Editorial Manager(tm) for Water, Air, & Soil Pollution Manuscript Draft
Manuscript Number: WATE2819R1 Title: Ecotoxicity of snowpack collected from selected sites in Québec, Canada Article Type: Full research paper Keywords: snowpack; atmospheric pollution; toxicity; bioassays; biomarkers; bacteria; micro-algae; micro-invertebrates; fish liver cells; mercury. Corresponding Author: Dr. Christian René Blaise, D.Sc. Corresponding Author's Institution: Fluvial Research section First Author: François Gagné, D.Sc. Order of Authors: François Gagné, D.Sc.; Laurier Poissant, Ph.D.; Christian René Blaise, D.Sc. Abstract: Sampling was conducted at seven sites on, and at varying distances adjacent to, the Island of Montréal (Québec, Canada), and as far as 1100 km away in Northern Québec, to explore the hazard potential of snowpacks in remote, rural and urban environments. Ecotoxic effects of melted snow were ascertained with a suite of small-scale bioassays representing several aquatic taxonomic groups (bacteria, micro-algae, micro-invertebrates, fish liver cells), as well as with biomarker measurements determined with a rainbow trout primary hepatocyte (RTPH) assay. Bioassays undertaken with the cnidarian Hydra attenuata and RTPH cell assays, and to a lesser extent with the micro-alga P. subcapitata, proved particularly sensitive to infer the presence of bioavailable pollutants in snow samples collected from all sites, thereby suggesting their contamination (at least) via atmospheric sources. Furthermore, biomarker responses indicated that snow samples presumably included metals (free Zn biomarker), organics (CYP 1A1 biomarker), estrogens (Alkali Labile Phosphate biomarker), as well as chemicals capable of causing oxidative stress (LPO biomarker), depending on the site being considered. Overall, effects data acquired during this preliminary investigation on the ecotoxicity of snowpacks submit that adverse impact
toward aquatic biota is conceivable at some sites during spring meltdown. Because snow has a recognized affinity for sequestering solids and contaminants of atmospheric origin, future studies aimed at identifying sources and chemicals implicated in observed effects are legitimate endeavors. Response to Reviewers: Response to reviewer comments We have considered all of the reviewer remarks and have amended our manuscript accordingly. COMMENT 1: Page 4 "Bioassay" Among endpoints of five bioassays, 48h-TEC of fish cell test seems to be different from other endpoints because the threshold value is more sensitive than IC50 or EC50. Is there a special reason for this ? RESPONSE 1: As explained in Gagné et al (2005), the 48h-TEC endpoint is traditionally employed with the fish cell test (rainbow trout primary hepatocyte test) which signals a graduated response of cell viability that is more sensitive than a 50 % effect.
COMMENT 2: Page 4 Weight scores based on effects noted for each toxicity test: It is difficult to understand the class weight score in Table.3 because there is no definition about the class score. And please check the formula of class weight score. RESPONSE 2: We have added additional detail in section 2.3 and given an example in Table 3 to clarify the understanding of the class weight score calculations.
COMMENT 3: Page 4 2.4.1 Cell viability and sublethal effects assessments It is better to give a brief introduction of the meaning of each biomarker to help readers' understanding. Some biomarkers such as labile zinc and alkali-labile phosphate are not so popular compared to other biomarkers. RESPONSE 3: We have added additional information in section 2.4.1 to better explain the significance of the biomarkers (labile zinc, LPO, CYP1A1 and vitellogenin) we measured in our study.
COMMENT 4: Page 5, line 29 Please check the following expression. .0.25 to 1.3 mg/L (NO2/NO3) and from 0.1/3.6 mg/L, respectively, RESPONSE 4: We have corrected the small error related to the inorganic phosphorus concentrations that the revieweer noted. COMMENT 5: Page 6. 3 rd paragraph related to benzene In this paragraph, benzene emission from industrial and traffic sources were discussed. This discussion implies the potential cause of bioassay results indirectly, but there is no date on sensitivity of each bioassay to benzene. It is better to show the direct evidence if these bioassays are sensitive enough to show the corresponding results shown in this study. RESPONSE 5 : We have added information and an accompanying reference that better infers a link between our bioassay results and the possible toxicity of benzene in snow.
COMMENT 6: Page 7. 2nd paragraph related to metal mobilization For biomarker assays, samples were extracted by C18 resin. The discussion in this paragraph seems to be doubtful if metals are supposed to go through C18 resin. RESPONSE 6: We have added a key reference to confirm that metals present in melted snow can be eluted with organic matter after C18 extraction
COMMENT 7: Table 3 100% (neat) or 100% (near) RESPONSE 7: OK, "neat" is the proper word COMMENT 8: Table 4. There are many zero increases in this table. However, it is unusual to get the precise zero increase because of the variations of responses. RESPONSE 8 : We have defined what the "0" values signify by adding a footnote to Table 4.
Gagné et al revised ms Click here to download Manuscript: Gagn et al WASP.doc
Click here to view linked References
1 Ecotoxicity of snowpack collected from selected sites in Québec, Canada
François Gagné, Laurier Poissant and Christian Blaise*
Fluvial Ecosystems Research, Aquatic Ecosystem Protection Research Division, Water Science and Technology Directorate, Environment Canada, 105 McGill Street, Montréal, Québec, Canada, H2Y 2E7. *Corresponding author:
[email protected]; Tel: (514) 496-7094; Fax: (514) 496-7398
Abstract
Sampling was conducted at seven sites on, and at varying distances adjacent to, the Island of Montréal (Québec, Canada), and as far as 1100 km away in Northern Québec, to explore the hazard potential of snowpacks in remote, rural and urban environments. Ecotoxic effects of melted snow were ascertained with a suite of small-scale bioassays representing several aquatic taxonomic groups (bacteria, micro-algae, micro-invertebrates, fish liver cells), as well as with biomarker measurements determined with a rainbow trout primary hepatocyte (RTPH) assay. Bioassays undertaken with the cnidarian Hydra attenuata and RTPH cell assays, and to a lesser extent with the micro-alga P. subcapitata, proved particularly sensitive to infer the presence of bio-available pollutants in snow samples collected from all sites, thereby suggesting their contamination (at least) via atmospheric sources. Furthermore, biomarker responses indicated that snow samples presumably included metals (free Zn biomarker), organics (CYP 1A1 biomarker), estrogens (Alkali Labile Phosphate biomarker), as well as chemicals capable of causing oxidative stress (LPO biomarker), depending on the site being considered. Overall, effects data acquired during this preliminary investigation on the ecotoxicity of snowpacks submit that adverse impact toward aquatic biota is conceivable at some sites during spring meltdown. Because snow has a recognized affinity for sequestering solids and contaminants of atmospheric origin, future studies aimed at identifying sources and chemicals implicated in observed effects are legitimate endeavors. Keywords: snowpack; atmospheric pollution; toxicity; bioassays; biomarkers; bacteria; micro-algae; microinvertebrates; fish liver cells; mercury.
2 1 Introduction Rain and snow precipitations are known to scavenge various types of atmospheric pollutants that are then transferred to terrestrial and aquatic environments (e.g., Poissant et al., 1994). Snow flakes and snow crystals in particular, owing to their respective larger surface areas and lower velocity over rain drops, can indeed prove potent in collecting particulates from the atmosphere (Schemenauer et al., 1981). Over the course of the winter season, snowpacks in urban and even in rural or uninhabited regions are likely to accumulate important quantities of solids and contaminants originating from sources such as airborne fallout, internal combustion vehicles, road maintenance (grit and salt applications), as well as from long range transport and deposition events (Oberts et al., 2000; Simonetti et al., 2000; Garbarino et al., 2002; Herbert et al., 2006). Heavy metals and Pb and Sr isotope ratios have been determined in abundance in precipitation collected between December 1997 and June 1999 in the vicinity of Montréal, Canada (St. Anicet and L’Assomption). Median enrichment factors for the heavy metals, relative to upper crustal abundances, are all >>10 indicative of an anthropogenic origin (Simonetti et al., 2000). The range in pH values was from ~3.9 to ~5.0. The mean abundances for heavy metals were higher at St. Anicet compared to those at L’Assomption, especially those for the 1997–1998 winter. This was especially intriguing since the St. Anicet station is located in a rural site, whereas L’Assomption is considered a semi-urban site (North-east of Montréal). Notably, most of the samples from St. Anicet contained the highest concentrations of Cd, Cu, and Zn. This finding along with higher mean abundances for heavy metals at St. Anicet was attributed to atmospheric emissions from the Zn refining plant (CEZinc) located 25 km northeast of St. Anicet (Simonetti et al., 2000). Atmospheric mercury depletion events (AMDEs) were first observed in the Canadian Arctic during the spring of 1995 (Schroeder et al., 1998). Throughout AMDEs, atmospheric Hg(0) levels drop from 1.6–1.8 ng m-3 to concentrations near zero, whereas Hg(II) species concentrations increase from 0.002 ng m-3 to levels as high as 0.9 ng m-3 (Lindberg et al., 2001). Field observations (Poissant and Hoënninger, 2004; Carpenter et al., 2005) suggested that atmospheric oxidation of Hg (0) to Hg(II) during AMDEs is caused by reactive bromine monoxide (BrO), bromine (Br) and iodine-containing compounds. These reactive halogens that originate from the photolysis of inorganic halogen species are released to the troposphere by the autocatalytic oxidation of sea-salt halides and the degradation of organohalogens. Constant et al., 2007 pointed out that following AMDEs snow total mercury (THg) concentrations increased by factors of 3-10. However, up to 80% of the THg is lost from the snow surface, over a short period (~12 h). The loss of THg may be attributable to a photoinduced reduction of Hg(II) to volatile Hg (0) within the snow surface (Poissant et al., 2002). Furthermore, the correlations between MeHg, sulfate (SO4=), and chlorine (Cl-) snow concentrations implicated marine aerosols as a significant source of MeHg, independent of AMDEs. However, the newly deposited MeHg was unstable in the snow cover as 15–60% of the MeHg was demethylated or otherwise ‘‘lost’’ during the nighttime period. During the snowmelt, although the THg snow concentrations remained at 8–9 ng L-1, the proportion of MeHg increased from 2.7 to 7.6%. Accordingly, potential impact of melting snow during subsequent spring runoff on water quality and its possible adverse effects on aquatic biota of receiving environments is clearly an undesirable consequence that is of concern. In fact, a few studies have documented the toxic (Hagen and Langeland, 1973; Marsalek et al., 1999) and genotoxic (White et al., 1995) potential of snow collected from diverse areas. To further address the issue of contaminated snow and to enhance understanding of the toxic potential it might contribute to aquatic systems in Eastern Canada (Québec), we conducted a preliminary investigation on the ecotoxicity of snowpack samples collected from seven sites reflecting remote, rural and urban characteristics, as well as a wide geographical span. Sites were specifically selected to represent various types of environmental impact such as urban and industrial air pollution plumes, long range transport, natural atmospheric fallout (especially the atmospheric mercury depletion event, AMDE) and background conditions. Furthermore, selected sites are close to routinely monitored meteorological stations to which they can be compared.
3 Hence, for this snowpack study, a suite of bioassays and biomarkers were employed to determine several types of adverse effects usually associated with mixed contaminants discharging to aquatic systems. The results of our study are presented herein.
2 Materials and Methods 2.1 Snow sample sites and collection Snow samples were collected from seven sites across Québec in March 2004. The Kuujjuarapik station is a remote site located in North-Western Québec along the East coast of the Hudson Bay. The Frelighsburg and St-Anicet stations are rural sites (~ 100 -150 km from Montréal). The former is located well away from nearby and other direct potential sources of atmospheric pollution impacting the city of Montreal, while the other sites are in, or adjacent to, this metropolitan area (Fig. 1). The rurally-located St.Anicet site is nevertheless impacted by nearby industrial activities (see Introduction). Other available data on the characteristics of snow-sampled sites are regrouped in Table 1. Acid-cleaned materials were used and full clean room clothing and polyethylene gloves were worn for all snow sample collection. The top 5 cm of snowpack were collected (Fig. 2) using a clean protocol (Quémerais and Cossa, 1997; Constant et al., 2007). Snow samples (placed in 2L Teflon bottles) were packed inside two sealed polyethylene bags. All samples were kept frozen in the dark prior to analysis.
2.2 Mercury analyses Continuous air TGM measurements were conducted at the Kuujjuarapik and St. Anicet sites with automatic Tekran mercury analyzers (Model 2537A). This instrument sampled the air and trapped the mercury vapor into a gold cartridge. Amalgamated mercury was then thermally desorbed and detected by using Cold Vapor Atomic Fluorescence Spectrometry (CVAFS). Two cartridges were mounted in parallel, allowing continuous TGM measurements by alternate sampling and desorption cycles. The analyzer was operated with 15 min sampling intervals at a flow rate of 1.5 L min-1 (Poissant, 1997). Snow samples for THg analyses at the Kuujjuarapik site were collected in duplicate (500 mL Teflon bottles) and measured in a class 100 clean room at the Environment Canada laboratories (Montréal, Canada). Mercury was dissociated from organic complexes with BrCl followed by a reduction of the monovalent and divalent mercury forms to Hg(0) with SnCl2. Volatilized Hg (0) was then concentrated on a gold trap that was finally heated to release mercury vapors detected with a Tekran Model 2500 atomic fluorescence spectrophotometer (Quémerais and Cossa, 1997) following the modified US EPA Method 2620. A detection limit of 10 pg L -1 (three times the standard deviation) was achieved. Mercury concentrations in precipitation were measured at the St. Anicet site using mercury the Deposition Network protocol and analyzed by Frontier Geosciences (Seattle, USA). 2.3 Bioassays The characteristics of liquid phase bioassays conducted to assess the toxicity of snow samples are highlighted in Table 2. Measurement endpoints of the bioassays were determined with statistical methods and software recommended for each procedure. Because bioassay responses with snow samples rarely yielded effects greater than 50% allowing for the determination of IC, EC or LC50s (see Table 2), we employed a toxicity classification system proposed by Persoone et al. (2003) to compare sites. It can be applied, for instance, for low toxicity response liquid samples that can be characteristic of slightly contaminated waters (e.g., groundwater, river water, drinking water). This scheme is based on toxicity responses generated with a battery of bioassays (similar to that used in the present study) on undiluted (neat) melted snow samples. It first assigns a toxicity class to each site according to the following criteria: Hazard Classes : Class I (no acute hazard): no toxicity tests in the battery showing toxic effects Class II (slight acute hazard): at least one test in the battery shows a 20% effect over control, but less than a 50% effect
4 Class III (acute hazard): at least one test in the battery shows a 50% effect over control, but less than a 100% effect Class IV (high acute hazard): at least one test in the battery shows a 100% effect Class V (very high acute hazard): all tests in the battery show a 100% effect Weight scores are then assigned for each snow sample (whose Hazard class has been determined) as follows: Weight scores based on effects noted for each toxicity test: 0: < 20% effect 1: > 20% but < 50% 2: > 50% but < 100% effect 3: ≥ 100% effect The class weight score is then the sum of all toxicity test scores divided by the number of tests performed. A class weight score in % for a particular class is then determined according to the following formula: Class weight score in % = [total class score / maximum class weight score] x 100 Sites toxicity reported with this Hazard Classification system are presented in Table 3. 2.4 Biomarkers The cytotoxic properties of snow samples were examined on a C18 solid phase extract. Snow samples were melted at room temperature and 100 mL were passed through a C18 solid phase cartridge (0.5 mg). The cartridge was washed with 5 mL of bi-distilled water, air dried under vacuum and the material eluted with 1 mL ethanol (100 fold concentrate). A blank was also prepared with 100 mL of bi-distilled water. Primary cultures of rainbow trout hepatocytes (Oncorhynchus mykiss) were prepared according to an adaptation of the double perfusion method using albumin and citrate to liberate cells (Gagné, 2005). Briefly, livers from n = 4 rainbow trout (8–12 cm long) were used for the preparation of primary cell cultures. Hepatocytes were plated (n = 6 wells per treatment) in 48-well microplates at a density of 0.5 x 106 cells/mL in Liebovitz (L-15) medium containing 10 mM Hepes-NaOH, pH 7.4, 100 units of penicillin, 100 µg/mL streptomycin and 0.25 µg/mL amphotericin B. They were then exposed to increasing concentrations of ethanol extract at 0, 0.008, 0.04, 0.2 and 1 % for 48 h at 15oC. In parallel, hepatocytes were exposed to 100 µg/mL of molecular cadmium (CdSO4) as a positive control. At the end of the exposure period, microplates were centrifuged at 500 x g for 2 min and the culture medium removed by aspiration. Cells were resuspended in phosphate-buffered saline containing 1 mM glucose, 1 mM citrate and 10 mM Hepes-NaOH, pH 7.4 for cell density and cell viability assessments. For cell density, absorbance at 600 nm was measured using a microplate reader (Powerwave Reader, USA). The remaining cell suspension was stored at -85oC awaiting later analysis. 2.4.1 Cell viability and sublethal effects assessments Cell viability was evaluated by the carboxyfluorescein diacetate dye retention assay as described elsewhere (Gagné, 2005). The levels of labile zinc were determined in the S15 fraction according to a fluorescent probe method (Gagné et al., 1996). The labile zinc biomarker reflects the release of zinc during heavy metal exposure and/or oxidative stress. Briefly, a portion of the cell suspension was thawed at 4oC and homogenized using a Teflon pestle tissue grinder (5 passes). The cell suspension was further homogenized by repeated mixing with a micro-pipette fitted with a 100 µL tip and was diluted 5-fold in water. A volume of 50 µL was mixed with 150 µL of 50 µM TSQ (N-[6-methoxy-8-quinolyl]-p-toluene sulfonamide) probe in 10% DMSO and phosphate-buffered saline (PBS: 140 mM NaCl, 5 mM KH2PO4 pH 7.4) and allowed to incubate for 10 min at room temperature. Fluorescence was measured at 400 nm excitation and 485 nm emission in a microplate reader (Bioscan, Chameleon-II). Standard solutions of zinc sulphate were prepared for calibration. Data were expressed as relative zinc equivalents/cell density. Lipid peroxidation, indicative of malonaldehyde formation following breakdown of oxidized unsaturated lipids, was determined in hepatocyte homogenates by the thiobarbituric acid method (Wills, 1987). Thiobarbituric acid reactants
5 (TBARS) were again determined by fluorescence (540 nm excitation and 590 nm emission) with the microplate reader. Data were expressed as µg of thiobarbituric acid reactants (TBARS)/cell density (absorbance at 600 nm). CYP1A1 activity was measured in cells using 7-ethoxyresorufin as the substrate (ethoxyresorufin O-deethylase or EROD) as described elsewhere (Kennedy et al., 1993). Enzymatic activities associated with this cytochrome reflect oxidative metabolism of polycyclic aromatic compounds to which primary hepatocytes have been exposed. The levels of vitellogenin-like proteins in the extracellular medium were determined using the alkali-labile phosphate assay with slight modifications (Gagné et al., 1999a). Vitellogenin is a precursor of the egg-yolk protein that is synthesized by 17βestradiol as well as by other estrogenic compounds (e.g., nonylphenol). It is therefore a useful biomarker to detect the presence of endocrine disruptors that act as estrogen mimics. Briefly, 500 µL of the cell-free culture medium were collected and centrifuged at 10 000 x g for 5 min to remove any cell debris. Acetone was added to the supernatant to give a final concentration of 30 %. The mixture was centrifuged at 10 000 x g for 5 min to collect high molecular weight proteins (i.e., vitellogenin). The pellet was mixed with 100 µL of NaOH 1 M for 30 min at 50oC to liberate alkali-labile phosphates. The levels of phosphates were then determined using the phosphomolybdenum reaction as described (Gagné et al., 1999a). Data were expressed as µg phosphate/cell density (absorbance at 600 nm). 2.5 Data analysis For primary hepatocyte biomarker measurements, data were checked for homogeneity of variance using Levene’s test. They were then subjected to an analysis of variance and critical differences between the unexposed cell groups vs. the treated group were appraised using the Least Square Difference test (LSD test). Data were next transformed as fold-response in respect to untreated cells (e.g. MT response of treatment/mean MT value of controls) and the 20 % effect concentration was calculated. Correlation analyses (Kendall’s concordance test, significance at p < 0.05) of bioassay, biomarker and chemical data, as well as Cluster analysis, were performed with Statistica® software, version 5.5.
3 Results and Discussion
3.1 Bioassay results Hazard classification of snow sites was based on results of the five toxicity tests described in Table 2. Three of these five bioassays were unable to discriminate sites in terms of their toxic potential. For the Microtox assay, all 100% (neat) melted snow concentrations displayed effects below 20% (i.e., 3.1 to 13.7% effects over controls) for the seven sites (Table 3). Likewise, the acute lethality T. platyurus test failed to show any toxic effects. While the algal bioassay also failed to elicit toxicity, it demonstrated marked stimulatory effects over controls at six sites (range of 20.8 to 116% stimulation over controls) suggesting the presence of chemicals in snow capable of enhancing the growth of primary producers. In a comprehensive review, atmospheric contribution of nutrients has been clearly demonstrated with nitrogenous (TKN, NO 2 and NO3) and inorganic phosphorus (P) snow pile concentrations reported to range from 1.2 to 6.6 mg/L (TKN), 0.25 to 1.3 mg/L (NO2/ NO3) and from 0.1 to 3.6 mg/L (P), respectively, all residential, commercial and industrial sites confounded (Oberts et al., 2000). More sensitive and varying effects observed with the Hydra sublethality and rainbow trout hepatocyte cytotoxicity assays enabled demarcation of snow sites based on the classification system employed (Persoone et al., 2003). Hydra bioassays have indeed been reported sensitive to a wide spectrum of contaminants of metallic, organic or mixed origin (Blaise and Kusui, 1997; Pardos et al., 1999; Arkhipchuk et al., 2000;; Holdway, 2005), as have cytotoxicity tests conducted with rainbow trout primary hepatocytes (Strmac and Braunbeck, 2002; Finne et al., 2007). Overall, these responses reflected slight acute (Class II), acute (Class III) and high acute (Class IV) hazard indicating the presence of varying degrees of contamination in snow collected from these sites. Regardless of their remote, rural or urban-
6 influenced location (Fig. 1), each site is subject to at least one or more sources of chemical contamination (Table 1). The post-AMDE event at the remote Kuujjuarapik station appears to have had a marked influence on snow contamination in comparison to pre-AMDE conditions, as shown by mercury concentrations that increase more than two-fold (Table 1). Bioassays are also clearly influenced by the post-AMDE situation, as Hydra and hepatocyte test responses increase 1.6- and 4.4-fold, respectively (Table 3). Notable as well is the response of the algal test which displays 35% stimulation in pre-AMDE snow and, in contrast, 3.6% inhibition under post-AMDE conditions, suggesting that an increase in snow contamination has offset the pre-AMDE auxinic effects of substances also likely prevalent in snow. During AMDE events, important loads of mercury fallout (100-200 ng m-2) can occur including halogen species that can potentiate mercury methylation and/or demethylation (Constant et al., 2007; Lahoutifard et al., 2006). Lahoutifard et al. (2006) pointed out that total organic carbon in spiked irradiated snow samples decreased (97 to 50 μM, and to 17 μM, at the detection limit, upon irradiation in the presence of H2O2), consistent with the formation of strongly oxidizing species in Kuujjuarapik’s snow. Indeed, the snow in Kuujjuarapik is saline as well as acidic: Chloride and sulfates in snow are typically < 5 mg L-1 and < 0.5 mg L-1 and pH ~5.1-5.4. (Constant et al., 2007). The snow likely becomes more acidic as spring progresses. Furthermore, both H+ and Cl- are concentrated in the quasi-liquid layer present on the surfaces of snow and ice crystals. Higher concentrations of both hydrogen ions and chloride ions cause the driving force for Hg(0) oxidation. The product Hg(II) is stabilized as HgCl 2. Laboratory evidence for the direct oxidation of Hg(0) by H2O2 is weak, both in the gas phase and in aqueous solution. In snow, the reaction may be indirect. Photolysis of H2O2 has been suggested to be an aqueous phase source of hydroxyl radicals. The oxidation of Hg(0) by OH has been demonstrated in the gas phase as well as the aqueous phase. The fate of HgOH is likely its oxidation by O2. Alternately, the photolytic reaction may be mediated by a naturally-occurring organic or inorganic trace contaminant. For example, a photo-initiated reaction of hydrogen peroxide with organic carbon may generate radicals which react readily with Hg(0). The observed decrease in total organic carbon in snow in irradiated and spiked snow samples by Lahoutifard et al. 2006 is consistent with the presence of highly oxidizing species (e.g., H 2O2 and/or OH). The peroxide may also transform halides (Cl-, Br-) to hypohalites (ClO-, BrO-), which then oxidize Hg(0) directly (Lahoutifard et al., 2006). Accordingly, the post-AMDE snowpack is reactive and may explain the above bioassay results in the post-AMDE situation, as Hydra and hepatocyte test responses indicate. As the only bioassay-designated Class IV hazard site, the Montréal-Est snow-sampled station is the only one in close proximity to the important industrial sector comprising numerous Petro-chemical plants located on the eastern side of the island of the City of Montréal (Fig. 1). Air pollution fallout from these industries is presumably the primary cause of snow contamination. For snow samples collected here, the Hydra test generated the most intense sublethal and lethal effects, while those of the hepatocyte test were the second highest of all sites (Table 3). The level of air pollution in the east end of the Island of Montreal is of concern. Volatile organic compounds, benzene, tropospheric ozone, sulfur oxide and nitrogen are routinely analyzed by Environment Canada at different stations. Motor traffic is responsible for the major part of benzene emissions on the Island of Montreal but not in the vicinity of Rivière des Prairies where benzene comes mostly from industrial sources located nearby. Indeed, the highest levels were recorded when the winds were blowing from the west-south-west and western sectors, which is where the industrial plants that reported major atmospheric releases of benzene are located. In 1994, an Environment Canada report stated that the Pointeaux-Trembles (PAT) station in Montreal had the highest mean (8.6 μg m-3) and maximum (126 μg m-3) levels of benzene in Canada during the 1989-1993 period (Dann, 1994). In 2000, Dann (2000) reported that benzene values measured at the PAT station (mean of 9.0 μg m-3; maximum of 126 μg m-3) were still the highest recorded for all Environment Canada stations in the 1989-1998 period. Between 1989 and 2000, however, benzene values declined in air samples collected on the Island of Montreal, with decreases of 60% reported at PAT. While regulation has helped to promote a drop in benzene releases to the atmosphere in Montreal in 2000 (Germain et al., 2001), this contaminant may yet be linked, to some extent, to observed toxicity responses elicited by the Hydra and rainbow trout hepatocyte assays (Table 3), as its toxic effects towards invertebrates and fish have been reported (CCME, 1999).
7
3.2 Biomarker results Trout hepatocytes exposed to ethanolic extracts of snow samples revealed a variety of sublethal effects (Table 4). Nearly all snow samples (6/8 or 75%) were able to produce oxidative stress as revealed by increased lipid peroxidation, regardless of site status (i.e., remote, rural or urban). Increased oxidative stress by (in)organic xenobiotics has been documented as an underlying mechanism leading to cellular damage dysfunction and death (Strmac and Braunbeck, 2002). Three sites were able to trigger cytochrome P4501A1 activity (Frelighsburg, St-Anicet, Montréal-Est), but snow collected from the Montréal-Est site proved to be the strongest inducer of EROD, suggesting the presence of polycyclic aromatic hydrocarbons (Navas and Segner, 2000). The latter site is under the influence of several petroleum refineries located east of the Island of Montreal. In such an effluent study, fractionated petroleum refinery products were previously shown to elicit 7-ethoxyresorufin O-deethylase in fish cell lines (Schirmer et al., 2001). Based on effluent C18 retention, the compounds were likely to be non-polar to moderately polar, and presumably associated with fine particles. Interestingly, three of the snow samples were able to increase the levels of alkali-labile phosphates and again snow collected from the Montréal-Est site gave the strongest response (Table 4). This suggests the presence of estrogen mimics in snow. The occurrence of estrogenic compounds in snow was also found in a previous study (Gagné et al., 1999a). Here, however, only plowed street snow samples were estrogenic as compared to undisturbed and clean snow collected in the present study. To our knowledge, this is the first study indicating that snow samples can elicit an estrogenic response in fish tissues. Yet, the presence of the endocrine disruptor 4-nonylphenol was detected in snow at a mean value of 0.24 µg dm-3 (Fries and Puttmann, 2004). A significantly higher mean level of 4-nonylphenol (0.48µg dm-3) was found in snow at urban sites, whereas in snow from suburban areas the concentration was 0.03 µg dm-3 which corroborates our findings as well. Four of eight snow extracts produced changes in IIb metals (i.e., zinc, cadmium and mercury) mobilization in trout hepatocytes (Table 4), which suggests exposure to metals in these snow samples. Indeed, even though our melted snow samples were extracted on a C18 column, organic matter can bind metals and allow it to be co-eluted (Gagné et al., 1999b). Increased antimony contamination in snow and ice samples from Devond Island in the Arctic was reported indicating that snow could represent a nonnegligible source of metals and metalloids (Krachler et al., 2005). In another study, the accumulation of cadmium and lead in snow and ice from the atmosphere was found to make up 40 % of the annual load in Bothnian Bay of the Baltic Sea (Granskog and Kaartokallio, 2004) further displaying the presence of metals in snow samples. Cluster analysis of (bio)analytical parameters revealed that effects on hepatocyte cell viability were more closely related to organic chemicals (cytochrome P4501A1 and vitellogenin-like protein (increases in the extracellular medium) than those related to heavy metal metabolism such as labile IIb metal (zinc) levels (Fig. 3). Although the reported snow Hg levels were closely related to hepatocyte viability (Fig. 3), concordance analysis revealed that snow Hg levels were not significantly correlated with either trout hepatocyte viability or changes in IIb metal levels (Kendall’s concordance = 0.33, p > 0.1 for each). However, given the paucity of compared data values available (n=3 for Hg concentrations in snow, see Table 1), confirmation with a greater sample size will be required in future to better comprehend the relationship between Hg in snow with cytotoxicity and the free Zn biomarker. Cluster analysis also indicates that H. attenuata appears sensitive to the presence in snow of estrogenic substances as suggested by its close proximity to the ALP biomarker in Fig. 3 and by concordance analysis (Kendall Τ value = 0.713, p < 0.1). Indeed, recent findings confirm the sensitivity of this hydrozoan toward several types of (xeno)estrogens. For one, it has been shown to be one of the most sensitive freshwater invertebrate species responding to acute and chronic toxicity of the estrogen, nonylphenol, a high production volume chemical used in synthesizing nonylphenol ethoxylates and in the production of resin plastics and stabilizers, with lowest observed effects based on lethal morphological criteria found to be close to 25 µg/L (Pachura et al., 2005). For another, a closely-related cnidarian, H. vulgaris, has also been found sensitive to the synthetic hormone 17β-estradiol and to bisphenol A, an
8 intermediate in the production of polycarbonate and epoxy resins, where structure and physiology of polyps were adversely affected at concentrations near (or under) the 100 µg/L range for both chemicals (Pascoe et al., 2002). Plasticizer metabolites in the environment are ubiquitous. Plasticizers such as di (2-ethylhexyl) phthalate (DEHP) or di-2-ethylhexyl adipate (DEHA) can be degraded by a common soil bacterium in the presence of an easily used carbon source. The degradation is not complete and resulted in the production of metabolites including 2-ethylhexanol,2-ethylhexanoic acid and monoesters such as 2-ethylhexyl phthalate. These compounds have been shown to exhibit acute toxicity using Microtox, Daphnia, rainbow trout and fathead minnow toxicity assays. Snow samples in Montreal have been shown to contain DEHP (130 ppb); DEHA (150 ppb) 2-ethylhexanol (0.0 ppb) 2-ethylhexanoic acid (6.7 ppb) (Horn et al., 2004). Occurrence of bisphenol A (BPA) and F (BPF) in the environment have been documented by Fromme et al. 2002. BPA measurements showed low concentrations from 0.0005 to 0.41 mg L -1 in surface water whereas measured concentrations of BPF were clearly lower than BPA in all environmental media.
4 Conclusions Our results have indicated that snow collected from either remote, rural or urban sites displayed effects suggesting the presence and bioavailability of (at least some) contaminants in snowpacks. This is established by toxicity responses observed with the H. attenuata sublethal test (22.2 to 100% effects over controls at sampled sites: Table 3) and with the rainbow trout primary hepatocyte (RTPH) cytotoxicity test (19 to 83% effects over controls at sampled sites: Table 3). Biomarkers measured with the RTPH test (Table 4) further intimate that snow from sampled sites harbours some quantities of mixed pollutants that are presumably inorganic (free Zn biomarker), organic (CYP 1A1 biomarker), estrogenic (ALP biomarker) or capable of causing oxidative stress (LPO biomarker). Collectively, these effects measurements certainly argue that snow from some of these sites could impact aquatic environments. This possibility would likely increase in situations of rapid spring melt-off, emission to small receiving systems and/or pollution loading (i.e., total volume of discharged snow). Snow indeed appears as a privileged medium for sequestering contaminants of atmospheric origin and future studies aimed at determining the chemicals (and their sources) implicated in noted effects are amply justified. . Acknowledgements: This study was funded by Environment Canada. Special thanks are extended to Conrad Beauvais and Michel Arseneau for snow sampling.
References Arkhipchuk, V., Malinovskaya, M. & Garanko, N. (2000). Cytogenetic study of organic and inorganic toxic substances on Allium cepa, Lactuca sativa and Hydra attenuata cells. Environmental Toxicology, 15, 338344. Blaise C. & Kusui T. (1997). Acute toxicity assessment of industrial effluents with a microplate-based Hydra attenuata assay. Environmental Toxicology and Water Quality, 12, 53-60. Blaise C. & Vasseur P. (2005). Algal Microplate Toxicity test. In C. Blaise & J.F. Férard (eds.), Small-scale Freshwater Toxicity Investigations, Volume 1, (pp. 137-179), Dordrecht, Springer Publishers. Carpenter, L. J., Hopkins, J.R., Jones, C.E., Lewis, A.C., Parthipan, R., Wevill, D.J., Poissant, L., Pilote, M. & Constant, P. (2005). Abiotic source of reactive organic halogens in the sub-Arctic atmosphere. Environmental Science and Technology, 39, 812- 8816. CCME (Canadian Council of Ministers of the Environment). (1999). Canadian water quality guidelines for the protection of aquatic life: benzene. In: Canadian environmental quality guidelines, 1999, Canadian Council of Ministers of the Environment, Winnipeg, Publication No. 1299 (pp. 1-4), ISBN 1-896997-34-1.
9
Constant P., Poissant, L., Villemur, R., Yumvihoze, E. & Lean, D. (2007). Fate of inorganic mercury and methyl mercury within the snow cover in the low arctic tundra on the shore of Hudson Bay (Québec, Canada). Journal of Geophysical Research D: Atmospheres 112 (8), art. no. D08309. Dann, T. (1994). Ambient Air Measurements of Benzene at Canadian Monitoring Sites (1987- 1993) Report Series No PMD 94-4 Environmental Technology Centre, Pollution Measurement Division, Environment Canada, Ottawa (Ontario) 55 p. Dann, T. (2000). Ambient Air Measurements of Benzene in Canada (1989-1998) in Proceedings of the Air & Waste Management Association’s 93rd Annual Conference & Exhibition. Salt Lake City, Utah, June 1822, 2000, 14 p. Environment Canada. (1992). Biological test method: toxicity test using luminescent bacteria (Photobacterium phosphoreum). Environmental Protection Series, Report EPS 1/RM/24, Environment Canada, Ottawa, 61 pages. Finne, E.F., Cooper, G.A., Koop, B.F., Hylland, K. & Tollefsen, K.E. (2007). Toxicogenomic responses in rainbow trout (Oncorhynchus mykiss) hepatocytes exposed to model chemicals and a synthetic mixture. Aquatic Toxicology, 81, 293-303. Fries, E. & Puttmann, W. (2004). Occurrence of 4-Nonylphenol in rain and snow. Atmospheric Environment, 38, 2013-2016. Fromme, H., Kuchler, T., Otto, T., Pilz, K., Muller, J. & Wenzel, A. (2002). Occurrence of phthalates and bisphenol A and F in the Environment. Water Research, 36, 1429-1438. Gagné, F. & Blaise, C. (1996). Available intracellular Zn as a potential indicator of heavy metal exposure in rainbow trout hepatocytes. Environmental Toxicology and Water Quality, 11, 319-325. Gagné, F., Pardos, M. & Blaise, C. (1999). Estrogenic effects of organic environmental extracts with the trout hepatocyte vitellogenin assay. Bulletin of Environmental Contamination and Toxicology, 62, 723-730. Gagné, F., Pardos, M., Blaise, C., Turcotte, P., Quémerais, B., Fouquet, A. (1999b). Toxicity evaluation of organic sediment extracts resolved by size exclusion chromatography using rainbow trout hepatocytes. Chemosphere, 39, 1545-1570. Gagné., F. (2005). Acute toxicity assessment of liquid samples with primary cultures of rainbow trout hepatocytes. In C. Blaise & J.F. Férard (eds.), Small-scale Freshwater Toxicity Investigations, Volume 1, (pp. 453-472), Dordrecht, Springer Publishers. Germain A., Rousseau, J. & Dann, T. (2001). Issues Related to Benzene in Eastern Montreal, Environment Canada, Catalogue number: EN40-644/2001E-IN, ISBN: 0-662-30990-1, 16p. Horn, O, Nalli, S., Cooper,_D. & Nicell, J. (2004). Plasticizer metabolites in the environment, Water Research, 38, 3693-3698. Krachler, M., Zheng, J., Koerner, R., Zdanowicz, C., Fisher, D. & Shotyk, W. (2005). Increasing atmospheric antimony contamination in the northern hemisphere: Snow and ice evidence from Devon Island, Arctic Canada. Journal of Environmental Monitoring, 7, 1169-1176. Garbarino, JR, Snyder-Conn, E., Leiker, TJ. & Hoffman, GL. (2002). Contaminants in arctic snow collected over northwest Alaskan sea ice. Water Air and Soil Pollution, 139, 183-214.
10 Granskog, M.A. & Kaartokallio, H. (2004). An estimation of the potential fluxes of nitrogen, phosphorus, cadmium and lead from sea ice and snow in the northern Baltic Sea. Water Air and Soil Pollution, 154, 331-347. Hagen, A. & Langeland, A. (1973). Polluted snow in Southern Norway and the effect of the meltwater on freshwater and aquatic organisms. Environmental Pollution, 5, 45-57. Herbert, B.M.J., Villa, S. & Halsall, C.J. (2006). Chemical interactions with snow: understanding the behavior and fate of semi-volatile organic compounds in snow. Ecotoxicology and Environmental Safety, 63, 3-16. Holdway, D. (2005). Hydra population reproduction toxicity test method. In C. Blaise & J.F. Férard (eds.), Small-scale Freshwater Toxicity Investigations, Volume 1, (pp. 395-412), Dordrecht, Springer Publishers. Kennedy, S.W., Lorenzen, A., James, C.A., & Collins, B.T. (1993). Ethoxyresorufin-O-deethylase and porphyrin analysis in chicken embryo hepatocyte cultures with fluorescence multiwell plate reader. Analytica Biochemica, 211, 102-112. Lahoutifard, N., Poissant, L., Scott, S.L. (2006). . Science of the Total Environment, 355 (1-3) 118-126. Lindberg, S. E., Brooks, S., Lin, C.J., Scott, K.J., Landis, M.S., Stevens, R.K., Goodsite, M. & Richter, A. (2002). Dynamic oxidation of gaseous mercury in the Arctic troposphere at polar sunrise. Environmental Science and Technology, 36, 1245-1256. Marsalek, J., Rochfort, Q., Bownlee, B., Mayer, T. & Servos, M. (1999). An exploratory study of urban runoff toxicity. Water Science Technology, 39, 33-40. Navas, J.M. & Segner, H. (2000). Antiestrogenicity of beta-naphthoflavone and PAHs in cultured rainbow trout hepatocytes: evidence for a role of the arylhydrocarbon receptor. Aquatic Toxicology, 51, 79-92. Oberts, G.L., Marsalek, J. & Viklander, M. (2000). Review of water quality impacts of winter operation of urban drainage. Water Quality Research Journal of Canada, 35, 781-808. Pachura, S., Cambon, J.P., Blaise, C. & Vasseur, P. (2005). 4-nonylphenol induced toxicity and apoptosis in Hydra attenuata. Environmental Toxicology and Chemistry, 24, 3085-3091. Pardos, M., Benninghof, C., Gueguen, C. Thomas, R., Dobrowolski, J, & Dominik, J. (1999). Acute toxicity assessment of Polish (waste) water with a microplate-based Hydra attenuata assay : a comparison with the Microtox test. Science of the Total Environment, 243/244, 141-148. Pascoe, D., Carroll, K., Karntanut, W. & Watts, M. (2002). Toxicity of 17β-estradiol and bisphenol A to the freshwater cnidarian Hydra vulgaris. Archives of Environmental Contamination and Toxicology, 43, 56-63. Persoone, G., Marsalek, B., Blinova, I., Törökne, A., Zarina, D., Manusadzianas, L., Nalecz-Jawecki, G., Tofan, L., Stepanova, N., Tothova, L. & Kolar, B. (2003). A practical and user-friendly classification system with microbiotests for natural waters and wastewaters. Environmental Toxicology, 18, 395-402. Poissant, L. (1997). Field observations of total gaseous mercury behaviour: Interactions with ozone concentration and water vapour mixing ratio in air at rural site. Water Air and Soil Pollution, 97, 341- 353. Poissant, L., Dommergue, A. & Ferrari, C.P. (2002). Mercury as a global pollutant. Journal of Physics, IV, 12, 143- 160.
11 Poissant, L. & Hoënninger, G. (2004). Atmospheric mercury and ozone depletion events observed at the Hudson Bay in northern Québec along to BrO (DOAS) measurements, RMZ Mater. Geoenvironment, 51, 1722-1725. Poissant, L., Schmit, J.-P. & Béron, P. (1994). Trace inorganic elements in rainfall in the Montreal Island Atmospheric Environment, 28, 339-346. Quémerais, B. & Cossa, D. (1997). Procedures for sampling and analysis of mercury in natural waters. Science and Technology Report ST-31E, 34 pp., St. Lawrence Centre, Environment Cananda, Montréal, Québec. Schemenauer, R.S., Berry, M.O. & Maxwell, J.B. (1981). Snowfall formation. In : D.M. Gray & D.H. Male (eds.), Handbook of snow. Principles, Processes, Management and Use, (pp. 129-151), Toronto, Pergamon Press. Schirmer K., Tom D.J. Bols N.C. & Sherry J.P. (2001). Ability of fractionated petroleum refinery effluent to elicit cyto- and photocytotoxic responses and to induce 7-ethoxyresorufin-O-deethylase activity in fish cell lines. Science of the Total Environment, 271, 61-78. Schroeder, W.H., Anlauf, K.G., Barrie, L.A., Lu, J.Y., Steffen, A., Schneeberger, D.R. & Berg, T. (1998). Arctic springtime depletion of mercury. Nature, 394, 331-332. Simonetti, A., , C., Carignan, J. & Poissant, L. (2000). Isotopic evidence of trace metal sources and transport in eastern Canada as recorded from wet deposition, Journal of Geophysical Research D: Atmospheres, 105 (D10), 12263-12278. Strmac, M., Braunbeck, T. (2002). Cytological and biochemical effects of a mixture of 20 pollutants on isolated rainbow trout (Oncorhynchus mykiss) hepatocytes. Ecotoxicology and Environmental Safety, 53, 293-304.
12
Table 1 Characteristics of snow-sampled sites Site
Sampling Date & time
GPS location
Site description and geographical location
Sources of atmospheric pollution likely to impact site
Kuujjuarapik, Qc Pre-MDE
11 March 04 10:00
55.28N 77.74W
Remote, 1100 Km North of Montréal
Kuujjuarapik, Qc Post-MDE
16 March 04 19 :00
55.28N 77.74W
Remote, 1100 Km North of Montréal
Frelighsburg, Qc
22 March 04 13 :10
45.05N 72.86W
Rural, 75 Km S-E Montréal
Electric Diesel power plant Pre-MDE (mercury depletion event) Electric Diesel power plant Post-MDE (mercury depletion event) Long range Air transport from USAOntario
St-Anicet, Qc
22 March 04 10 :10
45.12N 74.29W
Rural, 20 Km S-W Valleyfield and 70 Km from Montréal
Riv. des Prairies, Qc
22 March 04 09:45
45,65N 73,57W
Urban, N-E Montréal
Varennes, Qc
22 March 04 11:05
45,72N 73,38W
Rural, à 15 Km au N-E de Montréal
Ile Charron, Qc
22 March 04 10:30
45,59N 73,48W
Sub-urban, N-E Montréal
Zinc electrolytic plant at Valleyfield + Montréal urban plume + Long range Air transport from USA-Ontario N-E from Petro chemical plants + Montréal urban plume N-E from Petro chemical plants + Montréal urban plume Regional park under automotive impact
Approximate snow fall for 2004-05 winter season 2004-05 : N/A Normal : 241.3 cm 2004-05 : N/A Normal : 241.3 cm 2004-05 : 234.7 cm Normal : 281.4 cm (Sutton) 2004-05 : 171.9 cm Normal : 176.1 cm
Total Gaseous Mercury (TGM)
Hg in snow surface or precipitation
2.23±0.09 ng m-3
Snow surface: 8.04 ng L-1
0.88±0.38 ng m-3
Snow surface: 18.02 ng L-1
N/A
N/A
1.90±0.16 ng m-3
Precipitation : 5.66: ng/L or 50.72 ng/m2
2004-05 : 313.2 cm Normal : 169.3 cm 2004-05 : 299.0 cm Normal : 203.8 cm (Verchères) 2004-05 : 313.1 cm
N/A
N/A
N/A
N/A
N/A
N/A
13
(highways + Montreal urban plume) Montréal-Est, Qc
22 March 04 09:15
45,63N 73.51W
Urban, 1 km from Industrial sector
N-E from Petro chemical plants + Montréal urban plume
Table 2 Characteristics of the small-scale bioassays used to determine the toxic potential of snow Trophic level Toxicity test Assessment endpoint
Normal : 220.5 cm (Montréal/StHubert) 2004-05 : 313.2 Normal : 169.3 cm (Riv. Des Prairies)
Measurement endpoint
N/A
N/A
Reference
Decomposer
Bacterial test Vibrio fischeri (Microtox® toxicity test)
Acute sublethal light inhibition
15min-IC50
Environment Canada, 1992
Primary producer
Algal test (Pseudokirchneriella subcapitata microplate assay)
Chronic sublethal growth inhibition
72h-IC50
Blaise and Vasseur, 2005
Primary consumer
Micro-crustacean Thamnocephalus platyurus test (ThamnoToxkit assay)
Acute lethality
24h-LC50
Microbiotests Inc., http://www.microbiotests.be/
Secondary consumer
Cnidarian test
Acute sublethality indicated by morphological changes
96h-LC and EC50
(Hydra attenuata assay)
Blaise and Kusui, 1997
Secondary Acute cytotoxicity 48h-TECa Gagné, 2005 Fish cell test (rainbow trout primary consumer hepatocyte test) a) TEC (Threshold Effect Concentration) for cytotoxicity as manifested by a significant reduction in cell viability = (NOEC x LOEC)1/2, where NOEC = no observed effect concentration and LOEC = lowest observed effect concentration.
14
Table 3 Bioassay results showing toxic effects (in %) observed at the 100% (neat) concentration of melted snow and the corresponding hazard for each site based on a classification system proposed by Persoone et al. (2003) Station Microtox Algal test Hydra test T. platyurus Hepatocytes Hazardous class and (sublethality) class weight score in % % effect for each toxicity test (and corresponding test score) Kuujjuarapik Pre13.4 (0) -35 (0) 27.3 (1) 6.7 (0) 19 (0) II ((20%) MDE Kuujjuarapik Post10.1 (0) 3.6 (0) 44.4 (1) 0 (0) 83 (2) III (30%) MDE Frelighsburg 12.4 (0) -116 (0) 22.2 (1) 0 (0) 25 (1) II (40%) St-Anicet 3.9 (0) -32 (0) 54.6 (2) 3.3 (0) 20 (1) III (30%) Riv. des Prairies 13.7 (0) -107 (0) 30 (1) 0 (0) 20.5 (1) II (40%) Varennes 10.8 (0) -20.8 (0) 50 (2) 0 (0) 34 (1) III (30%) Ile Charron 13.7 (0) -40.6 (0) 40 (1) 3.3 (0) 26 (1) II (40%) Montréal-Est 12.3 (0) -8.8 (0) 100a (3) 0(0) 54 (2) IV (33%)b a) The Montréal-Est snow sample was the only one demonstrating clear 50% effects responses (with associated 95% confidence intervals) for the Hydra assay, where EC50 = 12.8 (8.4 – 18.1) and LC50 = 70.7 (50 – 100). b) For the Montréal-Est site, the calculation of class weight score as a percentage is as follows: first, the class weight score is (0 + 0 + 3 + 0 + 2)/5 = 1 (see explanation in section 2.3). Then, the class weight score as a percentage is (1/3) x 100 = 33%.
Table 4 Biomarker results showing effect-fold increases for Rainbow Trout Primary Hepatocytes exposed to a 100% (neat) melted snow concentration at each site over controls Station Lipid Peroxidation Free Zn CYP 1A1 Alkali labile phosphate Kuujjuarapik Pre2.21 1.31 0a 0 MDE Kuujjuarapik Post2.74 1.29 0 1.20 MDE Frelighsburg 0 0 2.71 0 St-Anicet 2.83 0 1.23 1.26 Riv. des Prairies 3.65 1.66 0 0 Varennes 2.42 0 0 0 Ile Charron 0 2.32 0 0 Montréal-Est 1.15 0 4.30 1.41 a) All "0" values indicate effect-fold increases that were ≤ 1.10 and thus insignificant.
15
16
Kuujjuarapik
Montréal
Snow Ecotoxic Pote ntial Proje ct Samping station
Varenne Rivière des Prairies
Montréal Est
Outaouais River
Ile Charron
Montréal
Ontario
Que be c
St.Law rence River St-Anicet 0
10
Kilometers
USA
Fig. 1 Sites in Québec, Canada, selected for snow collection
20
Frelighsburg
17
Fig. 2 Photo of snow sampling. The snow sample (top 5 cm) was collected using clean protocol and methods (see Materials and Methods) and placed in 2L Teflon bottles for ensuing analyses
18
Hydra
ALP
CYP1A1
PHRT via
Hg
Zn
LPO
0,0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
Relative distance (1- Pearson's correlation coefficient) Fig. 3 Cluster analysis of bioassay, biomarker and Hg parameters for the snow sites investigated. Hydra: H. attenuata sublethal endpoint; ALP: alkali labile phosphates of vitellogenin-like proteins in the primary hepatocyte rainbow trout (PHRT) bioassay; CYP1A1: cytochrome P450 1A1 induction of EROD in the PHRT bioassay; PHRT via: cytotoxicity endpoint (viability) in the PHRT bioassay; Hg: measured concentrations of mercury in snow samples; Zn: labile zinc levels in PHRT bioassay; LPO: lipid peroxidation levels (thiobarbituric acid reactants) in the PHRT bioassay
