomycetes Cyano-bacterial 16S rRNA

geo_cya543F geo_cya728R geo_act532F geo_act698R cya359F cya781R(a) cya781R(b) 16S rRNA_pA 16S RNA_pH 250 F 971R 288AF 288AR

5′ ATCGAATACATYGARATGCG 3′ 5′ ACTTCTCTYTGRTAGGA 3′ 5′ GARTACRTCGAGATGCGVCG 3′ 5′ GAGAAVAKGTCGTTRCGCAGRTG 3′ 5′ GGGGAATYTTCCGCAATGGGY 3′ 5′ GACTACTGGGGTATCTAATCCCATT 3′ 5′ GACTACAGGGGTATCTAATCCCTT T 3′ 5′ AGAGTTTCATCCTGGCTCAG 3′ 5′ AAGGAGGTGATCCAGCCGCA 3′ 5′-TTCTTCGACGAYCACTTCC-3′ 5′-CCCTYGTTCATGTARCGGC-3′ 5′-AACGACCTGTTCTCCTA-3′ 5′-GCTCGATCTCATGTGCC-3′

56 (−0.5)

186

This study

61

167

58

423

This study based on Auffret et al. (2011) Nübel et al. (1997)

56

1535

55

743

Giglio et al. (2008)

55

288

Giglio et al. (2008)

Actino-mycetes 16S rRNA Geosmin synthase Geosmin synthase

geoA_cya primers correspond to nucleotide sequences of the geosmin synthase in Nostoc punctiforme PCC73102.

Product size (bp)

Reference

Edwards et al. (1989)

O.A. Kutovaya, S.B. Watson / Journal of Great Lakes Research 40 (2014) 404–414

407

compared to those of commercially purchased geosmin, 2-MIB and biphenyl d-10 standards (Sigma Aldrich, Oakville, ON).

The remaining sequences that are shorter than 200 bases are available upon request.

Nucleic acid extraction

Results and discussion

DNA from planktonic environmental samples and cultures was isolated using phenol-chloroform extraction (Chomczynski and Sacchi, 1987) or the DNeasy Blood and Tissue Kit (QIAGEN) respectively, as described in Ilikchyan et al. (2009). DNA from surficial sediment samples was extracted with the DNA Stool Kit (QIAGEN) using the protocol for ‘Isolation of DNA from Larger Volumes’. The approximate volume of a sediment sample used was ~1 g. RNA was extracted from environmental samples following collection of seston onto 0.22-μm Sterivex cartridge filters. RNA extraction from both environmental samples and cultures was performed using the RNeasy Mini Kit with slight modifications as described elsewhere (Ilikchyan et al., 2009). The RNA was then treated with RNase-free DNase (QIAGEN) to remove any trace amounts of genomic DNA. RNA and DNA concentrations were measured with a NanoDrop ND-1000 spectrophotometer. All RNA samples were screened for genomic contamination by PCR.

Isolation of potential geosmin producers

PCR and RT-PCR conditions Table 2 describes the primers used for PCR and RT-PCR amplifications. PCR reactions were run on a T100 Gradient Thermal Cycler (Bio-Rad, Inc.). PCR amplification was performed with 1 μL of normalized DNA template (ca. 10 ng) in a final reaction mixture (25 μL) containing 1× PCR buffer (Promega), 0.2 mM L−1 of each deoxynucleotide (Promega), 0.5 μM L−1 of each primer, and 1.0 unit of GoTaq DNA polymerase (Promega). The PCR conditions for cyanobacterial geoA (geo_cya543F-728R) were 95 °C for 5 min, 30 cycles of 94 °C for 1 min, annealing at 56 °C with 0.5 °C decrease per cycle for 1 min, 72 °C for 1 min, followed by extension at 72 °C for 12 min. For actinomycete geoA and cyanobacterial 16S rRNA primers the conditions were the same with the annealing temperatures shown in Table 2. With actinomycetes 16S rRNA primers, in addition to the above-mentioned PCR master mix, 0.5 μg μL−1 bovine serum albumin (New England Biolabs) and 5% (v/v, final concentration) dimethyl sulfoxide (Sigma-Aldrich) were added. RNA was reverse-transcribed to single-stranded complementary DNA using an iScript cDNA Synthesis kit (Bio-Rad). To normalize the cDNA, the volume of RNA template added to the reaction mix was adjusted to yield a final concentration of 50 ng μL−1. Additional reactions were done without reverse transcriptase (RT) to ensure the absence of genomic DNA in all cDNA samples. PCR amplification was done in the 25 μL reaction mixture with the cDNA template normalized to ca. 10 ng. PCR and RT-PCR products were resolved on a 2% agarose gel. Selected gene products were cloned into TOPO TA Cloning® Kit (Invitrogen) and verified by colony-PCR. The PCR products were then purified with the Qiaquick spin PCR purification Kit (Qiagen) and sequenced at the Institute for Molecular Biology and Biotechnology, MOBIX, McMaster University using specific gene primers (forward primers of geo_cya, geo_act and 16S). Obtained sequences were analyzed manually and aligned and edited with Mega 5.05 software (Kumar et al., 2007). Phylogenetic analysis was performed using Mega 5.05 and Phylogeny.fr (Dereeper et al., 2008). Neighbor-joining phylogenetic analysis of geoA_cya amplicon sequences was performed using the default settings of Mega 5.05 software, with bootstrap values (1000 iterations) N 50%. While using Phylogeny.fr, we employed MUSCLE 3.7 alignment and Gblocks 0.91B alignment refinement. PhyML 3.0-Approximate Likelihood-Ratio Test (PhyML-268 aLRT) was applied to compute the phylogenetic trees. Nucleotide sequence accession numbers Sequences that met the minimum size requirement of 200 bases were deposited in GenBank (accession numbers KF735766–KF735788).

In order to test the developed probes and evaluate the phylogenetic diversity among the potential geosmin producers, thirty-two cyanobacterial cultures and three actinomycete strains were isolated from various environments (Table 1) and maintained in the culture collection held at the CCIW in Burlington, ON. The taxonomic identification of the cyanobacterial strains was verified by S. Watson and H. Kling (Algal Taxonomy Ecology Inc. (ATEI), Winnipeg, Manitoba) based on morphological traits using traditional microscopy, and was confirmed in this study using 16S rRNA sequences (Fig. S2). PCR amplification of geosmin synthase The presence of the geoA gene in both cyanobacteria and actinomycetes was detected by a PCR-based assay. Primers available from Giglio et al. (2008) and Auffret et al. (2011) (Table 2) were first tested with our cultures and environmental samples. The primers from Giglio and colleagues failed to amplify the geoA gene in environmental samples but did generate a PCR product from a few isolates facilitating the primer development. As a result, low-degeneracy oligonucleotide cyanobacterial geosmin synthase primers were designed incorporating the variability in the available geoA sequences (Table 2, Fig. S1). Primers for actinomycete geosmin synthase from Auffret et al. (2011) did not generate any PCR product with either the culture or environmental DNA samples. The original specific geoA_act primers from Auffret and colleagues were designed for q-PCR assay and likely were not suitable for environmental sample analysis performed in this study. However, the alignment from Auffret et al. was used to make degenerate primers for actinomycete (Table 2). Both cyanobacterial- and actinomycete geoA assays, were further used in this study to evaluate the physiology of the geosmin synthase gene in cultures and environmental samples. First, the assays were successfully tested on cultures that have been previously confirmed by GC–MS as positive (Table 1). Next, the assays were examined on the isolated cultures and environmental samples from Lake Erie, pelagic Lake Ontario and nearshore Lake Ontario (Bay of Quinte). As a control, all DNA and RNA samples were first tested for amplification of 16S rRNA to confirm successful nucleic acid isolation. Primer dimer formation by the cyanobacterial geoA assay prohibited further quantitative assays. Survey of geosmin synthase and GC–MS in laboratory cultures Twenty-six of the thirty-five cultures tested positive for the geosmin synthase (Table 1). Sequencing of the geosmin synthase from cyanobacteria and actinomycetes verified the results (Fig. S3 and Fig. S4). Aphanizomenon cf. flos-aquae EC HH06-01, Oscillatoria limosa GI LBD 305b, Pseudanabaena sp. LE 011-01, Pseudanabaena limnetica NIVA 111, and Synechococcus rhodobactron NIVA 8 did not possess the geosmin synthase gene. These species have been characterized previously as either variable in their ability to produce T/O compounds, such as Aph. cf. flos-aquae and O. limosa (Bafford et al., 1993; Yagi et al., 1988), or exclusive 2-MIB producers, such as Pseudanabaena (Izaguirre and Taylor, 1998; Izaguirre et al., 1999; Yagi et al., 1988; Watson, unpublished data). S. rhodobactron NIVA 8 is a unicellular freshwater picocyanobacterium that has never been reported as geosmin producer. Two of the fourteen strains of Anabaena sp. tested were also negative for the geosmin synthase gene, which is consistent with the previous findings about the strain (species) specificity among cyanobacteria to produce geosmin (Jüttner and Watson, 2007; Taylor

408

O.A. Kutovaya, S.B. Watson / Journal of Great Lakes Research 40 (2014) 404–414

et al., 2006). Isolated actinomycete cultures were confirmed as geosmin synthase positive. Based on the phylogenetic analysis of the actinomycete geoA gene (Fig. S4) the isolated strains were closely related to the geosmin synthase in the actinomycetes Streptomyces flavogriseus AMU14 (AEZ56842) and Streptomyces anulatus AMU11 (AEZ56841). GC–MS analysis was largely consistent with the PCR results (Table 1). Seventeen out of twenty strains with the geosmin synthase gene had detectable geosmin production in liquid media. GC–MS results corroborated the negative results of the PCR assay. No geosmin was detected in any of the seven strains found to be lacking the geoA gene in the PCR assay. The majority of the PCR-positive samples had detectable geosmin concentrations. Four cultures, A. lemmermannii EC LE011-02, A. lemmermannii LO 006-07, Anabaena planktonica and Synechococcus sp. GI CL 792 were PCR-positive but produced no measurable geosmin by GC–MS. Two of these, A. lemmermannii EC LE011-02 and Synechococcus sp. GI CL 792 were previously reported as geosmin producers but have apparently lost the capacity to produce after being held in culture (Watson, unpublished; G. Izaguirre personal communication; see also Table 1). Notwithstanding the evidence for constant expression of geoA in environmental samples (discussed below), it is possible that under some environmental or culture conditions, cyanobacteria possessing geoA either do not express the gene (e.g., due to gene mutation), produce the geosmin synthase protein, or direct photosynthate into terpenoid metabolism. Elucidating these anomalies and the biochemical regulation of geosmin production remains an important goal for future studies.

a

Detection of geosmin synthase in environmental samples Overall, the cyanobacterial primers that we developed were effective and versatile, amplifying the geoA gene from a wide range of geosmin-producers in both laboratory cultures and environmental samples. Sequencing of the geoA_cya amplicons from water samples and sediment DNA confirmed both their identity as geosmin synthase gene fragments and phylogenetic assignment as cyanobacteria (Fig. 1A, B). The cyanobacterial geoA assay did not detect the geosmin synthase gene in available Maumee River samples (data not shown), which was likely due to the dominance of other non-geosmin producing taxa of cyanobacteria such as Microcystis spp. and Planktothrix spp. (Kutovaya et al., 2011). Additionally, there is evidence for benthic geosmin sources in the Maumee River, such as mat-forming L. wollei (Watson et al., unpublished data), but unfortunately we were unable to follow this up as benthos samples from this site were not collected in this study. A total of fourteen Lake Erie and one Lake Ontario sediment samples were tested for geosmin synthase (Table S2). Cyanobacterial geoA was successfully amplified from the two Lake Erie eastern basin sites (LE/ 938 — September, LE/879 — October), two western basin sites (LE/341 and 881 — September), one sample from the Sandusky River (LE/1198 — September) and one from Sandusky Bay (LE/1163 — September). The relatively modest recovery of the geoA-positive sediment samples (six out of fourteen) is likely due to a low abundance of the geoA-positive cells in the one gram of sediment sample used for the analysis. Only the samples where 16S rRNA was successfully amplified were tested

LE Sed.1163 5 Lyngbya wollei LSL121 Calothrix sp.PCC7507 LE 1198 1 LE Sed.341 4 LE 1198 22 LE Sed.341 6 LE Sed.938 8 Anabaena lemmermannii GI CA799 LE Sed.1163 7 65 LE Sed.1163 15 83 LE Sed.341 5 LE Sed.938 7 Aphanizomenon gracile SAG31-79 LE 1198 5 LE 880 22 Anabaena ucrainica CHAB1432 Anabaena ucrainica CHAB2155 LE Sed.1163 1 96 74 LE 880 7 LE 880 5 LE Sed.1163 2 LE Sed.341 1 92 LE Sed.938 1 LE Sed.1163 9 LE Sed.341 12 LE 880 2 80 LE Sed.938 4 LE Sed.881 1

72

80

72

82

(21)

(2)

(3) (4) (7) (5)

(3) (17) (15) (10) (2) (15) (6)

0.2 Fig. 1. Neighbor-joining phylogenetic tree of geoA_cya amplicon sequences from Lake Erie water and sediment samples (A), and Lake Ontario and Bay of Quinte water samples (B). Bootstrap values (1000 iterations) N50% are presented. The number of the identical clones is indicated in the brackets.

O.A. Kutovaya, S.B. Watson / Journal of Great Lakes Research 40 (2014) 404–414

b 76 86

Aphanizomenon gracile SAG31-79 BQ MR 10 BQ MR 7 78 Anabaena ucrainica CHAB1432 Anabaena ucrainica CHAB2155 85 BQ MR 23 LO RLC 27 90 LO RLC 16 LO RLC 25 LO RLC 1 LO LV 1 77 BQ MR 9 BQ N 18

LO LV 24 LO RLC 5 BQ HB 4 Anabaena lemmermannii GI CA799 BQ HB 2 BQ HB 5 83 BQ MR 4 BQ N 2 94 BQ HB 14 BQ MR 3 BQ N 4 BQ HB 1 BQ N 1 Calothrix sp.PCC7507

62

74

409

(20) (22) (3)

(2) (6)

(2) (3) (5) (7) (7) (2) (2)

0.09 Fig. 1 (continued).

with the cyanobacterial and actinomycete geoA assays. The actinomycete geosmin synthase was not detected in any water or sediment samples tested in this study, although actinomycetes have been shown to make a significant contribution to geosmin and MIB production in sediments (Jensen et al., 1994; Sugiura and Nakano, 2000). Attempts to amplify the actinomycete geosmin synthase gene were unsuccessful, possibly because gene abundance was not sufficient to yield a PCR product, or primers were not specific enough to amplify the gene from the analyzed environmental samples. The actinomycete assay improvement was not pursued in this study and requires further work. Phylogenetic analysis of cyanobacterial geosmin synthase from Lake Erie, Lake Ontario and Bay of Quinte We used a phylogenetic analysis of geoA_cya amplified from the environment to understand the evolutionary relationships among dominant geosmin producers in the Great Lakes. Previous studies often rely on the coincidence of the metabolite geosmin with putative sources, but this phylogenetic analysis directly identifies the organisms likely responsible for its production. Lake Erie water and sediment samples revealed three major clusters (Fig. 1A). The largest cluster, representing most of the clones, was an A. ucrainica cluster and consisted of 25 water and 51 sediment clones. The second cluster, A. lemmermannii, was formed by the sediment samples exclusively and included 18 clones. These results are consistent with the previous observations about the dominance of Anabaena in the Great Lakes (Hartman, 1973; Makarewicz, 1993; Millie et al., 2009). The last cluster was identified with Calothrix sp. PCC7507 and was composed of the majority (22 out of 27) of Sandusky River clones (LE/1198) and one sediment clone. There was also one clone from Sandusky Bay (LE/1163) identified with L. wollei. Overall the phylogenetic analysis demonstrates predominance of the Nostocaceae among both water and sediment geosmin producers in Lake Erie. The results also reveal the distinct cyanobacterial community in Sandusky River (LE/1198) closely related to Calothrix sp. PCC7507.

The geosmin synthase gene sequences from the three sites in western Lake Ontario showed that 45 clones grouped with A. ucrainica, regardless of their location, and only three clones formed a cluster with A. lemmermannii (Fig. 1B). In contrast, samples from the Bay of Quinte (which flows into Eastern Lake Ontario) were mostly represented by A. lemmermannii (33 clones), only 5 clones grouped with A. ucrainica, four clones somewhat related to Calothrix sp. PCC7507 and two clones with Aph. gracile SAG 31–79 (Fig. 1B). Consistent with these results, long term studies of Bay of Quinte phytoplankton communities have reported a diversity of cyanobacteria including species belonging to the Nostocaceae and Rivulariaceae, while seasonal and spatial patterns in T&O indicate multiple planktic and benthic sources (Nicholls and Carney, 2011; Watson et al., 2005). It is interesting to note that whereas A. lemmermanni has been identified as a common geosmin producer in the Great Lakes, there is no report of A. ucrainica in these Lakes, nor is this species commonly reported from other source waters. A. ucrainica was originally described from Ukraine by Skorbatov (personal communication, Hedy Kling, ATEI Winnipeg, Manitoba). A. ucrainica is a synonym for Anabaena mucosa, which is more commonly reported globally and has been found in the Great Lakes. A. mucosa, based on 16S rRNA gene phylogenetic analysis, is very closely related to a number of planktonic Anabaena strains, such as Anabaena crassa, Anabaena mucosa, Anabaena spiroides, Anabaena smithii and Anabaena sigmoidea (N 98.6% gene similarity) (Rajaniemi et al., 2005). These results highlight the key importance of this kind of molecular approach as a new analytical tool for the detection of T/O episodes which can advance our understanding of the diversity and distribution of geosmin producers. Such studies are much needed because of cost and logistics of current monitoring practices, the limited number of the cyanobacterial geoA sequences available in GenBank and the reliance of cyanobacterial taxonomy on highly variable morphological traits (e.g. as seen here in the similarity in A. ucrainica and A. mucosa). This is particularly of relevance given the constantly evolving classification systems which are increasingly based on a polyphasic approach

410

O.A. Kutovaya, S.B. Watson / Journal of Great Lakes Research 40 (2014) 404–414

combining morphological, biochemical and molecular characteristics (e.g. Johansen and Casamatta, 2005). Cyanobacterial geosmin synthase gene expression in cultures and environmental samples The regulation of geoA gene expression in cyanobacterial cultures and environmental samples was examined by RT-PCR in order to understand and predict the production of geosmin in the environment.

Cultures of Pseudanabaena sp. ECR 601, A. lemmermanni CA799, Aph. gracile SAG 31–79, L. wollei BH-LE, and G. splendidum NIVA 244 were grown under standard conditions (Materials and methods). These cyanobacteria were chosen as representatives of different taxonomic groups with potential differences in geosmin production rates but confirmed to produce geosmin (Table 1) (Jüttner and Watson, 2007; Watson, 2003). All tested strains yielded a geoA_cya RT-PCR product (data not shown), and demonstrated the suitability of the assay for expression studies. To confirm that the primers do not generate false

44 N Bay of Quinte Lake Ontario

43 N

83 W

82 W

81 W

80 W

79 W

78 W

77 W

83 W

82 W

81 W

80 W

79 W

78 W

77 W

Ocean Data View

Lake Erie

42 N

44 N

42 N

Ocean Data View

43 N

44 N

42 N

- gsm expression + gsm expression

83 W

82 W

81 W

80 W

79 W

78 W

77 W

Ocean Data View

43 N

Fig. 2. geoA_cya gene expression pattern and locations of the analyzed samples from Lake Erie, Lake Ontario, and Bay of Quinte shown by month. Sample sites shown as: open circles (no geosmin expression detected) and closed circles (geosmin expression detected).

BQ/MR BQ/HB BQ/PH BQ/N LO/1297 LO/211 LO/403 LO/206

LE/879 LE/970 LE/880 LE/1156

Oct. 2011

LO/3 (17m) LO/3 (1m) LO/1001

LE/938 LE/882 LE/1163 LE/944

Sep. 2011

LE/880

Feb. 2012

LE/341 LE/1326 LE/970

O.A. Kutovaya, S.B. Watson / Journal of Great Lakes Research 40 (2014) 404–414

Fig. 3. Results of the geoA_cya gene expression assay for environmental samples from Lake Erie (LE), Lake Ontario (LO), and Bay of Quinte (BQ) collected in September and October 2011, and February 2012. Shown is the image of the entire gel used to isolate and visualize the copies of geoA_cya gene amplified from the environmental samples. Individual samples were loaded into vertical lanes on both the top (LE February 2012) and bottom (LE, LO, and BQ in September and October 2011) of the gel. The upper band in each lane shows the dyed geoA_cya gene copies amplified in the assay, whereas the lower band shows the non-specific loading of the dye. Samples with no visible bands, such as two of the Lake Erie September 2011 samples, and one February 2012 sample were interpreted as the absence of geoA_cya expression. The majority of samples showed appreciable geoA_cya expression.

positives, the strains confirmed as negative by PCR and GC–MS were tested as well (Aph. flos-aquae EC HH06-01 and Oscillatoria limosa GI LBD 305b). These strains did not yield a geoA_cya RT-PCR product. Surface water samples from Lake Erie, Lake Ontario and the Bay of Quinte in two seasons were analyzed to evaluate gene expression. In September, samples from Lake Erie (8 samples), Lake Ontario (3 samples) and the Bay of Quinte (3 samples) showed weak (11 samples) or no expression (3 samples) of cyanobacterial geoA (Fig. 2). A subset of these samples (Fig. 3) demonstrates the variation in gene expression. This weak gene expression result was unexpected because most of the reported T/O outbreaks in the Great Lakes have occurred in the late summer–fall period (Anderson and Quartermaine, 1998; Moore and Watson, 2007; Watson et al., 2007). Although the assays were not quantitative, the low expression was striking compared to other samples. The change in signal from the cyanobacterial geoA assay was not due to nucleic acid quantity, since there was a robust signal from 16S rRNA. The lower level of gene expression in the September samples may have been due to lower abundance of geosmin-producing cyanobacteria, for instance as a result of competition, grazing, or viral lysis. Most of these

411

sites also showed low total cyanobacterial biomass, which was dominated by different cyanobacterial communities (Table 3). For example, in Lake Erie, the eastern basin sites LE/944 and LE/938, which were confirmed negative, had only 215 and 38 μg L−1 cyanobacterial biomass, respectively. These samples showed a predominance of small chroococcales such as Radiocystis geminata, Aphanocapsa incerta, Aphanocapsa holsatica, Aphanocapsa delicatissima, Merismopaedia spp. and Synechococcus sp., which are common in the central and eastern basins of Lake Erie (Wilhelm et al., 2006). Most of these taxa have not been documented to date as geosmin producers (e.g. Jüttner and Watson, 2007), with the exception of a few Synechococcus isolates from US reservoirs which have since apparently lost odor production in culture (Izaguirre and Taylor, 2004; G Izaguirre personal communication; Table 1). Similarly, the September sample from the western basin site LE/882 also had a low cyanobacterial biomass (19 μg L−1), and this only had weakly positive RT-PCR results (Fig. 3). Microscopic analysis showed that this sample was dominated by Microcystis spp. (nongeosmin producers) with very few Anabaena spp., consistent with previous studies that have reported a prevalence of Microcystis spp. in late summer in this basin (Brittain et al., 2001; Ouellette et al., 2006; Rinta-Kanto et al., 2005). The September sample from the Bay of Quinte site (BQ/Moira River mouth) which also had only weakly positive RTPCR (Fig. 3) contained ~1100 μg L−1 cyanobacterial biomass, dominated by a mixed assemblage of Anabaena spp. (A. flos-aquae, A. planktonica, A. lemmermannii, A. crassa), which our strain data suggest may have been predominantly weak or non producers of geosmin (Tables 1, 3). In contrast, the geoA gene was expressed in all October samples collected from Lake Erie (four samples), Lake Ontario (six samples), and Bay of Quinte (three samples) (Fig. 2). Fig. 3 shows the variation in gene expression for the four Lake Erie stations and four of the Lake Ontario stations. RT-PCR analysis of four February samples from Lake Erie indicated one sample showing weak expression (LE/880) and three samples positive for geoA expression (LE/1326/Maumee Bay, LE/970, LE/341) (Fig. 2, 3). Taste and odor outbreaks during winter months have been previously reported from other systems (Dzialowski et al., 2009; Wang et al., 2005), although are not a major issue in the Great Lakes (Moore and Watson, 2007). It has been asserted that temperature, nutrients, and chlorophyll play significant role in geosmin production (Aoyama et al., 1995; Naes et al., 1989; Rosen et al., 1992; Saadoun et al., 2001; Utkilen and Froshaug, 1992). These parameters were evaluated in all analyzed samples to establish any potential correlation with the cyanobacterial geoA gene expression (Fig. 4). The samples with no gene expression (two September samples from LE and one BQ) had similar environmental values as the samples with strong gene expression. Overall, the examined environmental factors described a wide range of temperature, nutrient concentrations and phytoplankton biomass throughout the different seasons and various locations and showed no apparent correlation with gene expression.

Table 3 Phytoplankton biomass (μg/L−1) for major taxonomic groups and listing of dominant species at the two western basin sites, LE/882 and LE/1163, two eastern basin sites, LE/938 and LE/ 944, and Bay of Quinte BQ/MR; sample dates as in Table S1. Cyan: Cyanobacteria; Chlor: Chlorophyta; Chrys: Chrysophyceae (Chrysophyta); Hapt: Haptophyta; Diat: Bacillariophyceae (Chrysophyta); Crypt: Cryptophyta; Din: Dinophyta. Site

LE 1163 LE 882 LE 938 LE 944 BQ MR

Taxonomic group Cyan

Chlor

Chrys

Hapt

Diat

Crypt

Dino

Dominant species

23,133

810

151

–

1214

182

14

19

6

b1

–

6

0.3

–

Planktothrix cf. suspensa, Pl. aghardhii; Anabaena flos-aquae; Aulacoseira granulata Anabaena flos-aquae

38

289

–

3

12

3

21

Coelastrum sp., Synechococcus sp., Gymnodinium sp.

215

95

29

34

38

37

134

1111

565

22

–

254

–

19

Anabaena planctonica (=D. planctonicum), Ceratium furcoides, Synechoccus spp. Anabaena (=Dolichospermum) lemmermannii, A. planktonica, A. crassa, Spirogyra, Aulacoseira ambigua

412

O.A. Kutovaya, S.B. Watson / Journal of Great Lakes Research 40 (2014) 404–414

Feb

LE

Sep Oct Sep

LO Oct Sep

BQ Oct 0

10 20

Temp. (°C)

0

20

Chl. a (µg L-1)

20 60 100

0 500 1500

0 40 80

TP (µg L-1)

NO3 (µg L-1)

NH3 (µg L-1)

-

+

gsm expression

Fig. 4. The distribution of environmental parameters co-occurring with geoA_cya gene expression. The data are divided by the different regions and the month of sample collection. Each column displays the distribution of observations for water temperature, phytoplankton biomass (as chl a; μg L−1), total phosphorus (μg L−1), nitrate (μg L−1), and ammonia (μg L−1). The final column shows the number of samples with detectable geoA_cya expression, with the number of symbols in the “–” column representing the number of samples with no observed expression of geosmin synthase. Symbol type is alternated for clarity and some observations may not be visible because of overlapping symbols.

These observed patterns in gene expression pattern suggest geoA expression occurs under a range of environmental conditions, and geosmin synthesis may be a frequently expressed metabolic pathway in cyanobacteria. All geoA-positive cyanobacterial cultures expressed the gene in culture, 85% of cultures possessing geoA had measurable geosmin, and most environmental samples measured detectable geoA transcripts. Although not definitive, the evidence suggests that geoA is constitutively expressed in cyanobacteria, but geosmin synthesis is perhaps regulated either post-translationally or by the supply of substrate from terpenoid synthesis. While we cannot exclude the possibility that geoA-positive cyanobacteria ceased expression of the gene in September 2011, it is more likely that changes in the abundance of geoApositive taxa in the community were the cause for the change in the presence of geoA transcripts. No correlations between a suite of environmental factors and geoA expression were detected, which is consistent with constitutive gene expression. Similarly, Giglio et al. (2011) demonstrated constitutive gene expression in Anabaena circinalis cultures exposed to a range of light conditions. Geosmin synthase gene expression was also evaluated by Ludwig et al. (2007) and it was shown that both of the genes involved in geosmin production were consistently expressed at 20 °C and a light–dark cycle of 12:12 h. Correlational analysis of environmental geosmin concentrations with the results of the environmental sampling for the geoA gene and gene expression was not possible due to logistical constraints. However, if geoA is constitutively expressed, then the geosmin concentration will be a function of the abundance of geoA-possessing organisms and the organism-specific geosmin production rate. Both variables would be tractable targets for future research, and the molecular assay presented in this report may be valuable to those future efforts.

Conclusions We developed and validated a molecular assay for tracking geosminproducing cyanobacteria. The assay was successfully tested on cultures representing potential geosmin producers, and then environmental samples from regions of the Great Lakes that often experience taste and odor episodes. Phylogenetic analysis of the geosmin synthase in the environmental samples indicated that most of the potential geosmin-producers

are represented by Nostocaceae family, particularly Anabaena spp. The assessment of the geosmin synthase expression pattern in the environmental samples from various watersheds and seasons suggests constitutive gene expression. It is likely that the observed geosmin production pattern in environment depends on the presence and success of the strains that carry the gene. Therefore, a molecular approach is critical to study and monitor geosmin producers and allows to overcome the limitations of microscopic and biochemical analyses. Our results, showing Anabaena species as important sources of geosmin in the Great Lakes, can be used as guidance for the water industry to target these species for monitoring and further research into understanding the environmental and biological controls of geosmin production. Acknowledgments This research was supported by an NSERC Discovery Grant 250057 / 2007 to SW and by in kind support from Environment Canada. We are indebted to H. Kling (ATEI, Winnipeg) for species identification and sample enumeration and the captain and crew of CCGS Limnos and Environment Canada technical operations staff for their field support. We would also like to thank B. Beall, S. Giglio, G. Bullerjahn and M. McKay for their valuable suggestions on the research and manuscript. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jglr.2014.03.016. References Agger, S.A., Lopez-Gallego, F., Hoye, T.R., Schmidt-Dannert, C., 2008. Identification of sesquiterpene synthases from Nostoc punctiforme PCC 73102 and Nostoc sp. strain PCC 7120. J. Bacteriol. 190, 6084–6096. Andersen, R.A., 2005. Algal Culturing Techniques. Elsevier Academic Press, San Diego, CA, USA (578 pp.). Anderson, B.C., Quartermaine, L.-K., 1998. Tastes and odours in Kingston's municipal drinking water: a case study of the problem and appropriate solutions. J. Great Lakes Res. 24, 859–867. Aoyama, K., Kawamura, N., Saitoh, M., Magara, Y., Ishibashi, Y., 1995. Interactions between bacteria-free Anabaena macrospora clone and bacteria isolated from unialgal culture. Water Sci. Technol. 31 (11), 121–126.

O.A. Kutovaya, S.B. Watson / Journal of Great Lakes Research 40 (2014) 404–414 Auffret, M., Pilote, A., Proulx, E., Proulx, D., Vandenberg, G., Villemur, R., 2011. Establishment of a real-time PCR method for quantification of geosmin-producing Streptomyces spp. in recirculating aquaculture systems. Water Res. 45, 6753–6762. Bafford, R.A., Seagull, R.W., Chung, S.-Y., Millie, D.F., 1993. Intracellular localization of the taste/odour metabolite 2-methylisoborneol in Oscillatoria limosa (Cyanophyta). J. Phycol. 29, 91–95. Bowmer, K.H., Padovan, A., Oliver, R.L., Korth, W., Ganf, G.G., 1992. Physiology of geosmin production by Anabaena circinalis isolated from the Murrumbidgee River, Australia. Water Sci. Technol. 25, 259–267. Brittain, S., Wang, S., Babcock-Jackson, L., Carmichael, W., Rinehart, K., Culver, D., 2001. Isolation and characterization of microcystins, cyclic heptapeptide hepatotoxins from a Lake Erie strain of Microcystis aeruginosa. J. Great Lakes Res. 26, 241–249. Cane, D.E., He, X., Kobayashi, S., Omura, S., Ikeda, H., 2006. Geosmin biosynthesis in Streptomyces avermitilis, molecular cloning, expression, and mechanistic study of the germacradienol/geosmin synthase. J. Antibiot. 59 (8), 471–479. Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction. Anal. Biochem. 162, 156–159. 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. Dereeper, A., Guignon, V., Blanc, G., Audic, S., Buffet, S., Chevenet, F., Dufayard, J.-F., Guindon, S., Lefort, V., Lescot, M., Claverie, J.-M., Gascuel, O., 2008. Phylogeny.fr: robust phylogenetic analysis for the non-specialist. Nucleic Acids Res. 36 (Web Server Issue), W465–W469. Dzialowski, A.R., Smith, V.H., Huggins, D.G., 2009. Development of predictive models for geosmin-related taste and odour in Kansas, USA, drinking water reservoirs. Water Res. 43, 2829–2840. Edwards, U., Rogall, T., Blöcker, H., Emde, M., Böttger, E.C., 1989. Isolation and direct complete nucleotide determination of entire genes. Characterization of a gene coding for 16S ribosomal RNA. Nucleic Acids Res. 17 (19), 7843–7853. Environment Canada, 1994. Major ions and nutrients. Manual of Analytical Methods, Vol. 1. Environmental Conservation Service — ECD. Canadian Communications Group, Toronto, Ontario, Canada. Gerber, N.N., Lechevalier, H.A., 1965. Geosmin, an earthy-smelling substance isolated from actinomycetes. Appl. Microbiol. 13, 935–938. Giglio, S., Jiang, J., Saint, C.P., Cane, D.E., Monis, P.T., 2008. Isolation and characterization of the gene associated with geosmin production in cyanobacteria. Environ. Sci. Technol. 42 (21), 8027–8232. Giglio, S., Chou, W.K.W., Ikeda, H., Cane, D.E., Monis, P.T., 2011. Biosynthesis of 2methylisoborneol in cyanobacteria. Environ. Sci. Technol. 45 (3), 992–998. Hanson, T.R., 2003. Economic impact of off-flavor to the U.S. catfish industry. In: Rimando, A.M., Schrader, K.K. (Eds.), Off-flavors in Aquaculture. American Chemical Society, Washington, DC, pp. 13–29. Hartman, W.L., 1973. Effects of exploitation, environmental changes, and new species on the fish habitats and resources of Lake Erie. Technical Report #22. Great Lakes Fishery Commission, Ann Arbor, Michigan, pp. 26–27. Howgate, P., 2004. Tainting of farmed fish by geosmin and 2-methyl-isoborneol: a review of sensory aspects and of uptake/depuration. Aquaculture 234, 155–181. Ilikchyan, I.N., McKay, R.M., Zehr, J.P., Dyhrman, S.T., Bullerjahn, G.S., 2009. Detection and expression of the phosphonate transporter gene phnD in marine and freshwater picocyanobacteria. Environ. Microbiol. 11, 1314–1324. Izaguirre, G., Taylor, W.D., 1998. A Pseudanabaena species from Castaic Lake, California, that produces 2-methylisoborneol. Water Res. 32 (5), 1673–1677. Izaguirre, G., Taylor, W.D., 2004. A guide to geosmin- and MIB-producing cyanobacteria in the United States. Water Sci. Technol. 49 (9), 19–24. Izaguirre, G., Taylor, W.D., Pasek, J., 1999. Off-flavor problems in two reservoirs, associated with planktonic Pseudanabaena species. Water Sci. Technol. 40 (6), 85–90. Jensen, S.E., Anders, C.L., Goatcher, L.J., Perley, T., Kenefick, S., Hrudey, S.E., 1994. Actinomycetes as a factor in odour problems affecting drinking water from the North Saskatchewan River. Water Res. 28, 1393–1401. Jiang, J., Cane, D.E., 2008. Geosmin biosynthesis. Mechanism of the fragmentationrearrangement in the conversion of germacradienol to geosmin. J. Am. Chem. Soc. 130, 428–429. Jiang, J., He, X., Cane, D.E., 2007. Biosynthesis of the earthy odourant geosmin by a bifunctional Streptomyces coelicolor enzyme. Nat. Chem. Biol. 3, 711–715. Johansen, J.R., Casamatta, D.A., 2005. Recognizing cyanobacterial diversity through adoption of a new species paradigm. Algol. Stud. 117 (1), 71–93. Jüttner, F., 1984. Dynamics of the volatile organic substances associated with cyanobacteria and algae in a eutrophic shallow lake. Appl. Environ. Microbiol. 47 (4), 814–820. Jüttner, F., Höflacher, B., Wurster, K., 1986. Seasonal analysis of volatile organic biogenic substances (VOBS) in freshwater phytoplankton populations dominated by Dinobryon, Microcystis and Aphanizomenon. J. Phycol. 22, 169–175. Jüttner, F., 1995. Physiology and biochemistry of odourous compounds from freshwater cyanobacteria and algae. Water Sci. Technol. 31, 69–78. Jüttner, F., Watson, S.B., 2007. Biochemical and ecological control of geosmin and 2-methylisoborneol in source-waters. Appl. Environ. Microbiol. 73 (4), 4395–4406. Klausen, C., Nicolaisen, M.H., Strobel, B.W., Warnecke, F., Nielsen, J.L., Jørgensen, N.O.G., 2005. Abundance of actinobacteria and production of geosmin and 2-methylisoborneol in Danish streams and fish ponds. Appl. Environ. Microbiol. 52, 265–278. Komárek, J., 2002. Problems in cyanobacterial taxonomy; implication for most common toxin producing species. In: Melchiorre, S., Viaggiu, E., Bruno, M. (Eds.), Rapporti ISTISAN. Istituto Superiore di Sanitá, Roma, pp. 6–43. Komárek, J., Zapomělová, E., 2007. Planktic morphospecies of the cyanobacterial genus Anabaena = subg. Dolichospermum — 1. Part: coiled types. Fottea 7 (1), 1–31.

413

Komárek, J., Zapomělová, E., 2008. Planktic morphospecies of the cyanobacterial genus Anabaena = subg. Dolichospermum — 2. Part: straight types. Fottea 8 (1), 1–14. Kotai, J., 1972. Instructions for preparation of modified nutrient solution Z8 for algae. Publication B-11/69. Norwegian Institute for Water Research, Blindern, Oslo 3. Norway. Kumar, S., Nei, M., Dudley, J., Tamura, K., 2007. MEGA: a biologist-centric software for evolutionary analysis of DNA and protein sequences. Brief. Bioinform. 9, 299–306. Kutovaya, O.A., McKay, R.M., Beall, B., Wilhelm, S.W., Kane, D., Chaffin, J., Bridgeman, T., Bullerjahn, G.S., 2011. Evidence against fluvial seeding of recurrent toxic blooms of Microcystis spp. in Lake Erie's western basin. Harmful Algae 15, 71–77. Ludwig, F., Medger, A., Bornick, H., Opitz, M., Lang, K., Gottfert, M., Roske, I., 2007. Identification and expression analyses of putative sesquiterpene synthase genes in Phormidium sp. and prevalence of geoA-like genes in a drinking water reservoir. Appl. Environ. Microbiol. 73, 6988–6993. Lund, J.W.G., Kipling, C., Le Cren, E.D., 1958. The inverted microscope method of estimating algal numbers and the statistical basis of estimations by counting. Hydrobiologia 11, 143–170. Makarewicz, J.C., 1993. A lakewide comparison of zooplankton biomass and its species composition in Lake Erie, 1983–87. J. Great Lakes Res. 19 (2), 275–290. Matsumoto, A., Tsuchiya, Y., 1988. Earthy-musty odour-producing cyanophytes isolated from five water areas in Tokyo. Water Sci. Technol. 20, 179–183. Medsker, L.L., Jenkins, D., Thomas, J.F., 1968. Odourous compounds in natural waters. An earthy-smelling compound associated with blue-green algae and actinomycetes. Environ. Sci. Technol. 2, 461–464. Millie, D.F., Fahnenstiel, G.L., Bressie, J.D., Pigg, R.J., Rediske, R.R., Klarer, D.M., Tester, P.A., Litake, R.W., 2009. Late-summer phytoplankton in western Lake Erie (Laurentian Great Lakes): bloom distributions, toxicity, and environmental influences. Aquat. Ecol. http://dx.doi.org/10.1007/s10452-009-9238-7. Moore, L., Watson, S.B., 2007. The Ontario Water Works Consortium: a functional model of source water management and understanding. Water Sci. Technol. 55 (5), 195–201. Naes, H., Post, A.F., 1988. Transient states of geosmin, pigments, carbohydrates and proteins in continuous cultures of Oscillatoria brevis induced by changes in nitrogen supply. Arch. Microbiol. 150, 333–337. Naes, H., Utkilen, H., Post, A.F., 1989. Geosmin production in the cyanobacterium Oscillatoria brevis. Arch. Microbiol. 151, 407–410. Nicholls, K., Carney, E., 2011. The phytoplankton of the Bay of Quinte, 1972–2008: pointsource phosphorus loading control, dreissenid mussel establishment, and a proposed community reference. Aquat. Ecosyst. Health Manag. 14 (1), 33–43. Nielsen, J.L., Klausen, C., Nielsen, P., Burford, M., Jorgensen, N., 2006. Detection of activity among uncultured actinobacteria in a drinking water reservoir. FEMS Microbiol. Ecol. 55 (3), 432–438. Nübel, U., Garcia-Pichel, F., Muyzer, G., 1997. PCR primers to amplify 16S rRNA genes from cyanobacteria. Appl. Environ. Microbiol. 63, 3327–3332. Ouellette, A.J.A., Handy, S.M., Wilhelm, S.W., 2006. Toxic Microcystis is widespread in Lake Erie: PCR detection of toxin genes and molecular characterization of associated cyanobacterial communities. Microb. Ecol. 51, 154–165. http://dx.doi.org/10.1007/ s00248-004-0146-z. Persson, P.E., 1988. Odourous algal cultures in culture collections. Water Sci. Technol. 20, 211–213. Persson, P.E., 1996. Cyanobacteria and off-flavours. Phycologia 35 (6S), 168–171. Rajaniemi, P., Hrouzek, P., Kastovska, K., Willame, R., Rantala, A., Hoffmann, L., Komárek, J., Sivonen, K., 2005. Phylogenetic and morphological evaluation of the genera Anabaena, Aphanizomenon, Trichormus and Nostoc (Nostocales, Cyanobacteria). Int. J. Syst. Evol. Microbiol. 55, 11–26. Rinta-Kanto, J.M., Ouellette, A.J.A., Twiss, M.R., Boyer, G.L., Bridgeman, T.B., Wilhelm, S.W., 2005. Quantification of toxic Microcystis spp. during the 2003 and 2004 blooms in Western Lake Erie. Environ. Sci. Technol. 39, 4198–4205. Rosen, B., MacLeod, B., Simpson, M., 1992. Accumulation and release of geosmin during the growth phases of Anabaena circinalis (Kütz.) Rabenhorst. Water Sci. Technol. 25, 185–190. Saadoun, I., Schrader, K.K., Blevins, W.T., 2001. Environmental and nutritional factors affecting geosmin synthesis by Anabaena sp. Water Res. 35, 1209–1218. Schrader, K., Dennis, M.E., 2005. Cyanobacteria and earthy/musty compounds found in commercial catfish (Ictalurus punctatus) ponds in the Mississippi delta and Mississippi–Alabama blackland prairie. Water Res. 39 (13), 2807–2814. Smith, V.H., Sieber-Denlinger, J., deNoyelles, F., Campell, S., Pan, S., Randtke, S.J., Blain, G., Strasser, V.A., 2002. Managing taste and odour problems in a eutrophic drinking water reservoir. Lake Reservoir Manage. 18 (4), 319–323. Sugiura, N., Nakano, K., 2000. Causative microorganisms for musty odour occurrence in the eutrophic Lake Kasumigaura. Hydrobiologia 434, 145–150. Taylor, W.D., Losee, R.F., Torobin, M., Izaguirre, G., Sass, D., Khiari, D., Atasi, K., 2006. Early Warning and Management of Surface Water Taste-and-odour Events. AWWA Research Foundation, Denver, CO (237 pp.). Utkilen, H.C., Froshaug, M., 1992. Geosmin production and excretion in a planktonic and benthic Oscillatoria. Water Sci. Technol. 25 (2), 199–206. Wang, S.H., Dzialowski, A.R., Lim, N.C., Spotts, W.W., Huggins, D.G., deNoyelles, F., 2005. Relationships between cyanobacterial production and the physical and chemical properties of a Midwestern reservoir, USA. Hydrobiologia 541 (1), 29–43. Watson, S.B., 2003. Cyanobacterial and eukaryotic algal odour compounds: signals or byproducts? A review of their biological activity. Phycologia 42 (4), 332–350. Watson, S., Ridal, J., 2004. Periphyton: a primary source of widespread and severe taste and odour. Water Sci. Technol. 49, 33–39. Watson, S.B., Brownlee, B., Satchwill, T., Hargesheimer, E.E., 2000. Quantitative analysis of trace levels of geosmin and MIB in source and drinking water using headspace SPME. Water Res. 34, 2818–2828.

414

O.A. Kutovaya, S.B. Watson / Journal of Great Lakes Research 40 (2014) 404–414

Watson, S.B., Millard, S., Burley, M., 2005. Taste–odour and toxins in the Bay of Quinte. Project Quinte Monitoring Report #16. Bay of Quinte RAP Restoration Council/Project Quinte, Trenton, Ontario (164 pp.). Watson, S.B., Charlton, M., Yerubandi, R., Howell, T., Brownlee, B., Marvin, C., Ridal, J., Millard, S., 2007. Off-flavour issues in large waterbodies: physics, chemistry and biology in concert. Water Sci. Technol. 55 (5), 1–8. Watson, S.B., Ridal, J., Boyer, G.L., 2008. Taste and odour and cyanobacterial toxins: impairment, prediction, and management in the Great Lakes. Can. J. Fish. Aquat. Sci. 65 (8), 1779–1796. Wilhelm, S.W., Bullerjahn, G.S., Eldridge, M.L., Rinta-Kanto, J.M., Poorvin, L., Bourbonniere, R.A., 2006. Seasonal hypoxia and the genetic diversity of prokaryote populations in

the central basin hypolimnion of Lake Erie: evidence for abundant cyanobacteria and photosynthesis. J. Great Lakes Res. 32, 657–671. Yagi, M., Nakashima, S., Muramoto, S., 1988. Biological degradation of musty odour compounds, 2-methylisoborneol and geosmin, in a bio-activated carbon filter. Water Sci. Technol. 20, 255–260. Zaitlin, B., Watson, S.B., 2006. Actinomycetes in relation to taste and odour in drinking water: myths, tenets and truths. Water Res. 40, 1741–1753. Zaitlin, B., Watson, S.B., Ridal, J., Satchwill, T., Parkinson, D., 2003. Actinomycetes in Lake Ontario: habitats and geosmin and MIB production. J. Am. Water Works Assoc. 95, 113–118.