Alberta Environment Cyanotoxin Program Status Report Government A/\ L „ .. of Alberta ■ U€^V(^J Freedom To Create. Spirit To Achieve. Alberta Environment Cyanotoxin Program Status Report Prepared by: Ron Zurawell, Ph.D., P.Biol. Li m nolog ist/Water Quality Specialist Water Policy Branch Environmental Assurance Division Alberta Environment August 2010 ISBN: 978-0-7785-9013-2 (Printed Edition) ISBN: 978-0-7785-9014-9 (On-Line Edition) Web Site: http://environment.gov.ab.ca/info/home.asp Any comments, questions, or suggestions regarding the content of this document may be directed to: Water Policy Branch Alberta Environment 7th Floor, Oxbridge Place 9820 - 1 06th Street Edmonton, Alberta T5K 2J6 Phone: (780)427-6278 Fax: (780)422-6712 Additional copies of this document may be obtained by contacting: Information Centre Alberta Environment Main Floor, Oxbridge Place 9820 - 106th Street Edmonton, Alberta T5K 2J6 Phone: (780)427-2700 Fax: (780)422-4086 Email: env.infocent@gov.ab.ca EXECUTIVE SUMMARY Alberta’s nutrient-rich lakes and reservoirs often experience blooms of cyanobacteria (a.k.a. blue-green algae) during summer and early fall. Common species of cyanobacteria can produce potent liver and/or nerve toxins. Microcystins (MCYSTs) are thought to be the most common toxins produced by cyanobacteria and exert toxicity by severely damaging liver structure and function. They have been linked to high incidence of primary liver (hepatocellular carcinoma) and colorectal cancers in some countries. Cyanotoxin monitoring was incorporated into Alberta’s Integrated Lake and Reservoir Monitoring Program in 2005 with the goal of determining the prevalence of MCYST in Alberta. In addition, we investigate the occurrence of multiple MCYST analogues, a potent neurotoxin anatoxin-a (ATX-a), and a potentially toxic amino acid, p-A-methylamino-L-alanine (BMAA) recently reported to be produced by cyanobacteria. Monitoring data collected during 4 open water seasons (2005-2008), reveal that MCYSTs are prevalent in a majority of Alberta’s lakes and reservoirs. Though appearing more common and at higher concentrations in nutrient-rich, eutrophic and hypereutrophic waterbodies, MCYSTs did occur periodically in lower nutrient, oligo- and mesotrophic waters. Metalimnetic blooms of specific cyanobacteria may account for toxin in these low nutrient environments. Average off-shore (open-water) concentrations can be elevated, occasionally exceeding the draft recreational water quality (RWQ) guideline of 20 pg/L. Though not determined in this study, concentrations in near-shore bloom accumulations do become much (2 orders of magnitude) higher than that in open- water areas. MCYST concentration is correlated with the abundance of toxin-producing cyanobacteria. Cyanobacterial growth and reproduction and the onset, duration, and severity of surface blooms are likely influenced by climate-mediated water column warming and stability. Warmer years could have greater occurrence and concentrations of microcystins. In a subset of samples, multiple MCYST analogues were detected. While microcystin-LR (MCLR) appears to be most common, other toxin congeners occurred at higher concentrations and in the absence of MCLR. Toxicity of MCYST analogues can vary greatly with specific chemical structure. These findings challenge the applicability or suitability of the current Canadian Drinking Water Quality Guideline specifying only MCLR and not total MCYST concentration. In contrast to MCYST, ATX-a occurred infrequently and at low concentrations in Alberta’s lakes and reservoirs monitored in 2005 and was not detected during the summer of 2006. Analytical methods for determining BMAA in surface water samples have been developed. Initial work indicates this compound may occur in Alberta eutrophic surface waters. Studies on the prevalence of BMAA are ongoing. Alberta Environment Cyanotoxin Program Status Report l TABLE OF CONTENTS EXECUTIVE SUMMARY i TABLE OF CONTENTS ii LIST OF TABLES iv LIST OF FIGURES vi GLOSSARY OF TERMS AND ACRONYMS viii ACKNOWLEDGEMENTS x 1.0 INTRODUCTION 1 1 .1 Cyanobacteria Occurrence and Toxicity 1 1 .2 Drinking Water Guidelines 3 1.3 Monitoring Program and Objectives 4 2.0 METHODS 5 3.0 RESULTS AND DISCUSSION 6 3.1 Microcystin Monitoring 2005 - 2008 6 2005 Results 6 2006 Results 7 2007 Results 8 2008 Results 10 Microcystin Prevalence 2005 - 2008 Summary Data 10 Microcystin in Low Nutrient Lakes 11 Year-to-Year Microcystin Variability 12 Climate Change and Prevalence of Toxic Cyanobacteria 13 3.2 Microcystin Analogue Study 13 3.3 Anatoxin-a Study 17 3.4 p-AAMethylamino-L-Alanine (BMAA) Study 18 4.0 GENERAL DISCUSSION 21 5.0 CONCLUSIONS 23 6.0 REFERENCES.... 24 Alberta Environment Cyanotoxin Program Status Report ii APPENDIX I LABORATORY METHODS 27 Total Microcystin Analysis 28 Microcystin Analogue Analysis 28 Anatoxin-a Analysis 29 P-/V-Methylamino-L-Alanine (BMAA) Methods 29 Phytoplankton Community Analysis 30 APPENDIX II LIST OF LAKES/RESERVOIRS 31 APPENDIX III MAXIMUM MICROCYSTIN CONCENTRATIONS IN LAKES/RESERVOIRS 37 APPENDIX IV QUALITY ASSURANCE/QUALITY CONTROL 42 Microcystin Analysis QA/QC 43 APPENDIX V RAW DATA 45 Alberta Environment Cyanotoxin Program Status Report iii LIST OF TABLES Table 1 Structure and comparative toxicity of 20 primary microcystin analogues based on interperitoneal LD50 in mice (pg/kg); ND, not determined; Aba, L-aminoisobutyric acid; Hil, L-homoisoleucine; Hty, L-homotyrosine; M(O), methionine S-oxide; (H4)Tyr, 1 ,2,3,4- tetrahydrotyrosine (adapted from Zurawell, 2001) 2 Table 2 Summary of euphotic integrated composite samples collected during the open-water (May through October) seasons 2005 through 2008 6 Table 3 Summary of surface water samples with MCYST analogues detected in 2005 by ARCV 14 Table 4 Summary of surface water samples with MCYST analogues detected in 2007 by ACFT 14 Table 5 IC50 of PP1 by 7 MCYST analogues and mean inhibition relative to MCLR (ACFT unpublished data). Values for mean inhibition greater than 1 indicate less toxic than MCLR; values less than 1 indicate more toxic than MCLR 15 Table 6 Growth inhibition of OATP transfected cells lines by MCYST analogues (adapted from Monks et al. 2007) 16 Table 7 RT-CES based cytotoxicity of OATP1B3 cells lines by MCYSTs (ACFT unpublished data) 16 Table 8 Summary of cyanobacteria abundance, relative abundance and biomass in water samples analyzed for BMAA in 2005 19 Table Al List of lakes and reservoirs sampled for total microcystin in 2005 including trophic status, parent monitoring program and source of funding for toxin analyses 32 Table A2 List of lakes and reservoirs sampled for total microcystin in 2006 including trophic status, parent monitoring program and source of funding for toxin analyses 33 Table A3 List of lakes and reservoirs sampled for total microcystin in 2007 including trophic status, parent monitoring program and source of funding for toxin analyses 34 Table A4 List of lakes and reservoirs sampled for total microcystin in 2008 including trophic status, parent monitoring program and source of funding for toxin analyses 35 T able A5 List of lakes and reservoirs sampled for anatoxin-a in 2005 including trophic status and parent monitoring program 36 Alberta Environment Cyanotoxin Program Status Report iv Table A6 Results of paired analyses for total MCYST by PPI assay and LC- MS/MS analysis for specific MCYST analogues results for MCLR, MCRR and MCYR 44 Table A7 Raw Data 2005 46 Table A8 Raw Data 2006 49 Table A9 Raw Data 2007 52 Table A10 Raw Data 2008.. 55 Alberta Environment Cyanotoxin Program Status Report v LIST OF FIGURES Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Generalized chemical structure of MCYST: where position (1) is D- Alanine; (2) X is a variable L-amino acid; (3) is D-erythro-p- methylaspartic acid; (4) Y is another variable L-amino acid; (5) is Adda, (2S, 3S, 8S, 9S)-3-amino-9-methoxy-2, 6, 8-trimethyl- 10- phenyldeca-4,6-dienoic acid; (6) is D-Glutamic acid and (7) is N- methyldehydroalanine (adapted from Zurawell, 2001) 1 Number of surface water samples from 2005 with nondetectable (85 samples; blue portion of pie chart) vs. detectable (79 samples; red portion of pie chart) levels of MCYST. The bar summarizes the number of samples with MCYST at various incremental concentrations 7 Number of surface water samples from 2006 with nondetectable (85 samples; blue portion of pie chart) vs. detectable (97 samples; red portion of pie chart) levels of MCYST. The bar summarizes the number of samples with MCYST at various incremental concentrations 8 Number of surface water samples from 2007 with nondetectable (40 samples; blue portion of pie chart) vs. detectable (124 samples; red portion of pie chart) levels of MCYST. The bar summarizes the number of samples with MCYST at various incremental concentrations 9 Number of surface water samples from 2008 with nondetectable (94 samples; blue portion of pie chart) vs. detectable (126 samples; red portion of pie chart) levels of MCYST. The bar summarizes the number of samples with MCYST at various incremental concentrations 10 Prevalence of MCYST in lakes/reservoirs of varying trophic status from 2005-2008. *MCYST prevalence is the percentage of lakes/reservoirs containing detectable toxin (> 0.1 pg MCLReq./L) on at least one sampling occasion 11 Prevalence of MCYST in surface water samples grouped by sampling month (2005-2008). *MCYST prevalence is the percentage of samples containing detectable toxin (> 0.1 pg MCLReq./L) 12 Number of surface water samples with nondetectable (59 samples; blue portion of pie chart) vs. detectable (7 samples; red portion of pie chart) levels of ATX-a in 2005. The bar summarizes the number of samples with ATX-a at various incremental concentrations 17 Pie charts depicting the relative biomass of phytoplankton groups within samples collected for BMAA analysis from Winagami Lake, Alberta Environment Cyanotoxin Program Status Report vi Moonshine Lake Reservoir, Sturgeon Lake and Baptiste Lake South Basin. The bar presents the data for phytoplankton groups comprising less than 10% or the overall biomass 19 Figure A1 Maximum Microcystin Concentrations in Lakes/Reservoirs Sampled June - September 2005 38 Figure A2 Maximum Microcystin Concentrations in Lakes/Reservoirs Sampled June - September 2006 39 Figure A3 Maximum Microcystin Concentrations in Lakes/Reservoirs Sampled June - September 2007 40 Figure A4 Maximum Microcystin Concentrations in Lakes/Reservoirs Sampled June - September 2008 41 Alberta Environment Cyanotoxin Program Status Report vii GLOSSARY OF TERMS AND ACRONYMS \xg MCLR eq./L Unit of measure estimating micrograms of total microcystin concentration per liter water based on a toxicity response equivalent to a known concentration of microcystin-LR. ACFT The Alberta Centre for Toxicology, University of Calgary. AENV Alberta Environment ALMS The Alberta Lake Management Society ALS/PDC Amyotrophic Lateral Sclerosis/Parkinsonism-Dementia Complex: a neurodegenerative disorder with symptoms consistent with or similar to ALS and Parkinson’s diseases. API Alberta Peptide Institute, University of Alberta, Edmonton. ARC Alberta Research Council. ATX-a Anatoxin-a secondary, bicyclic amine alkaloid neurotoxin. BMAA P-N-methylamino-L-alanine amino acid neurotoxin DWQ Drinking water quality. Euphotic Zone The depth of the water column in a lake or reservoir that is exposed to sunlight intensity sufficient for supporting photosynthesis. It extends from the water’s surface (air- water interface) to a depth where light intensity is equal to 1 percent of that at the surface. Eutrophic Elevated trophic state. A lake or reservoir with high productivity (high concentrations of nutrients and resulting high algal/plant growth). GC-MS Gas chromatography linked mass spectrometry HPLC High performance liquid chromatography Hypereutrophic Highly elevated trophic state productivity. A lake or reservoir with very high productivity (very high concentrations of nutrients and resulting very high algal/plant growth). IC50 The concentration of a substance that inhibits a biological response or process (i.e., enzyme activity, cell growth) by 50%. LC-MS/MS Liquid chromatography linked tandem mass spectrometry. Alberta Environment Cyanotoxin Program Status Report viii LD50 The dose of a substance that causes mortality to 50% of the test organisms. LTLN Alberta Environment’s Long-term Lake Network MAC Maximum acceptable concentration. MCLR Microcystin analogue with (L)eucine and a(R)ginine in variable amino acid positions 2 and 4. MCYST Microcystin cyclic peptide hepato- (liver) toxin. Mesotrophic Moderate trophic state. A lake or reservoir with moderate productivity (moderate concentrations of nutrients and resulting moderate algal/plant growth). Metalimnetic Bloom An accumulation of buoyancy-regulating cyanobacteria, notably Planktothrix sp ., within a distinct depth stratum at or near the metalimnion (transition between the well-mixed upper illuminated layer and deep, isolated waters) of a stratified lake or reservoir. NOAEL No-observed- adverse-effects-level: denotes the level of exposure of a test organism, at which there is no biologically or statistically significant increase in the frequency or severity of any adverse effects in the exposed population when compared to its appropriate control. OATPs Organic anion transporter polypeptides are transmembrane proteins expressed in various organs/tissues that function in the uptake of compounds including enzymes, drugs and toxins. Oligotrophic Low trophic state. A lake or reservoir with low productivity (low concentrations of nutrients and resulting low algal/plant growth). PPI Protein phosphatase enzyme inhibition. PPLMP Alberta Environment’s Provincial Parks Lake Monitoring Program. TDI Tolerable daily intake: an estimate of the intake of a substance over a lifetime that is considered acceptable without appreciable health risk. Trophic State The overall level of biological productivity (or fertility) of a lake and is usually defined by concentrations of key nutrients (primarily phosphorus) and algae that are present WHO The World Health Organization. Alberta Environment Cyanotoxin Program Status Report ix ACKNOWLEDGEMENTS A significant portion of the data described in this report were collected through a joint research project between AENV and the Alberta Research Council (ARC) and funded by the Water Research Users Group. We acknowledge the collaborative efforts of Dave Humphries, Grant Prill, and Irene Gaudet (Organic Chemistry Laboratory) and Dr. Jim Davies (ToxTest Laboratory) of the ARC in Vegreville. We offer special gratitude and are indebted to Dr. David. Kinniburgh, Dr. Dorothy Yu Huang, Xu Zhang and others at The Alberta Centre for Toxicology for their extraordinary in-kind support, service and collaboration over the past 3 years. We gratefully recognize the Alberta Lake Management Society, Provincial Parks Staff (Alberta Department of Tourism, Parks, Recreation and Culture) and AENV field monitoring staff for their continued dedication to Alberta’s integrated lake and reservoir monitoring program. Special thanks Mary Raven (EAD) for formatting this report and Lisa Brodziak (EAD) for assisting with data management. Alberta Environment Cyanotoxin Program Status Report x 1.0 INTRODUCTION 1.1 Cyanobacteria Occurrence and Toxicity Eutrophic (nutrient-rich) lakes and reservoirs in Alberta often experience blooms of cyanobacteria during summer and early fall. It is well known that some common species of cyanobacteria produce potent liver and/or nerve toxins. Microcystins (MCYSTs) are the most common group of toxins produced by cyanobacteria and exert toxicity by severely damaging the structure and function of the liver representing a significant threat to humans (Dunn, 1996), pets and livestock (Gurney and Jones, 1997). They are tumor promoters and chronic exposure to MCYSTs has been linked to high incidence of primary liver (hepatocellular carcinoma) and colorectal cancers in rural human populations around the world (Zhou et al, 2002). Recent studies have also shown MCYSTs to induce oxidative DNA damage in liver cell isolates suggesting the toxin can initiate cancer (Zegura et al ., 2003). Several bloom-forming species of cyanobacteria (including Microcystis aeruginosa , M. flos- aquae , M. wesenbergii, Anabaena flos-aquae , A. circinalis, A. lemmermannii , Planktothrix agardhii and P. rubescens ) produce MCYSTs. To date, more than 80 toxin analogues of MCYST have been isolated and described globally. They are small monocyclic peptides composed of seven amino acids including an amino acid residue, abbreviated Adda, unique to MCYSTs (and Nodularin from marine cyanobacteria) and critical for toxicity (Figure 1). The 20 or so primary MCYST analogues differ with respect to variable L-amino acids at positions 2 and 4 - denoted by X and Y, respectively. Of these, microcystin-LR (MCLR), which possesses leucine (L) and arginine (R) at positions 2 and 4, was one of the first discovered and consequently is the most studied congener (Table 1). Alterations and substitutions of other constituent amino acids, including demethylation of D-MeAsp and/or Mdha (positions 3 and 7, respectively), methyl esterification of D-Glu (position 6) and geometric isomerization of Adda (position 5) result in numerous additional toxic and non-toxic congeners (reviewed in Zurawell, 2001). (6) (7) D-Glu Mdha Figure 1 Generalized chemical structure of MCYST: where position (1) is D-Alanine; (2) X is a variable L-amino acid; (3) is D-erythro-p-methylaspartic acid; (4) Y is another variable L-amino acid; (5) is Adda, (2S, 3S, 8S, 9S)-3-amino-9-methoxy-2,6,8-trimethyl-10-phenyldeca-4,6-dienoic acid; (6) is D-Glutamic acid and (7) is N-methyldehydroalanine (adapted from Zurawell, 2001). Alberta Environment Cyanotoxin Program Status Report 1 Table 1 Structure and comparative toxicity of 20 primary microcystin analogues based on interperitoneal LD50 in mice (pg/kg); ND, not determined; Aba, L-aminoisobutyric acid; Hil, L-homoisoleucine; Hty, L-homotyrosine; M(O), methionine S-oxide; (H4)Tyr, 1,2,3,4-tetrahydrotyrosine (adapted from Zurawell, 2001). Analogue Structure Toxicity Microcystin-AR cyclo (-D-Ala-L-Ala-D-MeAsp-L-Arg-Adda-D-Glu-Mdha-) 250 Microcystin-FR cyclo (-D-Ala-L-Phe-D-MeAsp-L-Arg-Adda-D-Glu-Mdha-) 250 Microcystin-HilR cyclo (-D-Ala-L-Hil-D-MeAsp-L-Arg-Adda-D-Glu-Mdha-) 100 Microcystin-HtyR cyclo (-D-Ala-L-Hty-D-MeAsp-L-Arg-Adda-D-Glu-Mdha-) 80-100 Microcystin-LA cyclo (-D-Ala-L-Leu-D-MeAsp-L-Ala-Adda-D-Glu-Mdha-) 50 Microcystin-LAba cyclo (-D-Ala-L-Leu-D-MeAsp-L-Aba-Adda-D-Glu-Mdha-) ND Microcystin-LF cyclo (-D-Ala-L-Leu-D-MeAsp-L-Phe-Adda-D-Glu-Mdha-) ND Microcystin-LL cyclo (-D-Ala-L-Leu-D-MeAsp-L-Leu-Adda-D-Glu-Mdha-) ND Microcystin-LM cyclo (-D-Ala-L-Leu-D-MeAsp-L-Met-Adda-D-Glu-Mdha-) ND Mlcrocystin-LR cyclo (-D-Ala-L-Leu-D-MeAsp-L-Arg-Adda-D-Glu-Mdha-) 50 Microcystin-LV cyclo (-D-Ala-L-Leu-D-MeAsp-L-Val-Adda-D-Glu-Mdha-) ND Microcystin-LY cyclo (-D-Ala-L-Leu-D-MeAsp-L-Tyr-Adda-D-Glu-Mdha-) 90 Microcystin-M(0)R cyclo (-D-Ala-L-Met(O)-D-MeAsp-L-Arg-Adda-D-Glu-Mdha-) 700-800 Microcvstin-RA cyclo (-D-Ala-L-Arg-D-MeAsp-L-Ala-Adda-D-Glu-Mdha-) ND Microcystin-RR cyclo (-D-Ala-L-Arg-D-MeAsp-L-Arg-Adda-D-Glu-Mdha-) 500-800 Microcystin-WR cyclo (-D-Ala-L-T ry-D-MeAsp-L-Arg-Adda-D-Glu-Mdha-) 150-200 Microcystin-YA cyclo (-D-Ala-L-Tyr-D-MeAsp-L-Ala-Adda-D-Glu-Mdha-) 60-70 Microcystin-YM(O) cyclo (-D-Ala-L-Tyr-D-MeAsp-L-Met(O)-Adda-D-Glu-Mdha-) 56-110 Microcystin-YR cyclo (-D-Ala-L-Tyr-D-MeAsp-L-Arg-Adda-D-Glu-Mdha-) 150-200 Microcystin-(H4)YR cyclo (-D-Ala-L-(H4)Tyr-D-MeAsp-L-Arg-Adda-D-Glu-Mdha-) ND In addition to MCYSTs, some species of cyanobacteria including A. flos-aquae and A. spiroides , produce a nerve toxin called anatoxin-a (ATX- a). This is a highly potent toxin that interferes with nervous system function by disrupting the normal propagation of nerve impulses from neurons to muscles, with the potential to cause paralysis and death via respiratory failure and suffocation in animals. Like MCYST, the occurrence of ATX-a is a global concern because it has been linked to the deaths of livestock, pets and wildlife around the world, including the Canadian prairies (Zurawell, 2001). In all previous cases within Alberta, the assumption of ATX- a poisoning has largely been based on the presence of cyanobacteria and symptoms in affected animals. The actual presence of ATX-a in these instances has not been documented prior to the work reported here. Beta-A-methylamino-L-alanine (BMAA) has recently been identified in terrestrial, marine and freshwater cyanobacteria (Cox et al , 2005). The amino acid has been linked to Alzheimer's-like neurodegeneration in human populations and is a possible causative agent of Amyotrophic Lateral Sclerosis/Parkinsonism-Dementia Complex (ALS/PDC; Kurland and Mulder, 1954; Spencer et al. , 1987). Additionally, it has been isolated and identified in brain tissue of Canadians who have succumbed to Alzheimer’s disease, though the source of BMAA in these instances remains unknown (Murch et al ., 2004a). Unlike MCYST or ATX-a, which elicit immediate acute toxicity, it appears BMAA can accumulate in tissues and bio-magnify through the food chain. It is hypothesized that accumulated BMAA is slowly released over years as proteins are naturally metabolized, causing recurrent neurological damage (Murch et al ., 2004b). Note that the proposed mechanism requires further assessment and an appropriate animal model relevant to humans must be demonstrated to validate the role of this putative neurotoxin. The question exists whether BMAA occurs in cyanobacteria inhabiting Alberta’s surface waters. Alberta Environment Cyanotoxin Program Status Report 2 1 .2 Drinking Water Guidelines Global concern over the implications of MCLR to human health peaked during the late- 1990’s in the wake of human tragedies in Caruaru, Brazil in 1996 - 49 kidney dialysis patients died from acute liver failure following microcystin exposure through dialysis treatment (Pouria et al, 1998). Following the lead of the World Health Organization (WHO), Health Canada sought to develop a drinking water quality (DWQ) guideline for MCLR, which was derived in accordance with the WHO’s approach. First, a tolerable daily intake (TDI) for MCLR was calculated based on: (1) an estimated no-observed-adverse-effects-level ( NOAEL ) following chronic exposure of mice to purified MCLR (study by Fawell et al., 1994) and (2) provision for uncertainties ( UF) due to intra- and interspecies variation and a less-than-lifetime study (see Equation 1). Equation (1): TDI = NOAEL UF 40 pg/kg bw/d 1000 = 0.04 pg/kg bw/d where: • NOAEL = a no-observed-adverse effects level of 40 pg/kg body weight per day derived from observed liver changes during a 13-week mouse study conducted by Fawell et al. (1994). • UF = an uncertainty factor of 1000; the product of applying a 10x factor each for: (1) intraspecies variation, (2) interspecies variation and (3) a less-than-lifetime study. Next, the TDI was multiplied with the body weight of an average adult in Canada (bw) and with the proportion of total toxin exposure attributed to the consumption of drinking water (P). This product, divided by the average volume of drinking water ( L ) consumed by an adult in Canada is equal to a maximum acceptable concentration (MAC) of 1.5 pg/L for total MCLR (applies to the sum of intra- and extracellular MCLR) in finished drinking water (see Equation 2; Health Canada, 2002). While the approach parallels that of the WHO, the derivation of the Canadian DWQ guideline differs with respect to values considered for: (1) the average body weight of an adult (70 kg Canadian vs. 60 kg WHO) and (2) the average daily consumption rate of drinking water (1.5 L/d Canadian vs. 2 L/d WHO); and using the values specified by the WHO yields a lower provisional guideline of 1 .0 pg/L MCLR. Equation (2): MAC = TDI x bw x P T~~ 0.04 pg/kg bw/d * 70 kg bw * 0.8 L/d = 1 .5 pg/L where: • TDI = the calculated tolerable daily intake of 0.04 pg/kg bw per day in equation 1 (above). • bw = the average body weight of an adult in Canada; 70 kg is typically used. Note: WHO used 60 kg - the international unit adult average weight. • P = the proportion of total toxin intake (exposure) attributed to ingestion of drinking water. Drinking water consumption is felt to be the major route of microcystin exposure, hence 80% was used. • L = the average daily consumption of drinking water for an adult in Canada is 1.5 L/d. Note: WHO used 2 L/d average daily consumption of drinking water. Alberta Environment Cyanotoxin Program Status Report 3 Note, however, that neither the WHO nor Health Canada have accounted for total MCYST, only MCLR in their derived MACs. In fact, the WHO guideline is stated as being ‘provisional’ citing insufficient toxicological data on other analogues to warrant inclusion (at the time). In contrast, Australia developed a guideline for total MCYST based on MCLR “toxicity equivalents” (i.e., to infer concentration of MCYST in a sample by determining concentration of purified MCLR required to generate an equivalent toxic response). Their rationale included the fact that blooms of M. aeruginosa (the most common bloom-forming cyanobacterium in Australia) contain several to many analogues (more than 20 in some cases) in an individual sample and that cumulative toxicity of MCYSTs represents the potential threat to human health via drinking water consumption (Nicholson and Burch, 2001). At the time of (DWQ) guideline development, the ability of MCLR to promote tumour growth was only suspected. Thus, provisions for this characteristic were not incorporated in the uncertainty factor of the TDI calculation. Subsequent studies not only confirm the ability of MCYSTs to promote tumour growth, but have established a link between toxin exposure and primary liver and colorectal cancers in human populations (Nishiwaki-Matsushima et al ., 1992; Humpage et al., 2000; Zhou et al, 2002). 1.3 Monitoring Program and Objectives Research into the occurrence of MCYSTs in Alberta’s lakes and reservoirs began in the late 1980’s. However, most of these studies focused on MCLR and as a result less information exists on the prevalence of other congeners - some of which possess potency similar to that of MCLR. In 2005, AENV incorporated sampling and analysis of water for total MCYST concentration into its Provincial Lake monitoring programs including the Long-term Lake Network (LTLN) and the priority-one lakes of the Provincial Parks Lake Monitoring Program (PPLMP). In addition, joint research between AENV and the Alberta Research Council (ARC, Vegreville) permitted the analysis of total microcystin in 100 samples collected from priority-two lakes of the PPLMP and lakes sampled by the Alberta Lake Management Society’s (ALMS) Lake Monitoring program - Lakewatch. MCYST analysis was added to all lake and reservoir monitoring programs province- wide in 2006 and continued through 2008. Provincial monitoring programs included the LTLN, PPLMP, Southern AB Lakes and Reservoirs Program, Central Recreational Lake Monitoring Program, Elk Island National Park Program and ALMS’ Lakewatch Program. Samples collected from lake monitoring programs in 2005 were analyzed for specific MCYST analogues including MCLR, MCYR and MCRR. This work was revisited in 2007 and a set of 44 samples were analyzed for 5 specific analogues including: MCLR, MCYR, MCRR, MCLF and MCLW. Analysis for ATX-a was also conducted on 66 water samples in 2005 and 1 0 additional samples collected in 2006. Research into the detection and quantification of BMAA in surface water samples was initiated in 2005 by AENV in partnership with the Alberta Peptide Institute (API) at the University of Alberta. Methods were developed to isolate and detect BMAA in routine surface water samples. Method development for the detection of BMAA continued in 2007 when a new research partnership was formed with The Alberta Centre for Toxicology, University of Calgary (ACFT). This report presents status and findings of the cyanotoxin monitoring program from 2005 - 2008. Alberta Environment Cyanotoxin Program Status Report 4 2.0 METHODS Euphotic (depth) integrated composite water samples were collected for chemical (total MCYST, MCYST analogue and ATX-a concentrations) and biological (phytoplankton species identification and enumeration) analyses. Euphotic integrated samples were, in most cases, collected from 1 0 sites on a lake/reservoir basin and combined, together, to form a composite water sample as specified in AENV (2006). Sites were selected to encompass both near- shallow near shoreline) and off-shore (open-water) areas and influences of embayment. Cyanobacteria are largely planktonic and can accumulate with wind and wave action resulting in significantly greater population density and toxin concentrations along leeward shorelines compared to those upwind (i.e., horizontal heterogeneity). Moreover, many cyanobacteria possess specialized gas vesicles that aid in regulating buoyancy allowing them to migrate vertically within the water column to depths of optimal light intensity and spectrum and nutrient concentrations. They produce excess gas vesicles during periods of wind-induced water-column mixing to counter the downward drag of water currents. When winds cease, cyanobacteria become over-buoyant and accumulate near the water’s surface producing what is commonly called a bloom and appearing as a visible paint-like scum. Buoyancy regulation and blooms result in significant vertical heterogeneity in cyanobacteria distribution. The euphotic (depth) integrated composite water samples accounts for vertical and horizontal spatial heterogeneity that may exist in a water body in an attempt to represent average whole-lake estimates of cyanobacteria density and toxin content. Note that this method of sampling does not determine peak cyanobacterial densities and toxin concentrations that occur near the surface and immediately adjacent to leeward shoreline areas; surface blooms may contain MCYSTs at concentrations 2 or even 3 orders of magnitude higher than open water estimates convey. During the 4 year period, samples were submitted to several laboratories depending on analyte of interest and laboratory capability. Sample handling and analytical methodology are detailed in Appendix I. Efforts were made to address aspects of quality assurance and control (QA/QC) pertaining primarily to sample analyses. Results are detailed in Appendix IV. Alberta Environment Cyanotoxin Program Status Report 5 3.0 RESULTS AND DISCUSSION 3.1 Mscrocystin Monitoring 2005 - 2008 2005 Results During the 2005 open water season (May through October), a total of 164 euphotic depth integrated composite samples were collected and analyzed for total MCYST concentration via PPI assay (detection limit of 0.1 pg MCLR eq./L). Of these samples, 140 were collected from 34 natural lakes and 24 were collected from 6 reservoirs across the province (Table 2). Waterbodies were chosen simply based on their inclusion in an existing monitoring program and not on their trophic status (Appendix II, Table Al) or the historical occurrence of blooms within them. Surface waters ranged in trophic status from unproductive and least likely to experience surface blooms of cyanobacteria (e.g., Gregg and Jarvis lakes; Upper and Lower Kananaskis reservoirs) to hypereutrophic, bloom-impacted lakes (e.g., Baptiste and Sturgeon lakes). Raw data are presented in Appendix V, Table A7. Microcystin was detected in 79 (48%) of the 164 samples (Table 2; Figure 2). The majority (80%) of these, contained MCYST at concentrations up to 0.5 pg/L and approximately 1% (1 sample) contained greater than 1.5 pg/L of MCYST - the current DWQ guideline for MCLR (Figure 2). Samples collected in 2005 never contained MCYST in excess of the proposed RWQ guideline of 20 pg/L (Table 2). These findings are consistent with previous studies that indicated the majority of surface waters contain low (i.e., up to 0.5 pg/L) concentrations of MCYST (Zurawell, 2002). Concentrations can be elevated during August and September and may exceed 10 pg/L (Kotak and Zurawell, 2006). Concentrations were less than 2 pg/L during the 2005 program (Appendix III, Figure Al). It is important to note these reported volumetric concentrations are from depth integrated water samples collected from the euphotic zone and not concentrated surface bloom samples. Grab samples from the surface during severe blooms can yield considerably higher toxin concentrations (i.e., > 5000 pg/L; Kotak and Zurawell, 2006). Table 2 Summary of euphotic integrated composite samples collected during the open-water (May through October) seasons 2005 through 2008. 2005 2006 2007 2008 Total # of samples collected 164 182 164 220 # (%) of samples with MCYST not detected 85 (52%) 85 (47%) 40 (24%) 94 (43%) # (%) of samples with MCYST detected (>0.1 ug/L) 79 (48%) 97 (53%) 124 (76%) 126 (57%) # (%) of samples over DW guideline (1 .5 ug/L) 1 (1%) 11 (11%) 39 (32%) 16(13%) # (%) of samples over proposed Rec. guideline (20 ug/L) 0 (0%) 0 (0%) 2 (2%) 0 (0%) Total # of lake samples 140 156 144 197 Total # of reservoir samples 24 26 20 23 # of lake basins sampled 34 42 43 45 # of reservoir basins sampled 6 7 8 7 Alberta Environment Cyanotoxin Program Status Report 6 □ Nondetectable □ MCYST up to 0.5 ug/L □ MCYST >1.5 ug/L □ MCYST >0.5 to 1 ug/L □ MCYST >1 to 1.5 ug/L Figure 2 Number of surface water samples from 2005 with nondetectable (85 samples; blue portion of pie chart) vs. detectable (79 samples; red portion of pie chart) levels of MCYST. The bar summarizes the number of samples with MCYST at various incremental concentrations. During the 2005 monitoring program, MCYST was detected on 1 occasion in 26 (76%) of the 34 lakes and in 4 (67%) of the 6 reservoirs. Toxin was not detected in Beauvais, Elkwater, Fishing, Fork, Gregg, Gregoire, Island, and Jarvis lakes or in Upper and Lower Kananaskis reservoirs in 2005. The fact that 75% of the waterbodies sampled during the 2005 open water season contained detectable levels of MCYST on at least 1 occasion (see Section 3.5 below, Figure 6) suggests the toxin is prevalent in Alberta’s surface waters. Microcystin was not detected in several eutrophic lakes (Fork, Gregoire and Island lakes) and one bloom-prone hypereutrophic lake (Fishing Lake) during this period (Appendix II, Table Al). Toxin was detected in a number of mesotrophic surface waters including Crimson, Dillberry, Frog, Gamier (North), Hilda, Miquelon, Whitefish and Wolf lakes and Spruce Coulee and Newell reservoirs (Appendix II, Table Al; Figure 6). 2006 Results A total of 182 composite samples were collected and analyzed for total MCYST concentration during the 2006 season. Of these samples, 156 were collected from 42 natural lakes and 26 from 7 reservoirs across the province (Table 2). As in 2005, these surface waters were chosen simply based on their inclusion in an existing monitoring program and not on their level of fertility (trophic status; Appendix II, Table A2) or the historical occurrence of blooms within them. The surface waters range in trophic status from oligotrophic (e.g., Gregg and Jarvis lakes; Upper and Lower Kananaskis and Gleniffer reservoirs) to hypereutrophic, bloom-impacted lakes (e.g., Baptiste, Steele and Sandy Lakes). Raw data are presented in Appendix V, Table A8. Microcystin was detected in 97 (53%) of the 182 samples (Figure 3). As in 2005, the majority (65%) of these, contained low MCYST concentrations (i.e., up to 0.5 pg/L); 15% contained greater than 0.5 and up to 1.0 pg/L of MCYST; 8% contained greater than 1.0 and up to 1.5 pg/L of MCYST; and approximately 1 1% (1 1 of the 97 samples) contained greater than 1.5 pg/L of MCYST. In general, these findings were consistent with those of 2005 and previous studies Alberta Environment Cyanotoxin Program Status Report 7 indicating the majority of surface waters contain low (i.e., < 0.5 jug/L) concentrations of MCYST, on a whole-lake average basis. Compared to the previous year, however, a greater percentage of samples collected in 2006 contained concentrations of MCYST in excess of the DWQ guideline of 1.5 pg/L (1 1% in 2006 compared to 1% in 2005). Also, peak MCYST concentrations were greater in 2006 than in the previous year, as 4 samples (2 from Cooking Lake and 1 each from Red Deer and Pigeon lakes) exceeded 2 jug/L (Appendix III, Figure A2). As in 2005, no samples contained MCYST in excess of the proposed RWQ guideline of 20 pg/L. Once again it is important to note these concentrations are from depth integrated, multiple-site, composite water samples and not from surface concentrated bloom samples (such as those regularly occurring along beaches), which yield considerably more toxin. □ Nondetectable □ MCYST up to 0.5 ug/L □ MCYST >0.5 to 1 ug/L □ MCYST >1 to 1 .5 ug/L □ MCYST >1 .5 ug/L Figure 3 Number of surface water samples from 2006 with nondetectable (85 samples; blue portion of pie chart) vs. detectable (97 samples; red portion of pie chart) levels of MCYST. The bar summarizes the number of samples with MCYST at various incremental concentrations. In 2006, MCYST was detected on at least 1 occasion in 34 (81%) of the 42 lakes and 3 (43%) of the 7 reservoirs. Toxin was not detected in Beauvais, Clear (Bams), Ethel, Frog, Moose, Sylvan, Tucker and Wolf lakes or Moonshine, Newell, Lower Kananaskis and Spmce Coulee reservoirs in 2006. Overall, 76% of waterbodies sampled in 2006 contained detectable MCYST on at least one occasion. This finding is consistent with the previous year (75% in 2005) and indicates that MCYST is prevalent in Alberta’s lakes and reservoirs. Toxin was detected in a number of oligotrophic (Gregg and Jarvis lakes and Upper Kananaskis and Gleniffer reservoirs) and mesotrophic (Beartrap, Crimson, Dillberry, Elkwater, Hilda, Miquelon and Muriel lakes) waterbodies, while not being detected in several eutrophic (Beauvais, Moose and Tucker lakes and Moonshine reservoir) and hypereutrophic, Clear (Bams) Lake, systems (Appendix II, Table A2; Figure 6). 2007 Results A total of 164 composite samples were collected and analyzed for total MCYST concentration in 2007 - 144 samples from 43 lakes and 20 from 8 reservoirs province-wide (Table 2). As in the previous 2 years of monitoring, the surface waters ranged in trophic status from oligotrophic to Alberta Environment Cyanotoxin Program Status Report 8 hypereutrophic (Appendix II, Table A3). MCYST was detected in 124 (76%) of the 164 samples (Figure 4). In contrast to the previous seasons, far fewer samples collected in 2007 contained low (i.e., up to 0.5 pg/L) concentrations of MCYST - 47% of samples in 2007 vs. 80% and 65% in 2005 and 2006, respectively. The percentages of samples containing moderately-low (i.e., greater than 0.5 and up to 1.0 pg/L) and moderate (greater than 1.0 and up to 1.5 pg/L) MCYST concentrations were 10% and 11%, respectively, and were similar to that observed in 2005 and 2006. The most striking distinctions in observed toxin levels in 2007 were the high proportion (30%) of MCYST-positive samples exceeding 1.5 pg/L of MCYST (the DWQ guideline) and the presence of 2 (sequential) samples from George Lake exceeding the newly proposed RWQ guideline of 20 pg/L. Concentrations in George Lake peaked on August 01, 2007 at nearly 46 pg/L making this the highest recorded whole lake composite MCYST levels over the 4 years of monitoring. A sequential sample taken from the lake on August 23, 2007 contained nearly 22 pg/L of MCYST and notably near the proposed RWQ guideline well into September (19.3 pg/L on September 18, 2007). □ Nondetectable □ MCYST up to 0.5 ug/L □ MCYST >0.5 to 1 ug/L □ MCYST >1 to 1 .5 ug/L □ MCYST >1 .5 to 20 ug/L □ MCYST >20 ug/L Figure 4 Number of surface water samples from 2007 with nondetectable (40 samples; blue portion of pie chart) vs. detectable (124 samples; red portion of pie chart) levels of MCYST. The bar summarizes the number of samples with MCYST at various incremental concentrations. Other lakes also appeared to have experienced severe blooms of toxic (MCYST-containing) cyanobacteria in 2007. A total of 15 samples collected from 9 lake basins (Baptiste North and South basins, Clear, Cooking, George, Saskatoon, Steele, Thunder and Winagami lakes) contained levels of MCYST near or in excess of 10 pg/L (Appendix III, Figure A3). Raw data are presented in Appendix V, Table A9. Toxin was detected on at least 1 occasion in all 43 (100%) of the lakes and 6 of the 8 reservoirs. Thus overall, 96% of waterbodies sampled in 2007 contained detectable MCYST on at least one occasion signifying an increase of about 20% in prevalence over previous years (Figure 6). Toxin was not detected in Newell Lake or Twin Valley reservoirs in 2007, but this may be an artifact of sampling frequency as sampling occurred on a single date for each. Given additional opportunity to collect samples over the entire open- water season, it is likely that toxin could have Alberta Environment Cyanotoxin Program Status Report 9 been detected in these two reservoirs bringing MCYST prevalence to 100% of the systems monitored. See section Year-to-Year Microcystin Variability below for further discussion. 2008 Results In 2008, a total of 220 samples were collected for total MCYST analysis - 197 samples from 45 lake basins and 23 samples from 7 reservoirs across the province (Table 2; Appendix II, Table A4). Toxin was detected in 126 (57%) of the 220 samples (Figure 5). Similar to the 2005 and 2006 seasons, the majority (69%) of these contained low MCYST concentrations (up to 0.5 pg/L); 1 1% contained greater than 0.5 and up to 1.0 pg/L of MCYST; 7% contained greater than 1.0 and up to 1.5 pg/L of MCYST. And like that observed in 2006, a portion (13%) of samples contained greater than 1.5 pg/L of MCYST (the DWQ guideline). Unlike 2007, no samples contained MCYST in excess of the proposed RWQ guideline of 20 pg/L - though several (3) from Oster and Sandy lakes exceeded 10 pg/L (Appendix III, Figure A4). Raw data are presented in Appendix V, Table A 10. □ Nondetectable □ MCYST up to 0.5 ug/L □ MCYST >0.5 to 1 ug/L □ MCYST >1 to 1 .5 ug/L □ MCYST >1 .5 ug/L Figure 5 Number of surface water samples from 2008 with nondetectable (94 samples; blue portion of pie chart) vs. detectable (126 samples; red portion of pie chart) levels of MCYST. The bar summarizes the number of samples with MCYST at various incremental concentrations. MCYST was detected on at least 1 occasion in 40 (89%) of the 45 lakes and 4 (57%) of the 7 reservoirs. Toxin was not detected in Beauvais, Elkwater, Gregg, Jarvis or Sylvan lakes, nor did it occur once in Upper and Lower Kananaskis and Reesor reservoirs in 2008. Overall, 85% of waterbodies sampled in 2008 contained detectable MCYST on at least one occasion - down 10% in prevalence from 2007, yet 10% greater than in both 2005 and 2006 (Figure 6). Also, MCYST was detected in Newell and Twin Valley reservoirs in 2008 - the only two locations with non- detectable levels in 2007. Microcystin Prevalence 2005 - 2008 Summary Data Microcystin was prevalent during the 4 years of monitoring. Toxin was detected on at least one sampling occasion in 75% (2005) to 96% (2007) of the lakes and reservoirs sampled in a given Alberta Environment Cyanotoxin Program Status Report 10 year (Figure 6); and was generally more prevalent in fertile (i.e., hypereutrophic and eutrophic) waterbodies than in less productive systems. However, toxin was common (i.e., occurring in > 50% of waterbodies) in mesotrophic waters in all 4 years and in oligotrophic systems in 2006 and 2007 (Figure 6). ■ Oligotrophic □ Mesotrophic □ Eutrophic □ Hypereutrophic ■ Overall * £ 100% § 90% o 70% 5 60% c © 50% O C © 40% (0 © 30% l. a. h 20% S2 a 10% 2 £ 0% 2005 2006 2007 2008 Year Figure 6 Prevalence of MCYST in lakes/reservoirs of varying trophic status from 2005-2008. *MCYST prevalence is the percentage of lakes/reservoirs containing detectable toxin (> 0.1 pg MCLReq./L) on at least one sampling occasion. Microcystin in Low Nutrient Lakes Oligotrophic and mesotrophic lakes generally do not support large populations of phytoplankton. As a result, surface blooms of cyanobacteria are rare in these environments. It is well established that less fertile, stratified lakes and reservoirs, experience periodic accumulations of cyanobacteria at depths several meters below the surface nearer the metalimnion (i.e., transition between the upper well-mixed and illuminated epilimnion and the deeper, isolated cold water hypolimnion). Metalimnetic blooms occur when cyanobacteria, notably Planktothrix (Oscillatoria) sp., congregate within a distinct 1 to 2 m thick stratum at depth. These so-called ‘metalimnetic ecostrategists’ possess phycobiliproteins - water soluble photosynthetic pigments (in addition to chlorophyll) that efficiently intercept the breadth of the light spectrum (400-700 nm). Some Planktothrix (particularly rubescens) perform photosynthesis at very low light and grow at light intensities 1-5% of surface irradiance. They are also able to effectively regulate buoyancy in order to remain at these depths of optimal light. Evidence from past field observations and other research in Alberta (Zurawell, unpublished) and elsewhere (e.g., in Finland by Lindholm and Meriluoto, 1991), indicate metalimnetic blooms occur in meso- and oligotrophic waters and often contain MCYST. This means even those lakes suffering little human impact, may occasionally become toxic. Sampling protocols employed for monitoring ensured water was collected from the entire euphotic zone. Toxin-producing metalimnetic cyanobacteria, if present, would have been collected and could account for the MCYST periodically contained within samples from meso- and oligotrophic lakes and reservoirs during the 4 years of study. Alberta Environment Cyanotoxin Program Status Report II Although not an initial objective of the monitoring program, provisions were made from 2006 onward to collect evidence indicating the presence of MCYST-producing metalimnetic cyanobacteria. Supersaturating (near 1 00%) oxygen conditions isolated at depth can denote the presence of photosynthetically active metalimnetic cyanobacteria. Field technicians were instructed to collect discrete samples for total MCYST analysis and phytoplankton identification should they encounter supersaturating conditions at depth. On one occasion supersaturating conditions were noted at 8-m depth (as determined by a Hydrolab oxygen sensor) in oligotrophic Jarvis Lake on September 4, 2006. A discrete sample collected from 8-m depth stratum contained 0.16 jug/L of MCYST. Unfortunately, a sample for the identification of phytoplankton was not collected and the assumption of the existence of metalimnetic cyanobacteria remains somewhat speculative. Notably, toxin was not detected (i.e., <0.1 jag MCLReq./L) in the euphoric integrated composite sample collected from Jarvis Lake on that date. The discrepancy could simply be due to dilution of MCYST within a discrete depth by toxin- free overlying water - all collected as part of the euphoric integrated sample protocol. □ May □ June □ July □ August □ September ■ October Year Figure 7 Prevalence of MCYST in surface water samples grouped by sampling month (2005-2008). *MCYST prevalence is the percentage of samples containing detectable toxin (> 0.1 pg MCLReq./L). Microcystin was detected in 48% (2005) to 76% (2007) of the samples collected in a given year (Table 2). As anticipated, MCYST was generally more prevalent during the months of August and September (Figure 7). Cyanobacterial blooms in Alberta lakes and reservoirs usually occur - and are more severe - from mid-summer through early fall as water column stability required for surface bloom formation is greatest during this period (Figure 7). Year-to-Year Microcystin Variability Our data suggest that the presence of MCYST in a given water body is highly variable year to year. For instance, toxin was not detected in Elkwater, Fishing, Gregg, Gregoire, Island, or Jarvis lakes or Upper Kananaskis reservoir in 2005, but was detected at least once in each during 2006. In contrast, Moose and Wolf lakes and three reservoirs, Moonshine, Newell and Spruce Coulee, contained MCYST at least once in 2005, but did not contain toxin in 2006. Alberta Environment Cyanotoxin Program Status Report 12 In some instances, low sampling frequency may explain year-to-year disparity in the presence/absence of MCYST, as only one sample may have been collected in a given year. Considering the dynamic nature of phytoplankton communities, a single sampling event over an open water season is not sufficient to capture the presence of toxin-producing cyanobacteria. This is especially true for less productive waters, as metalimnetic cyanobacteria reach high densities for short periods of time. This may have been the case for Gregg and Jarvis lakes, as both were only sampled once each in 2005 (MCYST not detected), yet did contain MCYST in 2006 when they were sampled four times each. Similarly, toxin was not detected in each single sample collected in either 2005 or 2006 from Beauvais Lake. However, MCYST was detected on 2 of 4 occasions in Beauvais Lake in 2007 - the first instance over the initial 3 sampling seasons when more than 2 samples were collected in a single season. This provides a clear example of the need to collect multiple samples over an open-water season in order to understand the true prevalence of MCYST in surface waters. Climate Change and Prevalence of Toxic Cyanobacteria The increase in toxin prevalence observed in 2007 is noteworthy (Figure 7) and raises questions as to what principal factors influence cyanobacteria growth and reproduction in Alberta’s lakes and reservoirs. Above-normal late-winter/spring climatic conditions and an early spring thaw could advance water column warming. This could cause early onset of water column stabilization and provide better conditions for cyanobacterial growth and surface accumulation. These conditions could allow cyanobacteria to dominate the phytoplankton community in June or July rather than later (Figure 7). Additionally, a protracted open-water season and higher degree of water column warming resulting from early thaw could potentially increase the duration and magnitude of cyanobacterial blooms. Historical surface water temperature data for lakes collected by AENV reveals only 1 1 instances out of 3895 recorded measurements (between 1980 through 2007), when water temperatures at lm depth reached or exceeded 25°C. Of these, 7 cases were recorded in 2007. The reproductive rate of cyanobacteria is dependent on water temperature. Maximum rate of reproduction for many species increases to an optimum temperature nearing 30°C. It is likely that above-normal water temperatures occurring in 2007 resulted in increased cyanobacterial abundance and hence greater MCYST prevalence (Figures 6 and 7) and higher peak concentrations (Figure 4; Appendix III, Figure A3). 3.2 Microcystin Analogue Study In addition to the 172 samples analyzed for total MCYST concentration in 2005, 87 of these were also analyzed for the concentration of specific MCYST analogues including MCLR, MCYR and MCRR via LC-MS/MS. Of the 3 toxin analogues, MCLR was the most prevalent being detected in 60 of the 87 samples (Table 3). The next most prevalent was MCYR and the least was MCRR being detected in 21 and 8 of the 87 samples, respectively. Notably, MCLR was present in all samples containing either MCRR or MCYR with the exception of one. Information on the occurrence of MCYST analogues (other than MCLR) in Canada’s surface waters was limited (prior to this study). For this reason, Health Canada (2002) only considered a maximum acceptable concentration of MCLR when developing its drinking water guideline, citing insufficient data on other MCYST analogues to suggest otherwise. Earlier research suggested MCLR is not the only MCYST produced by cyanobacteria in Alberta’s lakes and reservoirs. Bloom material collected from Little Beaver Lake, Alberta in 1991 contained myriad Alberta Environment Cyanotoxin Program Status Report 13 toxin analogues (in addition to MCLR) including: 1.5 pg/g of MCLL, < 20 ng/g of MCLV, MCLM, MCLF and MCLZ (Z being an unidentified amino acid) and undetermined concentrations of MCLA and MCFR (Boland et al ., 1993; Craig et al ., 1993). The analogues MCRR and MCLR occurred in a suite of northern Alberta boreal lakes from 1997-98 (Kotak and Zurawell, 2006). Over this period, MCRR was detected in only 3 of 38 phytoplankton samples, while MCLR was detected in 28 of these samples. In two instances however, the concentration of MCRR exceeded that of MCLR (Big Chief Lake: MCRR - 57 jag/g vs. MCLR = 27 pg/g; Cowper Lake: MCRR = 47 pg/g vs. MCLR = 20 pg/g). Similarly, an unidentified MCYST analogue, later determined to be MCLL, was found in concentrations greatly exceeding that of MCLR in phytoplankton collected from Formby Lake, Driedmeat Lake and a northern boreal lake (Kotak and Zurawell, 2006). In contrast, Murphy et al (2003) identified MCRR as the primary analogue in blooms collected from Hamilton Harbour of Lake Ontario, though MC YR and MCLR were also present. The results from 2005 (Table 3) are consistent with many of these previous findings as MCLR was the most prevalent MCYST analogue in samples collected. The low prevalence of MCRR (occurring in 9% of the samples) is comparable to that of the province’s northern boreal lakes (8% occurrence) determined earlier (Kotak and Zurawell, 2006). This is the first study to investigate the occurrence of MCYR in Alberta’s surface waters and the results are noteworthy as it was detected in nearly one quarter (24%) of the samples and occurred in the absence of MCLR. Table 3 Summary of surface water samples with MCYST analogues detected in 2005 by ARCV. MCYST Analogue Number of Samples with Detected Analogue % of Samples with Detected Analogue MCLR 60 69% MCYR 21 24% MCRR 8 9% A second round of MCYST analogue testing was conducted in 2007 and included MCLF and MCLW in addition to MCLR, MCRR and MCYR. Of the 44 samples tested, 36 (82%) contained MCLR, 6 (14%) contained MCRR, and 2 (5%) contained MCYR (Table 4). No samples contained detectable levels of MCLW or MCLF. Table 4 Summary of surface water samples with MCYST analogues detected in 2007 by ACFT. MCYST Analogue Number of Samples with Detected Analogue % of Samples with Detected Analogue MCLR 36 82% MCYR 2 5% MCRR 6 14% MCLF 0 0% MCLW 0 0% Microcystin-LR was the most prevalent MCYST analogue in Alberta’s surface waters; however, other analogues occur simultaneously and in some cases MCLR may not be the dominant toxin. Occasionally, MCYST analogues may occur in the absence of MCLR. Note that we currently possess very limited capacity to detect and quantify all known toxic analogues, as pure, certified MCYST standards required for analysis by LC-MS/MS or similar instrument-based analytical methods (e.g., HPLC; GC) do not exist in sufficient quantity. Thus, MCYSTs not included in the limited suite of analogues analyzed by ARCV in 2005 (3 analogues) and ACFT in 2007 (5 Alberta Environment Cyanotoxin Program Status Report 14 analogues) could have been present in samples. Comparing results from the analogue analyses conducted by ACFT in 2007 (suite of 5 MCYST analogues) with split-sample results from the PPI assay support this premise. Five of 8 (63%) samples deemed negative or ‘non-detected’ by the 5-analogue LC-MS/MS scan (12% of the 44 samples overall) did contain a detectable enzyme inhibition response by PPI assay. It is important to consider that MCYST analogues differ widely in terms of toxicity. Ultimately, MCYSTs cause cell damage and death by binding to, and inhibiting PPI and PP2A enzymes. These enzymes regulate intracellular signal transduction pathways responsible for a multitude of cell functions including cell division, cell-to-cell signaling and cell metabolism (reviewed in Zurawell et al , 2005). Binding affinity for PPI and PP2A enzymes varies with MCYST analogue and while influenced by the properties of constituent amino acids, inhibition is largely dependent on the integrity of Adda and D-Glu (Figure 1). In a comparative PPI assay, researchers at the ACFT demonstrated the difference in the relative ability of MCLR, MCYR, MCRR, MCLF, MCLW, MCLA and MCLY to inhibit PPI. In this study, MCLF was the most potent inhibitor of PPI with an IC50 (concentration of toxin resulting in 50% reduction in PPI enzyme activity) = 0.12 nMol, followed by MCLR, MCLA, MCLW, MCLY, MCRR and MCYR (Table 5). Mean inhibition - relative to MCLR (Table 5) - demonstrates several toxin analogues to be of similar inhibitory potency for PPI (i.e., MCLA, MCLW and MCLY - values close to 1) and others to be less so (i.e., MCRR and MCYR - values much greater than 1). It is important to note that relative inhibitory potency of various MCYST congeners to PP2A may differ from that to PPL Monks et al (2007) showed relative potency to PPI as MCLF>MCLR>MCLW>MCRR>MCYR and to PP2A as MCLR>MCLW>MCLF>MCYR>MCRR. Table 5 IC50 of PPI by 7 MCYST analogues and mean inhibition relative to MCLR (ACFT unpublished data). Values for mean inhibition greater than 1 indicate less toxic than MCLR; values less than 1 indicate more toxic than MCLR. MCYST Analogue PPI IC50 (nMol) Mean Inhibition Relative to MCLR MCLR 0.14 1 MCYR 0.41 2.95 MCRR 0.34 2.43 MCLF 0.12 0.89 MCLW 0.21 1.5 MCLA 0.18 1.25 MCLY 0.23 1.65 Critical to toxicity is the ability of MCYSTs to cross (biological) cell membranes. Though constituent amino acids include both polar (hydrophilic) and non-polar (hydrophobic) residues, all MCYSTs are, albeit to varying degrees, soluble in water (de Maagd et al ., 1999). Unlike lipophilic (fat-soluble) substances, the passive uptake of MCYSTs by cells/tissues is limited. Active uptake of MCYSTs into cells is required for toxicity and is mediated by organic anion transporter polypeptides (OATPs). Several OATPs - OATP1B1 and OATP1B3 - capable of mediating MCYST uptake are specifically expressed in hepatocytes (liver cells) for which bile acid salts among others (e.g., cholate and taurocholate) are the “natural” substrates (Dietrich and Hoeger, 2005; Konig et al , 2006). Yet another - OATP1 A2 - is expressed in a variety of cells including liver cholangiocytes (bile duct epithelial cells) and renal (kidney) and intestinal epithelial cells, but predominantly in blood capillary endothelium of the brain (Fischer et al, 2005; Lee et al, 2005). Alberta Environment Cyanotoxin Program Status Report 15 The occurrence, type and level of expression (i.e., number of trans-membrane proteins per cell) of OATPs may largely account for the hepato- (liver) toxic nature of MCYSTs. It also corroborates recent evidence indicating the toxins can cross the blood-brain and blood- cerebrospinal fluid barriers and explains observed acute neurotoxicity in those fatally exposed to MCYST in Caruaru, Brazil in 1996 (Dietrich and Hoeger, 2005; Pouria et al 1998). Here too, differences in constituent amino acids greatly influences the affinity of MCYSTs to OATPs and hence variation in analogue uptake and resulting toxicity. Those containing polar amino acids at variable positions 2 and 4 (Figure 1) are generally less toxic than those with one or more non-polar amino acids. For example, replacement of leucine in position 2 with another non-polar amino acid (e.g., alanine, phenylalanine or tryptophan) maintains toxicity, but substitution with a polar amino acid (e.g., arginine) dramatically reduces it (Stotts et al., 1993). MCYSTs with polar amino acids in both positions, such as MCRR (arginine, arginine) and MCM(0)R (methionine sulfoxide, arginine), are the least toxic (Zurawell, 2001). Those with only non-polar amino acids in position 2 and 4 are far more toxic. Monks et al. (2007) recently demonstrated differences in growth inhibition of OATP1B1 and OATP1B3 transfected cell lines to 5 MCYST analogues. In both instances, the non-polar congeners MCLW and MCLF were more cytotoxic than either MCLR or MCYR, which contain non-polar (leucine or tyrosine, respectively) and polar (arginine) substitutions. MCRR was by far the least cytotoxic (Table 6). Table 6 Growth inhibition of OATP transfected cells lines by MCYST analogues (adapted from Monks et al. 2007). MCYST Analogue OATB1B1 IC50 (nMol/L) OATB1B3 IC50 (nMol/L) MCLW 0.3 ±0.1 0.5 ±0.4 MCLF 0.4 ±0.1 0.9 ±0.9 MCLR 5 ±51 39 ±8 MCYR 90 ±20 45 ±30 MCRR 3,800 ± 2,300 580 ± 400 Moreover, a genetically modified cell line expressing OATP1B3 seeded on plates incorporating microelectronic sensor arrays allowed real-time monitoring of dynamic responses (i.e., proliferation, apoptosis and morphology change) to MCYSTs by analyzing changes in impedance measurements (Huang et al.. Accepted). Comparative toxicity studies based on this real-time cell electronic sensing system (RT-CES) indicate differences in cytotoxicity of 7 MCYST analogues (Table 7). The most hydrophobic of congeners (MCLF, MCLA, MCLY, and MCLW) were more toxic than MCLR, while MCYR and hydrophilic MCRR were less so. Table 7 RT-CES based cytotoxicity of OATP1B3 cells lines by MCYSTs (ACFT unpublished data). MCYST Analogue Mean Cytotoxicity IC50 (nMol) MCLW 0.30 MCLF 0.13 MCLR 1.00 MCYR 2.59 MCRR 7.76 MCLA 0.19 MCLY 0.19 Alberta Environment Cyanotoxin Program Status Report 16 3.3 Anatoxin-a Study A total of 66 samples were collected in 2005 and analyzed for ATX-a concentration. Of these, 55 were collected from 29 natural lakes and 1 1 were collected from 6 reservoirs across the province (Appendix II, Table A5). Anatoxin-a was detected in 7 (11%) of the samples collected in 2005 (Figure 8). Four of these contained ATX-a concentrations equal to the limit of detection, and the highest concentration was 0.5 pg/L. During this monitoring program, ATX-a was detected in Hilda, McLeod, Saskatoon, Skeleton (South Basin), Wabamun (West Basin) and Whitefish lakes and in Moonshine Reservoir. Little data exist on the occurrence of ATX-a in Alberta’s surface waters. In part, this may be due to the fact that few laboratories have the capacity to analyze for the toxin. Additionally, compared to MCYSTs, ATX-a is relatively unstable and degrades rapidly making sample collection and preparation difficult. Yet, ATX-a toxicity has been implicated in livestock and wildlife mortalities on several occasions throughout the province. In these historical cases, identification of ATX-a producing cyanobacteria from water samples along with symptoms and gross pathology of affected animals, has led to presumptive conclusions. The low occurrence and concentrations of ATX-a in samples collected in 2005 is not surprising, as the species of cyanobacteria that commonly produce the toxin rarely dominate the phytoplankton communities of Alberta’s bloom-prone lakes. These species comprise only a small percentage of the phytoplankton community. In 2006, a number of lakes and reservoirs experienced blooms of cyanobacteria (primarily Anabaena sp.) that could produce ATX-a. However, none of the (10) samples collected in 2006 contained detectable levels of ATX-a. □ Nondetectable □ ATX-a at 0.02 ug/L (Detection Limit) □ ATX-a >0.02 to 0.1 ug/L □ ATX-a >0.1 to 0.5 ug/L ^ 4 59 7 ^ 1 ^2 Figure 8 Number of surface water samples with nondetectable (59 samples; blue portion of pie chart) vs. detectable (7 samples; red portion of pie chart) levels of ATX-a in 2005. The bar summarizes the number of samples with ATX-a at various incremental concentrations. Alberta Environment Cyanotoxin Program Status Report 17 3,4 p-AA-Methylamino-L-Alanine (BMAA) Study Unlike the other studies presented above, a large component of the BMAA study in 2005 was devoted to developing methods for the isolation, identification and quantification of the compound from surface water samples (this constitutes Phase 1 of BMAA research). The first consideration was to confirm that a standard (reference) solution of BMAA could be detected by standard amino acid analysis (AAA) and LC-MS/MS. A standard solution of BMAA was detected by AAA at a detection limit of 0.02 pMol. The standard could also be detected and molecular weight determined using matrix-assisted laser desorption/ionization time-of-flight mass spectroscopy (MALDI ToF-MS). The next step was to analyze water samples collected from lakes with obvious cyanobacteria populations. In the first attempt (Prep A), a 100-ml water sample was lyophilized (freeze-dried) yielding about 10 mg of solid material. Results of AAA of Prep A indicated the total amino acid content was low and BMAA was not detectable. Next (Prep B), a 1 L water sample was filtered through cellulose fiber filter paper (Whatman #1, 1 1-pm pore size) yielding about 8 mg of solid material. Total amino acid content of Prep B was also low and BMAA could not be detected. At this point we acknowledged the need to further concentrate fresh water samples prior to analyses. Sample preparation methodology was altered to include filtering known volumes of freshly-collected water samples through 1.6-pm pore size glass fiber filters (Whatman GF/A) in the same fashion as preparing samples for the determination of chlorophyll-a. Four samples (one each from Baptiste Lake South Basin, Sturgeon Lake, Winagami Lake and Moonshine Reservoir) were analyzed following this procedure. Results of AAA indicated that a peak (compound) could be detected in these concentrated samples at a retention time consistent with that of pure BMAA standard (about 47 min elution time). Co-injection of a BMAA standard with sample was taken as further evidence the peak in the chromatogram was BMAA. Since these samples were not subjected to mass spectrometry analysis, the identity of the compound (peak) was not confirmed unequivocally and the presence of BMAA remains presumptive at this point. Concentrations of presumptive BMAA were determined by correlating peak area with that of a set of serially diluted BMAA standards. The concentrations of presumptive BMAA in the 4 samples were 41, 27, 23 and 12 pg per injection volume for Baptiste Lake South Basin, Sturgeon Lake, Winagami Lake and Moonshine Reservoir, respectively, equaling BMAA concentrations of 12, 8, 7 and 4 pg/sample or 48, 32, 28 and 16 pg/L lake water. Without exception, cyanobacteria were the most abundant of the phytoplankton groups in all 4 samples based on percent abundance (Table 8). In terms of absolute abundance (i.e., number of cells per L), Baptiste Lake South Basin had the greatest density of cyanobacteria. Its community was dominated by the cyanobacteria, Aphanizomenon flos-aquae and Coelosphaerium naegelianum. Sturgeon Lake contained the 2nd highest density of cyanobacteria, which comprised greater than 90% of the overall phytoplankton community. The dominant cyanobacterium in Sturgeon Lake was Aphanizomenon flos-aquae , but Oscillatoria sp., Chroococcus dispersus and an unidentified cyanophyte were abundant. Cyanobacteria density within Winagami Lake and Moonshine Reservoir was lower than in either Baptiste or Sturgeon lakes (Table 8), however species diversity was greater. Coelosphaerium naegelianum dominated the phytoplankton community within Moonshine Reservoir, but Aphanizomenon flos-aquae , Anabaena circinalis , Chroococcus dispersus and an unidentified species were abundant as well. Winagami Lake contained the most diverse assemblage of cyanobacteria including Aphanizomenon flos-aquae , Anabaena flos-aquae , Microcystis incerta , Oscillatoria sp ., Aphanocapsa delicatissima , Chroococcus dispersus , and 3 additional unidentified cyanophytes. Alberta Environment Cyanotoxin Program Status Report 18 Table 8 Summary of cyanobacteria abundance, relative abundance and biomass in water samples analyzed for BMAA in 2005. Location Abundance of Cyanobacteria (# cells/L) % Relative Abundance of Cyanobacteria Biomass of Cyanobacteria (ug/L) Winagami Lake 5,885,942 79% 20619 Moonshine Lake Reservoir 4,183,155 72% 753 Sturgeon Lake 13,517,320 90% 11744 Baptiste Lake South Basin 19,422,960 75% 16590 Though Baptiste and Sturgeon lakes had the greatest abundance of cyanobacteria of the 4 samples, Winagami Lake had the greatest biomass of cyanobacteria per L of sample (Table 8). With the exception of Moonshine Reservoir (15.4%), cyanobacteria comprised greater than 50% of the relative biomass in all of the samples (Figure 9). This was most evident in Baptiste Lake where 96.4% of the biomass was attributed to cyanobacteria. Winagami Lake Moonshine Lake Reservoir . — ^ 21.8% / □ Cyanobacteria ( flA/ ■».4% | Diatoms □ Chrysophytes 33.0% □ Cyanobacteria ■ Diatoms ■ Chlorophytes ■ Cryptophytes □ Chrysophytes □ Pyrrophytes Sturgeon Lake Baptiste Lake South Basin 54.1% □ Cyanobacteria ■ Diatoms ■ Chlorophytes ■ Cryptophytes □ Chrysophytes □ Pyrrophytes □ Cyanobacteria ■ Cryptophytes □ Chrysophytes Figure 9 Pie charts depicting the relative biomass of phytoplankton groups within samples collected for BMAA analysis from Winagami Lake, Moonshine Lake Reservoir, Sturgeon Lake and Baptiste Lake South Basin. The bar presents the data for phytoplankton groups comprising less than 10% or the overall biomass. The overwhelming prevalence of cyanobacteria has implications for the occurrence and concentration of BMAA in these lake water samples. Though data are limited, it appears the ambient concentration of BMAA may be highly correlated with the abundance and biomass of cyanobacteria. If future studies confirm the identity of the compound presumed to be BMAA, these findings will support recent studies by Cox et al. (2005) that suggested most species of cyanobacteria (90% of the species studied) produce BMAA. Method development for the analysis of BMAA continued in 2007 in collaboration with the ACFT. An LC-MS/MS method to identify and quantify a fluorescent derivative of BMAA following reverse-phase separation on a Cig column was established (see Appendix I - Alberta Environment Cyanotoxin Program Status Report 19 Laboratory Methods). BMAA was successfully identified and quantified in hydrolyzed cycad leaf by this method, which served as a positive biological reference material. Hydrolyzed cyanobacteria and fish filet tissue samples were spiked with BMAA (5 - 1000 pg/g) to validate the method. Method recovery was 85 - 115% with a relative standard deviation of 6 - 10%. In addition, ACFT compared results of derivatized BMAA with a modified LC-MS/MS method employing Hydrophilic Interaction LC (HILIC) normal-phase column without derivatization (Kubo et al, 2008; Rosen and Hellenas, 2008). Kubo et al. (2008) argue HILIC is more effective than traditional reverse-phase (Cig) columns for separating BMAA from other closely related amino acids due to an increased efficiency that results from the rapid evaporation of organic solvent during electrospray ionization. The 2 methods were in agreement. With these new methods established, ACFT is continuing testing of archived cyanobacteria samples in 2009 to confirm if the putative toxin exists in our surface waters. Testing of fish samples will also be conducted in the future. Data will be presented at a later time. Alberta Environment Cyanotoxin Program Status Report 20 4.0 GENERAL DISCUSSION Cyanotoxin monitoring was incorporated into Alberta’s Lake and Reservoir Monitoring Program in 2005 with the goal of determining the prevalence of MCYST in Alberta. In addition, funding was obtained to determine the occurrence and concentrations of ATX-a and several MCYST analogues in many of these lakes and reservoirs. MCYST was very prevalent over the period of study (2005 - 2008), occurring on at least one sampling occasion in 75% (2005) to 96% (2007) of lakes and reservoirs in a given year. In terms of samples overall, toxin was detected in 48% (2005) to 76% (2007) of those collected in any given open-water period. High prevalence of MCYST noted in 2007 is of concern and indicates that all Alberta lakes and reservoirs occasionally experience some degree of toxin-producing cyanobacteria regardless of trophic status. Climate could be an important factor in determining the abundance of cyanobacteria in our lakes and reservoirs. Warm surface water temperatures in 2007 likely contributed to higher cyanobacterial growth in all surface waters regardless of nutrient conditions. The period of onset and severity of cyanobacterial blooms may be, in part, determined by water temperature, as they occurred earlier (early July as opposed to August) and were more extensive in area and duration in many productive (eutrophic and hypereutrophic) systems than usual. Increasing air temperatures could result in earlier ice-off and lengthier ice-free seasons. Protracted periods of water column stability combined with warmer overall water temperatures would undoubtedly favor growth and reproduction of cyanobacteria and likely result in greater frequency, duration and severity of potentially toxic blooms. Though variable year to year, whole-lake concentrations of MCYST appear to be low (generally < 1.5 pg/L total MCYST). The highest concentrations were recorded in 2007; in that year, a greater proportion of samples overall, contained both detectable MCYST and elevated toxin levels (> 1.5 pg/L total MCYST) than in any other year studied. And in two instances, concentrations exceeded 20 pg/L total MCYST - the newly proposed recreational guideline value for Canada. Here again, factors controlling cyanobacterial growth, reproduction, community composition and toxin production, such as water temperature and nutrient availability, may principally influence MCYST concentrations. It is important to note these concentrations are based on depth integrated water samples collected from the surface to the bottom of the euphotic zone. While this provides an indication of the average whole-lake MCYST concentrations, it does not adequately assess peak concentrations that may occur in areas of densely accumulated cyanobacteria. Surface-grab samples (i.e., uppermost 10 cm of a water body) of severe blooms in embayment and near shoreline areas often yield much higher toxin levels. Notably, concentrations as high as 15,000 pg/L total MCYST - four to five orders of magnitude higher than typical whole-lake concentrations - have been documented in concentrated blooms samples from hypereutrophic recreational lakes. To determine risk of toxic blooms to recreational users, frequent (weekly) surface sampling of near- shore areas including public beaches and swimming areas would be more appropriate, as is conducted by Alberta Health Services. It is evident that MCYSTs do occur in less productive waters corroborating evidence that many of Alberta’s oligo- and mesotrophic lakes and reservoirs experience periodic blooms of metalimnetic cyanobacteria (e.g., Planktothrix sp.). In the one instance with sufficient evidence of a metalimnetic bloom, a discrete sample taken at depth contained MCYST. Metalimnetic Alberta Environment Cyanotoxin Program Status Report 21 blooms are sporadic in nature - they may be brief and yet re-occurring events in a given water body throughout the open water season. This implies site (lake/reservoir) selection based solely on nutrient status or chlorophyll-a concentration (i.e., oligo-, meso-, eu- and hypereutrophic) may be inappropriate when designing monitoring programs to assess the risk of MCYST exposure, as toxin occurs in all lentic (i.e., still or slow moving waters) systems regardless of trophic classification. Timing and frequency of monitoring is also an important consideration in terms of being able to observe the dynamic nature of toxin-producing cyanobacteria and detect their toxins. A single sample collected over a season is insufficient to draw conclusions or evaluate risk regarding the prevalence of cyanotoxin in Alberta’s lakes and reservoirs. Other factors including consumptive use by humans (e.g., raw source for drinking water) and livestock (watering) and degree of direct (public beaches for swimming) and indirect (boating and fishing) recreational exposure need to be considered. The prevalence of MCYST in oligo- and mesotrophic lakes and reservoirs suggests all drinking water treatment facilities drawing water from lentic (relatively still waters: lakes, ponds, reservoirs and irrigation canals) systems - regardless of trophic status - may risk toxin exposure. This includes utilities withdrawing and impounding water from lotic systems (i.e., fast-flowing streams and rivers) within reservoirs; conditions in storage reservoirs can be conducive for cyanobacteria growth. This also implies that no untreated water from these sources should be directly consumed or used for cooking (MCYSTs are heat stable and are not destroyed with boiling) or bathing (inhalation of water mist or spray is a recognized route of MCYST exposure). Results from the MCYST analogue studies support the premise that MCLR is a common toxin in Alberta’s surface waters. It is apparent that other analogues, such as MCRR and MCYR, may also be prevalent and can occur in the absence of MCLR or at concentrations exceeding that of MCLR. Recognizing we have a limited capacity to measure all of the known toxic congeners, it is not unrealistic to presume others occur in our lakes and reservoirs as well. The various toxin analogues differ in terms of toxicity as a result of inherent properties of individual constituent amino acids. Unequal inhibition of PP1 and PP2A by various MCYSTs only partly account for the disparities in overall toxicity. In addition, the active uptake of toxin by MCYST specific OATPs is required for toxicity. Binding and uptake rates of MCYSTs vary with analogue and specific OATP with the most hydrophobic of congeners (MCLF, MCLA, MCLY, and MCLW) being generally more toxic than MCLR and hydrophilic forms (MCYR and MCRR). These results call into question the current drinking water guideline based solely on the concentration of MCLR. It is evident that further consideration be given to other, potentially more toxic, MCYST analogues if a guideline is to be protective against this group of toxins. A guideline that considers cumulative toxicity of MCYSTs as representing the potential threat to human health via drinking water - perhaps based on total MCLR toxicity equivalents (as adopted in Australia) - appears more appropriate for use in Alberta. The neurotoxin, ATX-a, occurred infrequently and at low concentrations in Alberta’s lakes and reservoirs monitored in 2005. During the summer of 2006, blooms were uncharacteristically dominated by potentially neurotoxic species ( Anabaena sp.) and several water samples were collected for ATX-a analysis. However, none of these samples contained ATX-a at or above the analytical limit of detection. It appears the risk posed by ATX-a is much lower than that of hepatotoxic MCYSTs. Testing for ATX-a need only occur if circumstances dictate, such as wildlife or livestock mortality. Research into the detection and quantification of BMAA was initiated in 2005. The majority of effort during the 4 year period was devoted to developing methods of sample collection, Alberta Environment Cyanotoxin Program Status Report 22 preparation and analysis. Analysis of several samples indicates that BMAA may occur at detectable concentrations in Alberta’s lakes and reservoirs. Though data are limited, it appears that the ambient concentration of presumptive BMAA may be highly correlated with the abundance and biomass of cyanobacteria. Given that some research indicates most cyanobacteria can produce BMAA; this putative toxin may be more prevalent than other cyanotoxins in Alberta. Research on BMAA is continuing with support of the ACFT. If BMAA is confirmed in our lakes and research shows it to be a causative agent of ALS/PDC or other neurodegenerative diseases, the risk to human health will need to be assessed. 5.0 CONCLUSIONS • Microcystins occur in a high percentage of Alberta lakes and reservoirs. Several analogues are present. • Microcystins are found in oligo- and mesotrophic lakes, indicating that not only eutrophic/hypertrophic lakes with visible blooms contain toxins. • Average concentrations can periodically exceed 20 pg/L of total microcystin - the draft RWQ guideline value for surface waters. It is recognized that toxin concentrations in near-shore bloom accumulations can be significantly higher than levels in open-water. • Climate may influence MCYST prevalence and concentrations. • Anatoxin-^ is infrequent and at low concentrations and appears to pose less risk than MCYST. • BMAA analytical methods are under development. Risk to human health needs further assessment. Alberta Environment Cyanotoxin Program Status Report 23 6.0 REFERENCES Alberta Environment. 2006. Aquatic Ecosystems Field Sampling Protocols. Publication No. W0605, Edmonton. 149 pp. An, J.S. and W.W. Carmichael. 1994. Use of a colorimetric protein phosphatase inhibition assay and enzyme linked immunosorbent assay for the study of microcystins and nodularins. Toxicon, 32, 1495-1507. Boland, M.P., M.A. Smillie, D.Z. Chen and C.F.B. Holmes. 1993. A unified bioscreen for the detection of diarrhetic shellfish toxins and microcystins in marine and freshwater environments. Toxicon, 31:1393-405. Cox, P.A., S.A. Banack, S.J. Murch, U. Rasmussen, G. Tien, R.R. Bidigare, J.S. Metcalf, L.F. Morrison, G.A. Codd and B. Bergman. 2005. Diverse taxa of cyanobacteria produce b-N- methylamino-L-alanine, a neurotoxic amino acid. Proc. Natl. Acad. Sci., 102:5074-5078. Craig, M., T.L. McCready, H.A. Luu, M.A. Smillie, P. Dubord and C.F.B. Holmes. 1993. Identification and characterization of hydrophobic microcystins in Canadian freshwater cyanobacteria. Toxicon, 31:1541-1549. de Maagd, P.G.-J., A.J. Hendriks, W. Seinen and D.T.H.M Sijm. 1999. pH-Dependent hydrophobicity of the cyanobacteria toxin microcystin-LR. Water Res., 33: 677-680. Dietrich, D., S. Hoeger, 2005. Guidance values for microcystins in water and cyanobacterial supplement products (blue-green algal supplements): A reasonable or misguided approach?. Toxicol. Appl. Pharmacol., 203: 273-289. Dunn, J. 1996. Algae kills dialysis patients in Brazil. Brit. Med. J., 312:1 183-1 184. Fawell, J.K., C.P. James and H.A. James. 1994. Toxins from blue-green algae: toxicological assessment of microcystin-LR and a method for its determination in water. Report No. FR 0359/2/DoE 3358/2, Water Res. Centre, Medmenham, UK. 46 pp. Fischer, W.J., S. Altheimer, V. Cattori, P.J. Meier, D.R. Dietrich, and B. Hagenbuch. 2005. Organic anion transporting polypeptides expressed in liver and brain mediate uptake of microcystin. Toxicol. Appl. Pharmacol., 203: 257-63. Gurney, S. and G. Jones. 1997. Toxic algae occurrence and distribution in Manitoba surface waters. In: Proceedings of Rural Water Symposium. Canadian Water Resources Assoc. Winnipeg, Manitoba. 97-11 1 . Health Canada. 2002. Cyanobacterial Toxins - Microcystin-LR. Guidelines for Canadian Drinking Water Quality: Supporting Documentation. Federal - Provincial - Territorial Subcommittee on Drinking Water, Ottawa, Canada. 22 pp. Huang, Y., M. Mock, B. Hagenbuch, S. Chan, J. Dmitrovic, S. Gabos and D. Kinniburgh. (Accepted) Dynamic cytotoxic response to microcystins using microelectronic sensor arrays. Env. Sci. Technol. Humpage, A.R., S.J. Hardy, E.J. Moore, S.M. Froscio and I.R. Falconer. 2000. Microcystins (cyanobacterial toxins) in drinking water enhance the growth of aberrant crypt foci in the mouse colon. J. Toxicol. Environ. Health Part A. 61:155-165. Konig, J., A. Seithel, U. Gradhand,and M.F. Fromm. 2006. Pharmacogenomics of human OATP transporters. J. Naunyn-Schmiedeberg's Arch. Pharmacol., 372: 432-443. Alberta Environment Cyanotoxin Program Status Report 24 Kotak, B.G. and R.W. Zurawell. 2006. Cyanobacterial Toxins in Canadian Waters. LakeLine, 26:24-28. Kubo, T., N. Kato, K. Hosoya and K. Kaya. 2008. Effective determination method for a cyanobacterial neurotoxin, p-A-methylamino-L-alanine. Toxicon, 51:1264-1268. Kurland, L.T. and D.W. Mulder. 1954. Epidemiologic investigations of ALS. I. Preliminary reports on geographic distribution, with special reference to Mariana Islands, including clinical and pathologic observations. Neurology, 4: 355-378. Lee, W., H. Glaeser, L.H. Smith, R.L. Roberts, G.W. Moeckel, G. Gervasini, B.F. Leake and R.B. Kim. 2005. Polymorphisms in human organic anion-transporting polypeptide 1 A2 (OATP1A2): implications for altered drug disposition and central nervous system drug entry. J. Biol. Chem., 280: 9610-9617. Lindholm, T. and J.A.O. Meriluoto. 1991. Recurrent depth maxima of the hepatotoxic cyanobacterium Oscillatoria agardhii. Can. J. Fish. Aquat. Sci., 48:1629-1634. MacKintosh, C., and S. Klumpp. 1990. Tautomycin from the bacterium Streptomyces verticillatus. FEBS Lett., 277:137-140. Murch, S.J., P.A. Cox, S.A. Banack, J.C. Steele and O.W. Sacks. 2004. Occurrence of p- methylamino-l-alanine (BMAA) in ALS/PDC patients from Guam. Acta Neurologica Scandinavica, 110: 267-269. Murch, S.J., P.A. Cox and S.A. Banack. 2004b. A mechanism for slow release of biomagnified cyanobacterial neurotoxins and neurodegenerative disease in Guam. Proc. Natl. Acad. Sci., 101:12228-12231. Murphy, T.P., K. Irvine, J. Guo, J. Davies, H. Murkin, M. Charlton and S.B. Watson. 2003. New microcystin concerns in the Lower Great Lakes. Wat. Qual. Res. J. Can., 38:127-140. Nicholson, B.C. and M.D. Burch. 2001. Evaluation of analytical methods for detection and quantification of cyanotoxins in relation to Australian drinking water guidelines. National Health and Medical Research Council of Australia, the Water Services Association of Australia and the Cooperative Research Centre for Water Quality and Treatment, Canberra, Australia, http://www.nhmrc.gov.au/publications/synopses/_files/eh22.pdf. Accessed May 18, 2007. Nishiwaki-Matsushima, R., T. Ohta, S. Nishiwaki, M. Sugawama, K. Kohyama, T. Ishiwaki, W.W. Carmichael and H. Fujiki. 1992. Liver tumor promotion by the cyanobacterial cyclic peptide toxin microcystin-LR. J. Cancer Res. Clin. Oncol. 1 18:420-424. Pouria, S., A. Deandrade, J. Barbosa, R.L. Cavalcanti, V.S.T. Barreto, C.J. Ward, W. Preiser, G.K. Poon, G.H. Neild and G.A. Codd. 1998. Fatal microcystin intoxication in haemodialysis unit in Caruaru, Brazil. Lancet, 352: 21-26. Rosen, J. and K.-E. Hellenas. 2008. Determination of the neurotoxin BMAA (p-A-methylamino- L-alanine) in cycad seed and cyanobacteria by LC-MS/MS (liquid chromatography tandem mass spectrometry). Analyst, 133: 1785-1789. Spencer, P.S., P.B. Nunn, J. Hugon, A.C. Ludolph, S.M. Ross, D.N. Roy and R.C. Robertson. 1987. Guam amyotrophic lateral sclerosis-Parkinsonism-dementia linked to a plant excitant neurotoxin. Science, 237:517-522. Alberta Environment Cyanotoxin Program Status Report 25 Stotts, R.R., M. Namikoshi, W.M. Haschek, K.L. Rinehart, W.W. Carmichael, A.M. Dahlem and V.R. Beasley. 1993. Structural modifications imparting reduced toxicity in microcystins from Microcystis spp. Toxicon, 31: 783-789. Zegura, B., B. Sedmak and M. Filipic. 2003. Microcystin-LR induces oxidative DNA damage in human hepatoma cell line HepG2. Toxicon, 41: 41-48. Zhou, L., H. Yu and K. Chen. 2002. Relationship between microcystin in drinking water and colorectal cancer. Biomed. Environ. Sci., 15:166-71. Zurawell, R.W. 2001. Occurrence and Toxicity of Microcystins in the Freshwater Pulmonate Snail Lymnaea stagnalis. Ph.D. Thesis, University of Alberta, Edmonton. 221 pp. Zurawell, R.W. 2002. An initial assessment of microcystin in raw and treated municipal drinking water derived from eutrophic surface waters in Alberta. Publication No. T672, Prepared for Alberta Environment, Science and Standards Branch, Edmonton. 37 pp. Zurawell, R.W., H. Chen, J.M. Burke and E.E. Prepas. 2005. Hepatotoxic cyanobacteria: A review of the biological importance of microcystins in freshwater environments. J. Toxicol. Environ. Health. Part B, 8:1-37. Alberta Environment Cyanotoxin Program Status Report 26 Appendix I Laboratory Methods Alberta Environment Cyanotoxin Program Status Report 27 Total Microcystin Analysis Immediately following collection, aliquots of euphotic integrated composite water samples were poured into pre-rinsed 20mL plastic scintillation vials, cooled, and then stored at -20°C until analyzed. Total MCYST concentrations were determined with a colorimetric protein phosphatase inhibition (PPI) assay as specified by An and Carmichael (1994). Analyses were conducted by ARC Vegreville. The PPI assay was chosen for this study over instrument-based analytical methods, such as high performance liquid chromatography (HPLC), fast atom bombardment mass spectrometry (FABMS) and GC-MS, because of its sensitivity to detect trace MCYST concentrations and its ability to estimate ‘total’ MCYST concentration by quantifying an overall inhibitory response of the protein phosphatase Type lc enzyme by bioactive MCYSTs. The PPI assay is based on the actual mode of MCYST toxicity, that is, the irreversible binding and subsequent inhibition of protein phosphatase type lc and 2A (PPI and PP2A). It quantifies all unbound (i.e., those not bound to endogenous PPI or PP2A), bioactive (i.e., MCYSTs capable of inhibiting PPI) toxin analogues, though differentiation between and identification of various congeners is not possible (Zurawell, 2002). Total MCYST concentrations in composite water samples were extrapolated from curves plotting PPI inhibition by MCLR standards and are thus expressed as mass (pg) of MCLR equivalents/L. The assay detection limit (i.e., lowest measureable level of MCYST inferred from the standards dose-inhibition response curve) is approximately 0.07 pg MCLR eq./L. However, test plate conditions (e.g., PPlc enzyme activity and assay buffer concentrations and pH) can influence the inhibition response of the assay and lower limits can range from 0.05 to 0.09 pg MCLR eq./L. Considering this, the lower limit of detection for all samples analyzed by PPI assay was set at 0.1 pg MCLR eq./L and lesser concentrations were regarded as ‘non-detected’. It is recognized that other natural compounds of cyanobacterial (i.e., nodularin and motuporin) and non-cyanobacterial origin (i.e., okadaic acid, calyculin-A and tautomycin) also inhibit PPI; it was assumed they were not present in our samples for the following reasons. Nodularins occur in brackish and marine species of cyanobacteria including Nodularia spumigena typically occurring in the Baltic Sea and brackish water estuaries and coastal lakes. Similarly, motuporin (also known as nodularin- V) originates from cyanobacterial symbionts associated with the tropical sponge Theonella swinhoei. Okadaic acid, a causative agent of diarrhetic shellfish poisoning, is a polyether fatty acid (polyketide) originating from unicellular, marine dinoflagellates primarily of the genus Prorocentrum. Calyculin-A is a phosphorylated polyketide also isolated from a marine sponge, Discodermia calyx. Thus, the presence of these PPI inhibitors is highly unlikely. In contrast, tautomycin, a polyketide structurally related to okadaic acid, is produced by the soil bacterium Streptomyces spiroverticillatus (MacKintosh and Klumpp, 1990). Theoretically, soil bacteria within the watershed could be a source of this PP 1 inhibitor. However, as Streptomyces is closely associated with rooted terrestrial vegetation, it is unlikely that tautomycin occurs in sufficient concentrations to influence PP inhibition in surface water samples. Microcystin Analogue Analysis Microcystin analogues (MCLR, MCYR and MCRR in 2005 by ARC Vegreville; MCLR, MCLF, MCLW, MCYR and MCRR in 2007 by ACFT) were separated and quantified by liquid Alberta Environment Cyanotoxin Program Status Report 28 chromatography linked tandem mass spectrometry (LC-MS/MS). Aliquots of euphotic integrated composite water samples were poured into pre-rinsed 500mL plastic bottles immediately following collection and stored at -20°C until analyzed. Following thawing in the laboratory, samples were sonicated to disrupt cyanobacterial cell membranes and liberate toxin. 2005 Samples Sonicated samples were passed through Waters Oasis® HLB (Hydrophilic-Lipophilic Balance reverse-phase sorbent) solid-phase extraction (SPE) cartridge. MCYSTs were eluted with 6 mL methanol then evaporated to a final 1-mL volume. Samples were separated on reverse-phase Cjg column linked to MS with electrospray ionization (ESI). Multiple reactions monitoring (MRM) transition was: MCRR 520 - 135; MCLR 995 - 135; and MCYR 1045 - 135. 2007 Samples Sonicated samples were filtered through PES (polyethersulfone) syringe-type filter cartridge prior to direct injection on an Agilent 1 100 LC without further sample preparation. An 80 pL sample volume was injected on a BDS Hypersil reverse-phase Cjg column (100 mm x 2.1mm, 5p particle size) and MCYSTs eluted with a gradient of 0.1% formic acid in de-ionized water and 0.1% formic acid in acetonitrile (mobile phase). The column was at 40°C with a flow rate of 0.3 mL/min. The tandem MS analyses were conducted with a Sciex API 4000 Triple Quadrupole Mass Spectrometer operated in ESI positive mode. MCYST identification and quantification was performed based on the analogue’s retention time and two MRM transition (in standard and unknown sample). Anatoxin-a Analysis Aliquots of euphotic integrated composite water samples were poured into pre-rinsed 500mL plastic bottles and stored at -20°C until analyzed by ARC Vegreville. Following thawing in the laboratory, samples were sonicated to disrupt cyanobacterial cell membranes and liberate toxin. Samples were passed through Waters Oasis® HLB SPE cartridge and eluted with 6mL methanol. These were evaporated to dryness and derivatized in TFAA (trifluoroacetic anhydride) with heating to 95°C for 1 hour. Samples were evaporated to remove excess TFAA and reconstituted in 100 pL of ethyl acetate. ATX-a was separated and quantified by GC-MS ion trap. (3-AAMethylamino-L-Alanine (BMAA) Methods Researchers at ACFT developed an LC-MS/MS-based method for determining BMAA. Reference standards of BMAA and hydrolyzed cycad leaf, cyanobacteria and fish samples were derivatized, pre-column, using Waters’ AccQ-Fluor™ reagent. Derivatized BMAA and an internal standard of a-aminobutyric acid were separated with an Agilent 1 100 Liquid Chromatograph fitted with a Hypersil BDS reverse-phase C18 column (100 mm x 2.1mm i.d., 5- pm particle size) under gradient conditions maintained at 40°C. BMAA and a-aminobutyric acid were identified and quantified using a Sciex API 4000 Triple Quadrupole Mass Spectrometer in MRM mode. MRM transitions were: 459/171 and 459/289 for BMAA and 274/171 for a- aminobutyric acid. Identification and quantification was performed based on the two MRM transitions combined with the retention time. The detection limit for total BMAA was 2 pg/g dry Alberta Environment Cyanotoxin Program Status Report 29 weight (2 pg on column). Final BMAA concentrations were calculated based on known sample volume and expressed as pg/L of lake water. In addition, BMAA was analyzed with a modified LC-MS/MS method employing Hydrophilic Interaction LC (HILIC) normal-phase column without derivatization (Kubo et al., 2008; Rosen and Hellenas, 2008). The LC column was a TSK-gel Amide 80 HILIC column (150 mm x 2.0 mm i.d., 5-pm particle size, Tosoh Bioscience Gmbh., Japan). BMAA was separated and eluted from column in (A) 0.05% aqueous TFA and a 90-60% linear gradient of (B) acetonitrile (mobile phase) over 15 min; flow rate was 0.3 mL/min. at 40°C. MRM transitions were: 1 19.1/102, 1 19.1/44 and 1 19.1/88. The detection limit for total BMAA was 1 pg/g dry weight (1 pg on column). Phytoplankton Community Analysis Aliquots of composite water were poured in 100-mL amber bottles and preserved in Lugol’s solution. These were submitted to Dr. C. Earle (Edmonton, Alberta) for phytoplankton community analysis (identification and enumeration). Analyses were performed on subsamples following acclimation to room temperature conditions, adequate re-suspension and settling by the Utermohl method. Phytoplankton was identified to the lowest confirmed level of classification. Enumeration involved counting transects (strips) along the length of the chamber at several magnifications. Alberta Environment Cyanotoxin Program Status Report 30 Appendix II List of Lakes/Reservoirs Alberta Environment Cyanotoxin Program Status Report 31 Table A1 List of lakes and reservoirs sampled for total microcystin in 2005 including trophic status, parent monitoring program and source of funding for toxin analyses. Site Name Trophic Status Program Funding Baptiste Lake North Basin Hypereutrophic LTLN LTLN Baptiste Lake South Basin Hypereutrophic LTLN LTLN Beauvais Lake Mesotrophic Prov. Parks Prov. Parks Bluet Lake (Gamier Lake South) Eutrophic ALMS WRUG Crimson Lake Mesotrophic Prov. Parks Prov. Parks Dillberry Lake Mesotrophic Prov. Parks WRUG Elkwater Lake Mesotrophic Prov. Parks WRUG Fishing Lake Hypereutrophic ALMS WRUG Fork Lake Eutrophic ALMS WRUG Frog Lake Mesotrophic ALMS WRUG Gamier Lake North Mesotrophic ALMS WRUG Goose Lake Hypereutrophic ALMS WRUG Gregg Lake Oligotrophic Prov. Parks WRUG Gregoire Lake Eutrophic Prov. Parks WRUG Hilda Lake Mesotrophic ALMS WRUG Island Lake Eutrophic ALMS WRUG Jarvis Lake Oligotrophic Prov. Parks WRUG Kananaskis Lake Lower Oligotrophic Prov. Parks WRUG Kananaskis Lake Upper Oligotrophic Prov. Parks WRUG Kehiwin Lake Hypereutrophic ALMS WRUG Long Lake (Near Boyle) Eutrophic Prov. Parks WRUG McLeod Lake (East) Eutrophic Prov. Parks Prov. Parks Miguelon Lake Mesotrophic Prov. Parks WRUG | Moonshine Lake Reservoir Eutrophic Prov. Parks Prov. Parks Moore (Crane) Lake Eutrophic ALMS WRUG [ Moose Lake Eutrophic ALMS WRUG Newell Lake Reservoir Mesotrophic Prov. Parks Prov. Parks Pine Lake Eutrophic ALMS WRUG Reesor Lake Reservoir Eutrophic Prov. Parks Prov. Parks . Saskatoon Lake Hypereutrophic Prov. Parks Prov. Parks Skeleton Lake North Basin Eutrophic ALMS WRUG Skeleton Lake South Basin Eutrophic ALMS WRUG Spruce Coulee Reservoir Mesotrophic Prov. Parks Prov. Parks Steele (Cross) Lake Hypereutrophic Prov. Parks WRUG Sturgeon Lake Hypereutrophic Prov. Parks Prov. Parks Wabamun Lake East Basin Eutrophic LTLN LTLN Wabamun Lake West Basin Eutrophic LTLN LTLN Whitefish Lake Mesotrophic ALMS WRUG Funded Winagami Lake Hypereutrophic Prov. Parks Prov. Parks Wolf Lake Mesotrophic ALMS WRUG Trophic status based on mean summer chlorophyll-a concentration. ALMS = Alberta Lake Management Society’s Lakewatch Program; LTLN = Long-term Lake Network Program; WRUG = Water Research User’s Group funded project. Alberta Environment Cyanotoxin Program Status Report 32 Table A2 List of lakes and reservoirs sampled for total microcystin in 2006 including trophic status, parent monitoring program and source of funding for toxin analyses. Site Name Trophic Status Program Funding Baptiste Lake North Basin Hypereutrophic BISL Cyanotoxin Program Baptiste Lake South Basin Hypereutrophic BISL Cyanotoxin Program Bear Trap Lake Mesotrophic ALMS Cyanotoxin Program Beauvais Lake Eutrophic Prov. Parks Cyanotoxin Program Big Lake Eutrophic ALMS Cyanotoxin Program Buck Lake Eutrophic Central AB Lakes Central AB Lakes Clear (Barnes) Lake Hypereutrophic ALMS Cyanotoxin Program Cooking Lake Hypereutrophic ALMS Cyanotoxin Program Crimson Lake Mesotrophic Prov. Parks Cyanotoxin Program Dillberry Lake Mesotrophic Prov. Parks Prov. Parks Elkwater Lake Mesotrophic Prov. Parks Prov. Parks Ethel Lake Mesotrophic LTLN LTLN Fishing Lake Hypereutrophic ALMS Cyanotoxin Program Frog Lake Mesotrophic ALMS Cyanotoxin Program Gleniffer Lake (Dickson Dam Res.) Oligotrophic Central AB Lakes Central AB Lakes Gregg Lake Oligotrophic Prov. Parks Prov. Parks Gregoire Lake Eutrophic Prov. Parks Prov. Parks Gull Lake AENV Eutrophic Central AB Lakes Central AB Lakes Gull Lake ALMS Eutrophic ALMS Cyanotoxin Program Hilda Lake Mesotrophic ALMS Cyanotoxin Program Island Lake Eutrophic BISL Cyanotoxin Program Jarvis Lake Oligotrophic Prov. Parks Prov. Parks Kananaskis Lake Lower Oligotrophic Prov. Parks Prov. Parks Kananaskis Lake Upper Oligotrophic Prov. Parks Prov. Parks Lac Sante Eutrophic ALMS Cyanotoxin Program Long Lake (Near Boyle) Eutrophic Prov. Parks Prov. Parks McLeod Lake (East) Eutrophic Prov. Parks Cyanotoxin Program Miquelon Lake Mesotrophic Prov. Parks Prov. Parks Mons Lake Eutrophic ALMS Cyanotoxin Program Moonshine Lake Reservoir Eutrophic Prov. Parks Cyanotoxin Program Moore (Crane) Lake Eutrophic ALMS Cyanotoxin Program ' Moose Lake Eutrophic ALMS Cyanotoxin Program Muriel Lake Mesotrophic ALMS Cyanotoxin Program Nakamun Lake Hypereutrophic LTLN LTLN Newell Lake Reservoir Mesotrophic Prov. Parks Cyanotoxin Program Pigeon Lake Eutrophic Central AB Lakes Central AB Lakes Pine Lake Eutrophic ALMS Cyanotoxin Program Red Deer Lake Hypereutrophic Central AB Lakes Central AB Lakes Reesor Lake Reservoir Eutrophic Prov. Parks Cyanotoxin Program Sandy Lake Hypereutrophic ALMS Cyanotoxin Program Saskatoon Lake Hypereutrophic Prov. Parks Cyanotoxin Program Skeleton Lake South Basin Eutrophic ALMS Cyanotoxin Program Spruce Coulee Reservoir Mesotrophic Prov. Parks Cyanotoxin Program Steele (Cross) Lake Hypereutrophic Prov. Parks Prov. Parks Sturgeon Lake Hypereutrophic Prov. Parks Cyanotoxin Program Sylvan Lake Mesotrophic ALMS Cyanotoxin Program Tucker Lake Eutrophic ALMS Cyanotoxin Program Winagami Lake Hypereutrophic Prov. Parks Cyanotoxin Program Wizard Lake Hypereutrophic ALMS Cyanotoxin Program Wolf Lake Mesotrophic ALMS Cyanotoxin Program Trophic status based on mean summer chlorophyll-a concentration. ALMS = Alberta Lake Management Society’s Lakewatch Program; BISL = Baptiste, Island and Skeleton Lakes Stewardship Society; LTLN = Long-term Lake Network Program. Alberta Environment Cyanotoxin Program Status Report 33 Table A3 List of lakes and reservoirs sampled for total microcystin in 2007 including trophic status, parent monitoring program and source of funding for toxin analyses. Site Name Trophic Status Program Funding Baptiste Lake North Basin Hypereutrophic LTLN LTLN Baptiste Lake South Basin Hypereutrophic LTLN LTLN Bear Trap Lake Mesotrophic ALMS Cyanotoxin Program Beauvais Lake Mesotrophic Prov. Parks Prov. Parks Lac Bellevue Mesotrophic ALMS Cyanotoxin Program Bittern Lake Eutrophic Central AB Lakes Central AB Lakes Clairmont Lake Hypereutrophic ALMS Cyanotoxin Program Clear Lake Eutrophic Southern L&R Cyanotoxin Program Clear (Barnes) Lake Oligotrophic ALMS Cyanotoxin Program Cooking Lake Hypereutrophic ALMS Cyanotoxin Program Crimson Lake Mesotrophic Prov. Parks Prov. Parks ! Dillberry Lake Mesotrophic Prov. Parks Prov. Parks Driedmeat Lake Hypereutrophic Central AB Lakes Central AB Lakes Eagle Lake Eutrophic Southern L&R Cyanotoxin Program Elkwater Lake Mesotrophic Prov. Parks Prov. Parks George Lake Hypereutrophic ALMS Cyanotoxin Program Gregg Lake Oligotrophic Prov. Parks Prov. Parks Gregoire Lake Eutrophic Prov. Parks Prov. Parks Hilda Lake Mesotrophic ALMS Cyanotoxin Program Jackfish Lake (Near Carvel) Eutrophic Central AB Lakes Central AB Lakes Jarvis Lake Oligotrophic Prov. Parks Prov. Parks Kananaskis Lake Upper Oligotrophic Prov. Parks Prov. Parks Kehiwin Lake Hypereutrophic ALMS Cyanotoxin Program Laurier Lake Mesotrophic ALMS Cyanotoxin Program Long Island North Basin Mesotrophic ALMS Cyanotoxin Program Long Island South Basin Mesotrophic ALMS Cyanotoxin Program Long Lake (Near Boyle) Eutrophic Prov. Parks Prov. Parks Marie Lake Mesotrophic ALMS Cyanotoxin Program Matchayaw Lake Hypereutrophic ALMS Cyanotoxin Program McLeod Lake (East) Mesotrophic Prov. Parks Prov. Parks ! Miquelon Lake Mesotrophic Prov. Parks Prov. Parks Moonshine Lake Reservoir Hypereutrophic Prov. Parks Prov. Parks j Moore (Crane) Lake Mesotrophic ALMS Cyanotoxin Program Newell Lake Reservoir Oligotrophic Prov. Parks Prov. Parks Pine Coulee Res. North Basin Eutrophic Southern L&R Cyanotoxin Program Pine Coulee Res. South Basin Mesotrophic Southern L&R Cyanotoxin Program ; Pine Lake Hypereutrophic Southern L&R Cyanotoxin Program Reesor Lake Reservoir Hypereutrophic Prov. Parks Prov. Parks Lac Sante Mesotrophic ALMS Cyanotoxin Program Saskatoon Lake Hypereutrophic Prov. Parks Prov. Parks Shorncliffe Lake Eutrophic Central AB Lakes Central AB Lakes Siler (Stoney) Lake Hypereutrophic ALMS Cyanotoxin Program Spruce Coulee Reservoir Mesotrophic Prov. Parks Prov. Parks Steele (Cross) Lake Hypereutrophic Prov. Parks Prov. Parks Sturgeon Lake Hypereutrophic Prov. Parks Prov. Parks Thunder Lake Hypereutrophic Central AB Lakes Central AB Lakes Tucker Lake Hypereutrophic ALMS Cyanotoxin Program Twin Valley Reservoir Eutrophic Southern L&R Cyanotoxin Program Wapasu Lake Mesotrophic ALMS Cyanotoxin Program Winagami Lake Eutrophic Prov. Parks Prov. Parks Wolf Lake Mesotrophic ALMS Cyanotoxin Program Trophic status based on mean summer chlorophyll-a concentration. ALMS = Alberta Lake Management Society’s Lakewatch Program; LTLN = Long-term Lake Network Program; Southern L&R = Southern Region Lakes and Reservoirs. Alberta Environment Cyanotoxin Program Status Report 34 Table A4 List of lakes and reservoirs sampled for total microcystin in 2008 including trophic status, parent monitoring program and source of funding for toxin analyses. Site Name Trophic Status Program Funding Adamson Lake Hypereutrophic Elk Island Park Elk Island Park Amisk Lake Mesotrophic ALMS Cyanotoxin Program Astotin Lake Hypereutrophic Elk Island Park Elk Island Park Bear Trap Lake Mesotrophic ALMS Cyanotoxin Program Beauvais Lake Mesotrophic Prov. Parks Prov. Parks Beaver Lake Mesotrophic ALMS Cyanotoxin Program Blackfalds Lake Hypereutrophic ALMS Cyanotoxin Program Clear Lake Mesotrophic Southern L&R Cyanotoxin Program Clear (Barnes) Lake Mesotrophic ALMS Cyanotoxin Program Cow Lake Oligotrophic Central AB Lakes Central AB Lakes Crimson Lake Mesotrophic Prov. Parks Prov. Parks Dillberry Lake Mesotrophic Prov. Parks Prov. Parks Elkwater Lake Mesotrophic Prov. Parks Prov. Parks Ethel Lake Mesotrophic LTLN LTLN Goose Lake Hypereutrophic ALMS Cyanotoxin Program Gregg Lake Oligotrophic Prov. Parks Prov. Parks Gregoire Lake Eutrophic Prov. Parks Prov. Parks Gull Lake Eutrophic Central AB Lakes Central AB Lakes Hastings Lake Hypereutrophic ALMS Cyanotoxin Program Jarvis Lake Oligotrophic Prov. Parks Prov. Parks Kananaskis Lake Lower Oligotrophic Prov. Parks Prov. Parks Kananaskis Lake Upper Oligotrophic Prov. Parks Prov. Parks Kehiwin Lake Eutrophic ALMS Cyanotoxin Program Lac La Nonne Hypereutrophic ALMS Cyanotoxin Program Laurier Lake Eutrophic ALMS Cyanotoxin Program Long Lake (Near Boyle) Eutrophic Prov. Parks Prov. Parks Marie Lake Mesotrophic ALMS Cyanotoxin Program McLeod Lake (East) Mesotrophic Prov. Parks Prov. Parks Minnie Lake Mesotrophic ALMS Cyanotoxin Program Miquelon Lake Mesotrophic Prov. Parks Prov. Parks Moonshine Lake Reservoir Hypereutrophic Prov. Parks Prov. Parks ' Moore (Crane) Lake Oligotrophic ALMS Cyanotoxin Program Nakamun Lake Hypereutrophic LTLN LTLN Newell Lake Reservoir Mesotrophic Prov. Parks Prov. Parks Oster Lake Hypereutrophic Elk Island Park Elk Island Park Pigeon Lake Eutrophic Central AB Lakes Central AB Lakes Pine Lake Hypereutrophic ALMS Cyanotoxin Program Reesor Lake Reservoir Eutrophic Prov. Parks Prov. Parks Sandy Lake Hypereutrophic Central AB Lakes Central AB Lakes Lac Sante Mesotrophic ALMS Cyanotoxin Program Saskatoon Lake Hypereutrophic Prov. Parks Prov. Parks Siler (Stoney) Lake Hypereutrophic ALMS Cyanotoxin Program Skeleton Lake South Basin Eutrophic ALMS Cyanotoxin Program Spruce Coulee Reservoir Oligotrophic Prov. Parks Prov. Parks Steele (Cross) Lake Eutrophic Prov. Parks Prov. Parks Sturgeon Lake Hypereutrophic Prov. Parks Prov. Parks Sylvan Lake Mesotrophic Central AB Lakes Central AB Lakes Twin Valley Reservoir Eutrophic Southern L&R Cyanotoxin Program Tyrell Lake Eutrophic Southern L&R Cyanotoxin Program Wapasu Lake Hypereutrophic ALMS Cyanotoxin Program Winagami Lake Hypereutrophic Prov. Parks Prov. Parks Wizard Lake Eutrophic ALMS Cyanotoxin Program Trophic status based on mean summer chlorophyll-a concentration. ALMS = Alberta Lake Management Society’s Lakewatch Program; LTLN = Long-term Lake Network Program; Southern L&R = Southern Region Lakes and Reservoirs. Alberta Environment Cyanotoxin Program Status Report 35 Table A5 List of lakes and reservoirs sampled for anatoxin-a in 2005 including trophic status and parent monitoring program. Site Name Trophic Status Program Baptiste Lake North Basin Hypereutrophic LTLN Beauvais Lake Mesotrophic Prov. Parks Bluet Lake (Gamier Lake South) Mesotrophic ALMS Crimson Lake Mesotrophic Prov. Parks Dillberry Lake Mesotrophic Prov. Parks Elkwater Lake Mesotrophic Prov. Parks Fishing Lake Hypereutrophic ALMS Fork Lake Eutrophic ALMS Frog Lake Mesotrophic ALMS Gamier Lake North Mesotrophic ALMS Goose Lake Hypereutrophic ALMS Gregoire Lake Eutrophic Prov. Parks Hilda Lake Mesotrophic ALMS Island Lake Eutrophic ALMS Kananaskis Lake Lower Oligotrophic Prov. Parks ! Kananaskis Lake Upper Mesotrophic Prov. Parks Long Lake Eutrophic Prov. Parks McLeod Lake Eutrophic Prov. Parks Miquelon Lake Mesotrophic Prov. Parks Moonshine Lake Reservoir Eutrophic Prov. Parks Newell Lake Reservoir Mesotrophic Prov. Parks Pine Lake Eutrophic ALMS Reesor Lake Reservoir Eutrophic Prov. Parks Saskatoon Lake Hypereutrophic Prov. Parks Skeleton Lake North Basin Eutrophic ALMS Skeleton Lake South Basin Eutrophic ALMS Spruce Coulee Reservoir Mesotrophic Prov. Parks Steele (Cross) Lake Hypereutrophic Prov. Parks Sturgeon Lake Hypereutrophic Prov. Parks Wabamun Lake East Basin Eutrophic LTLN ! Wabamun Lake West Basin Eutrophic LTLN j Whitefish Lake Mesotrophic ALMS Winagami Lake Hypereutrophic Prov. Parks Trophic status based on mean summer chlorophyll-a concentration. ALMS = Alberta Lake Management Society’s Lakewatch Program; LTLN = Long-term Lake Network Program. Alberta Environment Cyanotoxin Program Status Report 36 Appendix III Maximum Microcystin Concentrations in Lakes/Reservoirs Alberta Environment Cyanotoxin Program Status Report 37 Maximum Recorded Microcystin Concentrations in 2005 if) O o CM v_ 0 -Q E 0 +■» Q. 0 CO 1 0) ■D 0 a E (0 CO (/> L_ o £ 0 0 0 g *0 0 0 0 C o 2 ■»-> c 0 o c o o 0 > o o < e 3 O) il 00 ro Alberta Environment Cyanotoxin Program Status Report Maximum Recorded Microcystin Concentrations in 2006 ON m CO o o CN h. 0) -Q E a a CD co i CD £ 3 ■o _a> a E re co 0) v. o £ CD CO CD To CD J C re co £ O 3 re £ a a £ O O (0 > o o 04 < £ 3 O) Alberta Environment Cyanotoxin Program Status Report Maximum Recorded Microcystin Concentrations in 2007 / Q?>PZ *$k\ z&rtbASs % h%\ M+&> s$>, t> * V°xV°n % % v *$>< ;VvV , rw □ ;Wo te m4 v^ %?>%*> ?o 'b S3& vs& %%$ *oA \7^ < P ' *>><- S (~|/ba anoi/\| Bn) uoj;ej)U33UOQ uhsAoojo|i/\| Figure A3 Maximum Microcystin Concentrations in Lakes/Reservoirs Sampled June - September 2007 Maximum Recorded Microcystin Concentrations in 2008 Figure A4 Maximum Microcystin Concentrations in Lakes/Reservoirs Sampled June - September 2008 Appendix IV Quality Assurance/Quality Control Alberta Environment Cyanotoxin Program Status Report 42 Microcystin Analysis QA/QC Efforts were made to address aspects of QA/QC pertaining to the analysis of water samples for total microcystin via colorimetric PPI assay. In order to ensure comparability of data over the 4- year period, a standard operating procedure (SOP) developed by AENV was adopted by both laboratories involved in testing. Laboratories were required to use standard measurement units in reporting data. Any changes or modifications of the testing protocol by the laboratories were submitted in advance to the Program Coordinator for review. As a measure of Laboratory QC, the PPI assay protocol requires that for each sample, 3 subsamples (in 3 microtiter plate test wells) be tested for inhibitory activity to the PPI enzyme. Values for individual subsamples (wells) are scrutinized and may be disregarded at the discretion of the Laboratory Technician if issues with plating of samples are noted. Sample analysis may be deemed invalid if precision (reproducibility) exceeds %CV >5. These are re-analyzed. The mean of the subsamples is reported as the value for total microcystin. Lastly, it was desirable to assess whether the inhibitory responses (to the PP 1 enzyme) of samples quantified by the PPI assay was due to microcystins. Forty four (44) samples collected during 2007 were submitted to the ACFT Laboratory. Subsamples of each were analyzed by: (1) PPI assay to estimate total microcystin concentration (based on toxicity equivalents to MCLR toxin standards) and (2) LC-MS/MS analysis of MCLR, MCRR, MCYR, MCLW and MCLF concentrations. Three (7%) samples lacked MCYST as determined by PPI assay (i.e. <0.1 pg MCLReq./L) and LC-MS/MS analysis (Table A6). Microcystin was detected in 41 (93%) of the 44 samples by PPI assay (Table A6). In contrast, only 36 (82%) of the 44 samples analyzed by LC-MS/MS contained detectable concentrations of MCLR (Table A6). Samples analyzed by LC-MS/MS also contained detectable levels of MCRR and MCYR (Table A6), but no MCLW or MCLF (data not shown). Thus in 5 samples, no specific toxin analogues were detected by LC-MS/MS analysis, yet inhibitory responses to PPI enzyme were noted by PPI assay. Given that more than 70 MCYSTs have been described globally (Zurawell, 2001) - only 5 of which were quantified by LC-MS/MS is this study - it is quite possible the toxic responses (as measured by PPI assay) of these 5 samples may be attributed to unmeasured analogues (see Section 3.2 - Microcystin Analogue Study). However, there remains a remote possibility that a toxic response in the PPI assay may result from the presence of tautomycin - a naturally occurring PPI inhibitor originating from soil bacteria ( Streptomyces spiroverticillatus). Although it is highly unlikely that tautomycin occurs in lakes at concentrations required to elicit an inhibitory response (in the PPI assay), this possibility is being addressed. Alberta Environment Cyanotoxin Program Status Report 43 Table A6 Results of paired analyses for total MCYST by PPI assay and LC-MS/MS analysis for specific MCYST analogues results for MCLR, MCRR and MCYR. Sample Date MCYST by PPI MCLR MCRR MCYR (Mg MCLReq./L) ( pg/L) (m g/L) (mq/l) Baptiste Lake South 25/06/07 0.36 <0.1 <0.2 <0.1 Baptiste Lake South 18/07/07 0.9 0.6 <0.2 <0.1 Baptiste Lake South 08/08/07 1.56 2.4 0.24 <0.1 Baptiste Lake South 30/08/07 5.2 6.5 <0.2 <0.1 Baptiste Lake South 27/09/07 1.23 1.1 <0.2 <0.1 Baptiste Lake South 29/10/07 <0.07 <0.1 <0.2 <0.1 Baptiste Lake North 18/07/07 1.43 0.8 <0.2 <0.1 Baptiste Lake North 08/08/07 1.42 2.3 <0.2 <0.1 Baptiste Lake North 30/08/07 3.63 4.2 <0.2 <0.1 Baptiste Lake North 27/09/0 7 2.37 1.9 <0.2 <0.1 Baptiste Lake North 29/10/07 0.19 0.1 <0.2 <0.1 Bittern Lake 11/08/07 0.19 <0.1 <0.2 <0.1 Clairmont Lake 22/10/07 2.45 4.5 <0.2 <0.1 Clairmont Lake 19/09/07 2.45 5.4 <0.2 <0.1 Clear Lake 17/07/07 1.39 1.7 <0.2 <0.1 Clear Lake 29/08/07 11.2 17.2 <0.2 <0.1 Clear Lake 21/09/07 0.89 2.1 <0.2 <0.1 Driedmeat Lake 13/08/07 0.89 2.1 0.89 <0.1 Driedmeat Lake 19/08/07 0.31 0.2 <0.2 <0.1 George Lake 01/08/07 24.33 39.6 <0.2 <0.1 George Lake 23/08/07 6.84 11.6 <0.2 <0.1 Jackfish Lake 15/08/07 0.08 <0.1 <0.2 <0.1 Jackfish Lake 18/08/07 0.36 <0.1 <0.2 <0.1 Miquelon Lake 27/06/07 0.29 <0.1 <0.2 <0.1 Moonshine Lake 31/07/07 0.23 0.6 <0.2 <0.1 Moonshine Lake 23/08/07 0.41 0.5 <0.2 <0.1 Moonshine Lake 23/08/07 0.1 0.2 <0.2 <0.1 Moonshine Lake 18/09/07 <0.07 <0.1 <0.2 <0.1 Moonshine Lake 31/07/07 0.26 0.7 <0.2 <0.1 Saskatoon Lake 02/08/07 3.1 8.9 <0.2 <0.1 Saskatoon Lake 22/08/07 9.8 12.2 <0.2 <0.1 J Pine Coulee Res. North 15/08/07 0.31 0.3 <0.2 <0.1 Pine Coulee Res. North 10/07/07 0.08 0.1 <0.2 <0.1 Pine Coulee Res. South 10/07/07 <0.07 <0.1 <0.2 <0.1 Pine Lake 25/07/07 0.26 0.4 <0.2 <0.1 Pine Lake 15/08/07 0.5 0.8 <0.2 <0.1 Shorncliffe Lake 15/08/07 3.68 3.4 <0.2 <0.1 Sturgeon Lake 21/09/07 0.13 0.3 <0.2 <0.1 Thunder Lake 16/07/07 1.09 5.6 1.39 <0.1 Thunder Lake 25/08/07 4.1 13.6 4.89 0.16 Thunder Lake 24/09/07 3.93 9.1 4.12 <0.1 Twin Valley Reservoir 31/07/07 0.51 1.0 <0.2 <0.1 Twin Valley Reservoir 24/09/07 0.16 0.3 <0.2 <0.1 Twin Valley Reservoir 29/08/07 223.5 206.0 19.8 1.1 n.d. = no toxin analogue detected. Alberta Environment Cyanotoxin Program Status Report 44 Table A7 Raw Data 2005 Alhoria Pn\/irnnmQnt C' wonntnvin Drnnrom Q+oti ic Donnrt A Table A7 Raw Data 2005 Alberta Environment Cyanotoxin Program Status Report Table A7 Raw Data 2005 Alhorta Fn\/irnnmont r'\/onr>tr»vin Drnnrcim Qtotiic Darw-iri- /IQ Table A8 Raw Data 2006 Aiberta Environment Cyanotoxin Program Status Report Table A8 Raw Data 2006 o 10 Alberta Environment Cyanotoxin Program Status Report Table A8 Raw Data 2006 Alberta Environment Cyanotoxin Program Status Report Table A9 Raw Data 2007 Alberta Environment Cyanotoxin Program Status Report Table A9 Raw Data 2007 Aiberta Environment Cyanotoxin Program Status Report Table A9 Raw Data 2007 V) Alberta Environment Cyanotoxin Program Status Report Table A10 Raw Data 2008 Aiberta Environment Cyanotoxin Program Status Report Table A10 Raw Data 2008 Alberta Environment Cyanotoxin Program Status Report Table A10 Raw Data 2008 i i Alberta Environment Cyanotoxin Program Status Report Table A10 Raw Data 2008 00 IT) Alberta Environment Cyanotoxin Program Status Report i I I i ! f i i