Ecotoxicology and Environmental Safety 95 (2013) 104–112

Contents lists available at SciVerse ScienceDirect

Ecotoxicology and Environmental Safety

0147-65http://d

AbbreOrganic(PWE),Osmosis

n CorrE-m

dub.mo

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

Influence of elevated alkalinity and natural organic matter (NOM)on tissue-specific metal accumulation and reproductive performancein fathead minnows during chronic, multi-trophic exposures to ametal mine effluent

Jacob D. Ouellet a,n, Monique G. Dubé b, Som Niyogi a

a Department of Biology, University of Saskatchewan, 112 Campus Drive, Saskatoon, SK, Canada S7N 5Eb Canadian Rivers Institute, University of New Brunswick, NB, Canada

a r t i c l e i n f o

Article history:Received 2 April 2013Received in revised form16 May 2013Accepted 17 May 2013Available online 19 June 2013

Keywords:Metal mine effluentWater chemistry modificationAlkalinityNatural organic matterMetal bioaccumulationEgg production

13/$ - see front matter & 2013 Elsevier Inc. Alx.doi.org/10.1016/j.ecoenv.2013.05.027

viations: Metal Mine Effluent (MME), Biotic LMatter (NOM), Dissolved Organic Carbon (DOEnvironmental Effects Monitoring (EEM); Refe(RO).esponding author. Fax: +1 306 966 4461.ail addresses: jake.ouellet@usask.ca (J.D. Ouelln@hotmail.com (M.G. Dubé), som.niyogi@usas

a b s t r a c t

Metal bioavailability in aquatic organisms is known to be influenced by various water chemistry parameters.The present study examined the influence of alkalinity and natural organic matter (NOM) on tissue-specificmetal accumulation and reproductive performance of fathead minnows (Pimephales promelas) duringenvironmentally relevant chronic exposures to a metal mine effluent (MME). Sodium bicarbonate (NaHCO3)or NOM (as commercial humic acid) were added to a Canadian MME [45 percent process water effluent(PWE)] in order to evaluate whether increases in alkalinity (3–4 fold) or NOM (�1.5–3 mg/L dissolvedorganic carbon) would reduce metal accumulation and mitigate reproductive toxicity in fathead minnowsduring a 21-day multi-trophic exposure. Eleven metals (barium, boron, cobalt, copper, lithium, manganese,molybdenum, nickel, rubidium, selenium, and strontium) were elevated in the 45 percent PWE relative tothe reference water. Exposure to the unmodified 45 percent PWE resulted in a decrease of fathead minnowegg production (�300 fewer eggs/pair) relative to the unmodified reference water, over the 21-day exposureperiod. Water chemistry modifications produced a modest decrease in free ion activity of some metals(as shown by MINTEQ, Version 3) in the 45 percent PWE exposure water, but did not alter the metal burdenin the treatment-matched larval Chironomus dilutus (the food source of fish during exposure). The tissue-specific metal accumulation increased in fish exposed to the 45 percent PWE relative to the reference water,irrespective of water chemistry modifications, and the tissue metal concentrations were found to be similarbetween fish in the unmodified and modified 45 percent PWE (higher alkalinity or NOM) treatments.Interestingly however, increased alkalinity and NOM markedly improved fish egg production both in thereference water (�500 and �590 additional eggs/pair, respectively) and 45 percent PWE treatments (�570and �260 additional eggs/pair, respectively), although fecundity over 21 day exposure consistently remainedlower in the 45 percent PWE treatment groups relative to the treatment-matched reference groups.Collectively, these findings suggest that metal accumulation caused by chronic 45 percent PWE exposurecannot solely explain the reproductive toxicity in fish, and decrease in food availability (decrease in C. dilutusabundance in 45 percent PWE exposures) might have played a role. In addition, it appears that NaHCO3 orhumic acid mitigated reproductive toxicity in fish exposed to 45 percent PWE by their direct beneficial effectson the physiological status of fish.

& 2013 Elsevier Inc. All rights reserved.

1. Introduction

Metal mines release effluents that can be a significant source ofa variety of metals in receiving waterbodies. Although metal

l rights reserved.

igand Model (BLM); NaturalC); Process Water Effluentrence Water (RW), Reverse

et),k.ca (S. Niyogi).

concentrations are often elevated in areas that receive metal mineeffluents (MMEs), the metals in those ecosystems might not beavailable or cause toxicity to resident biota. Currently, risk-assessment models that are used to predict metal bioavailabilityand toxicity in aquatic organisms [e.g., Biotic Ligand Model (BLM)]are mainly focused on assessing toxicity of single metals (Di Toroet al., 2001; Paquin et al., 2000; Santore et al., 2001), and do notinclude the effects of metal mixtures. The BLM incorporates theinfluence of water chemistry parameters, such as water hardness,pH, alkalinity, and dissolved organic carbon (DOC), on the avail-ability and binding of free metal ions to the biotic ligand (e.g., fishgill), and predicts toxicity based on the critical accumulation of

J.D. Ouellet et al. / Ecotoxicology and Environmental Safety 95 (2013) 104–112 105

metal(s) on the biotic ligand (Niyogi and Wood, 2004). However,there is a general lack of understanding of how the waterchemistry parameters influence metal bioavailability, bioaccumu-lation and toxicity in chronic conditions, particularly duringexposure to complex metal mixtures, such as MMEs.

Water chemistry is likely to influence the toxicity of MMEs toaquatic organisms. In addition to a mixture of metals, MMEs oftenalso contain high concentrations of various ions (e.g., Ca2+, Mg2+

and Na+, and Cl− and SO4−) (Dubé et al., 2005; Ouellet et al., 2013a,

2013b; Rozon-Ramilo et al., 2011a), which can strongly influencethe bioavailability of metals. For example, hardness cations Ca2+

and Mg2+ can reduce metal bioavailability through competitionwith free metal ions, whereas natural anions such as Cl− and SO4

−

can also decrease metal bioavailability through complexation offree metal ions (Niyogi and Wood, 2004). On the other hand,MMEs, as well as many metal contaminated lakes in the CanadianShield, generally have relatively low pH/alkalinity (low CO3

−/HCO3

−) and DOC concentrations (Dubé et al., 2005; Ouellet et al.,2013a, 2013b; Pyle et al., 2005; Rozon-Ramilo et al., 2011a), whichcan contribute to increased bioavailability of metals. During acuteexposure to single metals, metal complexation associated withincreased alkalinity or DOC is known to reduce metal bioavail-ability and toxicity to aquatic organisms (Pagenkopf et al., 1974;Playle et al., 1993; Schwartz et al., 2004). Similar protective effectsof DOC have also been reported in fish exposed to short-termexposures to metal-mixtures (Richards et al., 2001), although theevidence is sporadic. Importantly though, the water chemistryparameters, which are known to reduce acute metal toxicity, havebeen reported to be either less effective or ineffective duringchronic exposure to metal(s) (Brauner and Wood, 2002a, 2002b;De Schamphelaere and Janssen, 2004a, 2004b). Moreover, it islargely unknown whether water chemistry parameters can influ-ence metal bioavailability and toxicity to fish during chronicexposure to metal mixtures. Recent evidence indicates no appar-ent interrelationship among the free metal ion activity in theexposure water, and bioaccumulation and chronic toxicity in fishexposed to metal mixtures (Kamunde and MacPhail, 2011). Expo-sure to metals in mixture is likely to elicit competitive or additiveinteractions, considerations that are not included in the currentBLM and therefore limit its ability to predict toxicity of metals incontaminated aquatic ecosystems, where organisms are almostalways exposed to metals in mixture (Borgmann et al., 2008). Inaddition, fish inhabiting the metal-contaminated environmentsare exposed to metal mixtures via both water and diet (Wang,2011), and the latter pathway is not included in the current BLMs.

It is apparent that further investigations are needed to betterunderstand the influence of water chemistry on the bioavailabilityof metals to fish during chronic exposure to MMEs. This isimportant since MMEs have been found to elicit a variety of toxiceffects to fish in both field-based and laboratory studies, and theeffects included delayed development and larval deformities(Driessnack et al., 2011; Jezierska et al., 2009), decreased lipidstorage and growth (Bennett and Janz, 2007), as well as decreasedfecundity (Driessnack et al., 2011; Franssen, 2009; Rozon-Ramiloet al., 2011a, 2011b), and behavioral changes (Gerhardt, 1998;McPherson et al., 2004). Generally, the effects observed in fishexposed to MMEs are believed to be induced by the accumulationof metals in target organs (e.g., gill, liver and gonad), however acausal relationship between metal bioaccumulation and toxicityhas not been clearly established in any previous studies.

The main objective of this study was to examine whetherincreased alkalinity (achieved by adding NaHCO3) or naturalorganic matter (NOM) (added as humic acid) ameliorate tissue-specific metal accumulation and reproductive toxicity in a modelfish species, the fathead minnow (Pimephales promelas), duringenvironmentally relevant chronic exposure to an MME. Since both

alkalinity and NOM are believed to reduce metal bioavailability byincreased complexation of free metals in the exposure water, wehypothesized that increased NaHCO3 or humic acid content wouldcause a decrease of metal accumulation in target organs andprotect against the MME-induced reproductive impairment infathead minnows. The MME used in the present study was processwater effluent (PWE), which has been found to cause metalaccumulation in benthic invertebrate and fish tissues, and impairreproductive output of fish in a number of previous studies(Ouellet et al., 2013a, 2013b; Rickwood et al., 2006; Rozon-Ramilo et al., 2011a, 2011b). PWE is released at concentrations ofapproximately 51,000,000 m3/yr (2009) into the Junction CreekWatershed, near Sudbury, Ontario, Canada, and is diluted to aconcentration of approximately 45 percent in the receiving envir-onment, based on discharge and stream conditions (Rickwoodet al., 2006).

2. Materials and methods

2.1. Multi-trophic mesocosms

This research was performed at the University of Saskatchewan in Saskatoon,SK, Canada, from January to March 2011. Multi-trophic mesocosms with 6 replicates(10.3-L circular polyethylene streams) were used to mimic natural stream systems.Detailed descriptions of the multi-trophic systems can be found in Rickwood et al.(2006) and Hruska and Dubé (2004). Each replicate stream was made up of asediment layer (�2.5 cm of pre-cleaned silica sand), a feeding barrier for providingconsistent numbers of Chironomus dilutus as food to fathead minnows duringexposure, a spawning tile, and a mesh-screen for preventing adult C. dilutus andfathead minnows from escaping the streams. Water was exchanged at a rateof one turnover/day into an 85-L reservoir, which evenly recirculated the treatmentwater to each of the mesocosm streams via a March pump (Model LC-3CP-MD,March Manufacturing, Glenview, IL, USA). Streams were aerated with air-stonesand heated to 2572 1C with submersible aquarium heaters under conditionsof 16 h light:8 h dark photoperiod. The mesocosm system allowed for bothwater-borne and dietary exposures to the MME, where daily egg production andtissue-specific metal accumulation in fathead minnows were primary endpoints.The bioaccumulation of metals from MMEs to target organisms and tissues(i.e., C. dilutus, fathead minnow liver, gonad, gill, and carcass tissues) have beenevaluated using these mesocosms in many regions, including Ontario (Dubé et al.,2006; Hruska and Dubé, 2004, 2005; Rickwood et al., 2006; Rozon-Ramilo et al.,2011a, 2011b), New Brunswick (Dubé et al., 2005), Saskatchewan (Driessnacket al., 2011), and the Northwest Territories (Spencer et al., 2008). The 21-daymulti-trophic mesocosms used in our study are an accepted component ofEnvironment Canada’s Environmental Effects Monitoring (EEM) program. Animaluse was approved by the University of Saskatchewan Committee on Animal Careand Supply, and Animal Research Ethics Board.

2.2. Process water effluent and reference water

The effluent used in this study was the process water effluent (PWE) from anOntario metal mine (Vale Canada Limited). This effluent was chosen because it isdischarged at high volumes (�51,000,000 m3/yr in 2009 in the Junction Creekwatershed) and has been shown to cause significant decreases in cumulative eggproduction of fathead minnows in several studies (Ouellet et al., 2013a, 2013b;Rickwood et al., 2006; Rozon-Ramilo et al., 2011a, 2011b). Process water effluentexposures were performed at a concentration of 45 percent dilution based onenvironmentally relevant concentrations in the Junction Creek stream watershed.Effluent was shipped weekly from Sudbury, Ontario to the University of Saskatch-ewan. For the present study, reference water (RW) was made up of a mixture ofreverse osmosis (RO) water and dechlorinated laboratory water at concentrationsof approximately 65 percent RO and 35 percent laboratory water in order to matchthe water chemistry conditions (pH, alkalinity, hardness, background metal con-centrations) of a previously used reference river site (Vermillion River), locatednear the Junction Creek Watershed (Rozon-Ramilo et al., 2011a). The samereference water was also used as dilution water for the 45 percent PWE.

2.3. Treatments

This study consisted of six treatment groups (see Table 1): (i) Reference Water(RW), (ii) RW+increased alkalinity, (iii) RW+increased DOC, (iv) 45 percent PWE,(v) 45 percent PWE+increased alkalinity, and (vi) 45 percent PWE+increased DOC.In order to increase the alkalinity of the respective RW or 45 percent PWE

Table 1Summary of general water quality (WQ) measurements and dissolved metal concentrations from unmodified and modified (increased alkalinity and increased DOC)reference water (RW) and 45 percent process water effluent (PWE) treatments.

Parameter

General WQ RW RW+inc. alk. RW+inc. DOC 45 % PWE 45 % PWE+inc. alk. 45 % PWE+inc. DOC

Alkalinity mg/L 32.270.8ab 84.473.6c 37.071.4b 23.672.5ae 103.874.3d 19.872.9e

pH pH 7.770.0a (7.6) 8.170.0b (8.1) 7.870.0a (7.6) 7.270.1c (6.9) 8.170.0b (8.1) 7.170.1c (6.9)DOC* mg/L 4.070.2a (3.0) 4.070.4a (3.0) 6.470.1b (6.0) 5.070.3a (3.0) 4.970.2a (3.0) 6.270.2b (6.0)Temperature 1C 23.970.0a 24.170.0a 24.770.0a 24.470.2a 23.870.2a 24.370.1a

Ammonia mg/L 0.1670.10a 0.8070.27ab 0.8970.27ab 2.7070.51c 1.7470.47bc 1.0870.44abc

Chloride mg/L 3.0770.31a 3.2670.39a 4.7371.69a 37.8770.41b 37.8770.32b 36.6770.50b

Conductivity mg/L 146.672.6a 234.575.5b 154.772.7a 1590.0716.8c 1617.4715.0c 1579.778.1c

Nitrate mg/L 0.4570.33a 0.1970.03a 0.2470.12a 0.5070.00a 0.5070.00a 0.5070.00a

Sodium mg/L 9.6470.47a 32.2370.92b 10.3770.72a 64.1771.74c 113.6777.51d 67.0772.07c

Sulfate mg/L 25.0374.91a 26.9371.98a 26.3073.41a 663.00744.84b 665.00721.28b 644.67748.20b

Total Hardness (as CaCO3) mg/L 49.676.3a 48.473.1a 43.774.1a 700.3723.2b 644.7743.1b 707.7749.15b

Calcium mg/L 12.8572.53a 12.1571.52a 10.9371.79a 237.33710.81b 214.67715.45b 241.67721.28b

Magnesium mg/L 4.2670.14a 4.4170.43a 4.0070.32a 26.1371.34b 26.3771.43b 25.2371.52b

Barium mg/L 7.7370.41a 9.9072.08a 7.5770.55a 29.9371.30b 26.4070.97b 30.3771.29b

Boron mg/L 23.7370.78a 23.9071.25a 23.4372.09a 69.3071.17b 59.9772.85b 68.4773.24b

Cadmium mg/L 0.2870.23a 0.0770.02a ≤0.0570.00a 0.1970.04a 0.0770.02a 0.1370.01a

Cobalt mg/L ≤0.5070.00a ≤0.5070.00a ≤0.5070.00a 3.4570.76b 2.1970.10b 3.0870.61 b

Copper mg/L 11.2071.87a 9.6071.82a 6.8372.74a 56.8773.79b 49.4076.07b 23.0072.25c

Lithium mg/L 4.4370.99a 3.4770.97a 3.4070.90a 30.6771.76b 28.0072.00b 32.6771.86b

Manganese mg/L 0.9370.43a 1.4370.93a 0.8370.33a 30.20712.88b 4.3770.32ab 25.33711.87b

Molybdenum mg/L ≤0.5070.00a ≤0.5070.00a ≤0.5070.00a 3.6770.38b 3.1370.12b 3.5370.23b

Nickel mg/L 2.5370.33a 1.9070.15a 2.4070.10a 77.3375.88b 61.6773.84b 80.27710.18b

Rubidium mg/L ≤0.5070.00a ≤0.5070.00a 0.7070.20a 31.3771.35bc 29.1770.76b 32.6770.44c

Selenium mg/L ≤0.5070.00a ≤0.5070.00a ≤0.5070.00a 7.2770.48b 6.6770.38b 8.2370.58b

Strontium mg/L 80.87712.88a 77.6077.58a 71.60710.20a 600.67722.98b 541.00741.33b 598.00746.92b

Thallium mg/L ≤0.0570.00a ≤0.0570.00a ≤0.0570.00a 0.0870.03a 0.0970.02a 0.0970.02a

Zinc mg/L 14.1375.36a 20.0078.89a 15.0076.34a 20.3375.78a 13.3373.61a 18.4074.46a

Data is presented as mean7S.E.; n¼21 for pH, alkalinity, temperature, ammonia, and conductivity, and n¼3 for all remaining parameters.Data in parentheses represent nominal values. Means that do not share letters are statistically different from each other (ANOVA; Tukey's HSD posthoc test, po0.05).

n Dissolved organic carbon.

J.D. Ouellet et al. / Ecotoxicology and Environmental Safety 95 (2013) 104–112106

treatments, sodium bicarbonate (NaHCO3; 99 percent) (Alfa-aesar, Heysham, Lancs,UK, 99 percent) was added until a pH of 8.1 was achieved. Humic acid (as sodiumsalt) (Alfa-aesar, Ward Hill, MO, USA, 50–60 percent) was added to the respectiveRW or 45 percent PWE treatments in order to increase the DOC by a nominalconcentration of 3 mg/L. We chose a modest increase in DOC level in our exposuresbecause of the low solubility of commercial humic acid used in our experiment.Depending on the treatment, NaHCO3 or humic acid was added to RWor 45 percentPWE in 330 L polyethylene holding tanks, and allowed to reach chemical equili-brium over 24 h. During equilibration, the exposure waters were stirred constantlywith a Stir-Paks mixer (Cole-Parmer, Montreal, QC, Canada) to facilitate solubility.Subsequently, the respective exposure water was pumped (Pulsatron Series E,Viking Pump of Canada, Edmonton, AB, Canada) into the multi-trophic mesocosms.

2.4. Trophic-transfer system

C. dilutus larvae were cultured directly in each of the six reference andtreatment streams in order to provide a dietary exposure to fathead minnows inaddition to the waterborne exposure. Egg sacs of C. dilutus were isolated fromlaboratory held brood stocks every seven days for three weeks, and added directlyto the replicate streams to provide two eggs sacs per stream (equivalent to 1 g/dayof food for the duration of the 21-day exposure period) (Rickwood et al., 2006;Rozon-Ramilo et al., 2011a). During the C. dilutus culturing period, water exchangeswere performed at a rate of half exchanges every second day, except during the firstweek of culturing when the first water exchange was carried out on day 4, tominimize disturbances while C. dilutus cultures were being established.

2.5. Pre-exposure period

The pre-exposure period was performed in order to ensure all pairs of fatheadminnows selected for the exposure period were capable spawners, as well as toprovide similar reproductive potential for each replicate stream. Six to nine-monthold fathead minnows were obtained from Osage Catfisheries Inc. (Osage Beach, MO,USA). On day-0 of the pre-exposure, body weight, total length, and secondary sexcharacteristics (banding, nuptial tubercles, dorsal pad, fin dot, and ovipositor) wererecorded for each fathead minnow (83 pairs). Fathead minnows were placedrandomly into each stream until one male and one female were present in each(one breeding pair/stream). Water exchanges were fixed at one turnover/day.

Breeding pairs were fed frozen bloodworms (Sally’s bloodwormsTM, San FranciscoBay Brand, Inc., Newark, CA, USA) twice daily at a feeding rate of approximately1 g/day. Egg production was monitored daily for seven days by removing thebreeding tile from each stream, scraping eggs onto a petri dish, and photographingthe eggs with a Canon Powershot A620 digital camera mounted to a Vista VisionTM

(Model 48402-00, VWR International, Mississauga ON, Canada). Breeding pairs thatspawned at least once in the pre-exposure period with 480 percent fertilizationsuccess were used in the exposure period. This consisted of thirty-six pairs beingdistributed among the six treatments (n¼6 for each treatment). Infertile eggs wereeither opaque, had a visibly precipitated yolk, or contained no yolk (Ankley et al.,2001). One-way ANOVA was performed on the pre-exposure egg production (totalnumber of eggs, eggs/female/day, and breeding attempts) to verify that there wereno significant differences between egg production for fathead minnows in each ofthe treatments (α¼0.05, n¼6). Breeding pairs were distributed randomly to astream within the treatment.

2.6. Exposure period

Fathead minnows were exposed to the treatments for 21 days. Daily waterquality measurements were performed at the University of Saskatchewan. Tem-perature, dissolved oxygen, conductivity [YSI meter (Yellow Springs Instruments,Yellow Springs, OH, USA)], ammonia (Rolf C. Hagen, Edmonton, AB, Canada), pH(Oakton pHTestr, Oakton Instruments, Vernon, IL, USA), and alkalinity (LaMotteCompany, Chestertown, MD, USA) were each measured from one stream, randomly,per treatment. Weekly water samples were taken on days 7, 14, and 21 from eachtreatment, randomly, in pre-labeled high density polyethylene sample bottles.Samples were taken directly from a single stream which was chosen randomlyfrom each mesocosm table, filtered into the sample bottles with a 0.45 mm celluloseacetate filter (Corning Incorporated, Corning, NY, USA), sealed in a ziplock bag, andshipped out to Testmark Laboratories (Sudbury, Ontario, Canada) in a cooler chilledwith ice for further analysis. These water samples were analyzed for dissolvedmetals using inductively coupled plasma-mass spectrometry (ICP-MS). Similarly,Ca2+ and Mg2+ concentrations were also measured using ICP-MS without acidifica-tion, and converted to total water hardness [expressed as the equivalent of calciumcarbonate (CaCO3)]. In addition, anions were analyzed by Ion Chromatography, anddissolved organic carbon (DOC) was measured using a Dohrman total organiccarbon (TOC) analyzer. Minimum detection limits for DOC and TOC were 0.4 mg/L.All of these water quality measurements were conducted by Testmark Laboratories,

J.D. Ouellet et al. / Ecotoxicology and Environmental Safety 95 (2013) 104–112 107

following the analytical methodology of the American Public Health Association(APHA) and US Environmental Protection Agency (EPA).

During the exposure period, egg production, egg sizes, fathead minnow larvae,and C. dilutus emergence were monitored daily. Each day, breeding tiles werechecked for eggs, and eggs were scraped onto a petri dish and photographed. Teneggs per brood were selected and analyzed with Image Pro Plus 6.1 (MediaCybernetics Inc., Maryland, USA) for egg size determinations. Subsequently, theeggs were placed into egg cups, returned to the respective treatment, and aerated.Unfertilized eggs and dead eggs were counted and removed daily until hatchingwas complete. Once hatched, larvae were moved into petri dishes and photo-graphed with the microscope to assess larval deformities. At the end of theexposure period, fathead minnows were anesthetized with tricaine methanesulfo-nate (MS-222, Sigma-Aldrich, St Louis, MO, USA), assessed for secondary sexcharacteristics, total body weight, total length, and dissected to obtain livers, gills,gonads, and the carcass. On the final day of the exposure period, three 9-cm2 coresper stream were taken to determine densities of C. dilutus larvae. The number ofthird and fourth instar C. dilutus was counted from the three cores to determine anaverage number of C. dilutus per stream. One gram (wet weight) of C. dilutus werealso collected from three randomly selected streams per treatment and stored in acooler of dry ice along with the fish tissue samples, and shipped out to TestmarkLaboratories for metal analyses. Metal analysis in the fish tissues and C. dilutus wascarried out using ICP-MS following microwave digestion. In order to maintainquality control and quality assurance of metal analysis, Testmark Laboratories usedmethod blanks, positive controls, and blank spikes, and also analyzed a certifiedreference material [DOLT-3 dogfish (Squalus acanthias) liver, National ResearchCouncil of Canada]. Percentage recovery for the elements analyzed ranged from84.06 to 111.04 percent.

2.7. Metal speciation analysis

We used the geochemical speciation model, Visual MINTEQ, version 3.0 (KTH,Department of Land and Water Resources Engineering, Stockholm, Sweden), toestimate the free ion concentrations of various metals in the 45 percent PWEtreatments. The mean values presented in Tables 1 and 2 were used to run thespeciation modeling. The dissolved organic carbon content of each treatment wasassumed to be 60 percent humic acid and 40 percent fulvic acid (Kamunde andMacPhail, 2011).

2.8. Exposure analysis and statistics

Data were analyzed and graphed using IBM SPSS Statistics 20.0.0 (SPSS,Chicago, IL) and Sigmaplots Version 11.0 (San Jose, CA, USA). Water chemistry,and metal burdens in C. dilutus and fish tissues were analyzed using one-wayanalysis of variance (ANOVAs). Adult survival, condition factor (k), liver somaticindex [LSI (percent)], and gonadal somatic index [GSI (percent)], as well as meanegg sizes, mean total deformities (percent), mean fertilization success (percent),and mean densities of C. dilutus per replicate were also analyzed using one-wayANOVAs. The Shapiro–Wilk test was used to test parametric assumptions fornormality and Levine’s test was used to test for homogeneity over variance priorto the one-way ANOVA analysis. Data that failed these assumptions were eithertransformed [arcin(percent) or log10] or analyzed using the non-parametricKruskal–Wallis test. If significant differences were detected by ANOVAs, Tukey’spost-hoc test was used to determine the differences between the treatment and thereference(s) and/or among the treatments. Cumulative mean daily egg production(calculated as: # eggs/breeding pair each day, summed cumulatively for 21 days)and cumulative total spawning events were compared between each treatment byperforming multiple two-sample Kolmogorov–Smirnov tests.

Table 2Proportions ( percent) of free metal ions from the unmodified and modified(increased alkalinity and increased DOC) 45 percent process water effluent (PWE)treatments determined from Visual MINTEQ speciation modeling.

Parameter

Metal 45 %PWE

45 % PWE+inc.alk.

45 % PWE+inc.DOC

Barium Free metals (%) 66.17 65.90 66.64Cobalt Free metals (%) 56.14 52.08 56.60Copper Free metals (%) 2.34 0.13 0.72Lithium Free metals (%) 96.52 96.50 96.61Manganese Free metals (%) 59.58 53.12 60.18Nickel Free metals (%) 54.36 48.06 54.26Rubidium Free metals (%) 93.35 93.32 93.52Strontium Free metals (%) 57.08 56.93 57.64Thallium Free metals (%) 83.72 83.55 84.07Zinc Free metals (%) 49.93 39.99 50.18

3. Results

3.1. Water chemistry

3.1.1. RW+increased alkalinity and RW+increased DOC vs. RWAlkalinity, pH, conductivity, and sodium concentrations were

all significantly greater in the RW+increased alkalinity treatmentrelative to the other RW treatments, while DOC was significantlygreater (�60 percent increase) in the RW+increased DOC treat-ment relative to the other RW treatments (Tukey’s HSD posthoctest; po0.05) (Table 1). The measured DOC concentration wassimilar to nominal concentrations in RW+increased DOC. Therewere no significant differences in dissolved metal concentrationsamong any of the RW treatments (ANOVA; p40.05) (Table 1).

3.1.2. 45 percent PWE vs. RWThe pH was significantly lower in 45 percent PWE relative to

RW (Tukey’s HSD posthoc test; po0.05), while ammonia, chloride,conductivity, nitrate, sodium, sulfate, total hardness, calcium, andmagnesium measurements were significantly greater in the 45percent PWE treatment relative to the RW treatment (Tukey’s HSDposthoc test; po0.05) (Table 1). The 45 percent PWE exposurewater contained elevated concentrations of eleven metals (barium,boron, cobalt, copper, lithium, manganese, molybdenum, nickel,rubidium, selenium, and strontium) relative to the RW exposure(Tukey’s HSD posthoc test; po0.05 for each metal) (Table 1).

3.1.3. 45 percent PWE+increased alkalinity and 45 percentPWE+increased DOC Vs. 45 percent PWE

The alkalinity increased by more than four-fold in the 45percent PWE+increased alkalinity treatment relative to the other45 percent PWE treatments, and both pH, and sodium concentra-tions were significantly greater, as well (Tukey’s HSD posthoc test;po0.05). Similarly, DOC was significantly greater (�25 percentincrease) in the 45 percent PWE+increased DOC treatment relativeto the other 45 percent PWE treatments (Tukey’s HSD posthoctest; po0.05) (Table 1). Dissolved metal concentrations weremostly similar among 45 percent PWE treatments, except forconcentrations of manganese which were significantly lower inthe 45 percent PWE+increased alkalinity treatment relative to theother 45 percent PWE treatments (Tukey’s HSD posthoc test;po0.05), and concentrations of copper which were significantlylower in the 45 percent PWE+increased DOC treatment relative tothe other 45 percent PWE treatments (Tukey’s HSD posthoc test;p≤0.002) (Table 1).

3.1.4. Predicted proportions of free metal ions in 45 percentPWE+increased alkalinity and 45 percent PWE+increased DOC vs. 45percent PWE

In general, the proportion of free ions for all of the metalsexamined were 440 percent in each of the 45 percent PWEtreatments, except for copper (o3 percent) (Table 2). Increasedalkalinity in 45 percent PWE contributed to a predicted increase incarbonate-bound proportion of cobalt (�6.5 percent increase),manganese (�11 percent increase), nickel (�11 percent increase),and zinc (�10 percent increase) (data not shown), which alsocoincided with a decrease (7–20 percent) in predicted free ionconcentrations of the same metals (Table 2). Notably, free copperion concentration reduced to an almost negligible proportion (0.13percent) in the 45 percent PWE+increased alkalinity treatment(Table 2). In contrast, increased DOC level in the 45 percent PWEdid not affect the free ion concentration of any metals exceptcopper, which showed a �70 percent decrease relative to that inthe unmodified 45 percent PWE treatment (Table 2). Due to thehigh sulfate level in the 45 percent PWE, the predicted proportion

J.D. Ouellet et al. / Ecotoxicology and Environmental Safety 95 (2013) 104–112108

of sulfate-bound metals in each of the 45 percent PWE treatmentswas relatively high. The sulfate species of barium, cobalt, manga-nese, nickel, strontium, and zinc were 430 percent for each ofthese metals, while the proportion of sulfate-bound rubidium andthallium were �5 percent and �16 percent, respectively (data notshown).

3.2. Metal burdens of C. dilutus

There were no significant differences in the metal concentra-tions of C. dilutus larvae among RW treatments (Tukey’s HSDposthoc test; p40.05). Although eleven metals were found to beat elevated concentrations in the 45 percent PWE exposure waterrelative to the reference water, only six metals (cobalt, copper,nickel, rubidium, selenium and thallium) were present at signifi-cantly elevated concentrations in C. dilutus larvae in the 45 percent

Fig. 1. Mean concentrations of metals (7S.E.) that were elevated in C. dilutuslarvae exposed to 45 percent process water effluent (PWE). Letters indicatesignificant differences, where means that do not share letters are statisticallydifferent from one another for that metal (Tukey HSD posthoc test, po0.05).

Fig. 2. Mean concentrations of metals (7S.E.) that were elevated in various female fatwater effluent (PWE). Letters indicate significant differences, where means that do not(Tukey HSD posthoc test, po0.05). oMDL indicates less than minimum detection limi

PWE treatments relative to the RW treatments (Tukey’s HSDposthoc test; po0.05) (Fig. 1). There were, however, some differ-ences in the tissue metal concentrations of C. dilutus larvae among45 percent PWE treatments. Concentrations of cobalt and nickelin C. dilutus larvae were approximately two-fold greater in the45 percent PWE+increased alkalinity treatment relative to the45 percent PWE treatment (Tukey’s HSD posthoc test; p¼0.002and p¼0.012, respectively). Similarly, concentrations of copper inC. dilutus larvae were �50 percent greater in the 45 percentpercent PWE+increased DOC treatment relative to the unmodified45 percent PWE treatment (Tukey's HSD posthoc test; p¼0.036).

3.3. Tissue-specific metal burdens in fathead minnows

In general, there were no significant differences in the con-centration of any metals in any tissues examined among femalefathead minnows from the RW treatments (Tukey’s HSD posthoctest; p40.05). There were, however, significant differences intissue-specific metal concentrations in female fathead minnowsexposed to 45 percent PWE relative to RW treatments (Fig. 2). Thetissue concentrations of rubidium (liver, gonads, gills, and carcass),selenium (liver, gonads, and gills), nickel (gonads and gills),thallium (gonad and carcass), and copper (gills) were significantlyelevated in fish exposed to 45 percent PWE relative to RW (Tukey’sHSD posthoc test; po0.05). In addition, the tissue-specific metalconcentrations in fish were similar among all three 45 percentPWE treatments except for copper. The gill copper concentrationwas significantly greater in fish exposed to the 45 percent PWEwith increased DOC relative to that in the unmodified 45 percentPWE treatment (Tukey's HSD posthoc test; p¼0.033).

3.4. Biological endpoints and fecundity

The survival of adult fish was not impacted by any of thetreatments (ANOVA; p40.05) (data not shown). Similarly, therewere no statistically significant differences in GSI (percent), LSI

head minnow liver, gonads, carcass, and gill tissues exposed to 45 percent processshare letters are statistically different from one another for that metal and tissuets.

Fig. 3. Cumulative daily mean egg production (measured as eggs/pair) for breeding pairs of fathead minnows over the 21-day exposure period (A) and mean brood size (7S.E.) of fathead minnow breeding pairs (B) in unmodified and modified (increased alkalinity and increased DOC) reference water (RW) and 45 percent process water effluent(PWE) treatments. (A) Treatment groups that do not share letters are statistically different from one each other (two-sample Kolmogorov–Smirnov test; po0.05). (B) Therewere no statistically significant differences among the treatments (Kruskal–Wallis; p¼0.102).

Fig. 4. Mean densities of C. dilutus larvae (7S.E.) sampled from unmodified andmodified (increased alkalinity and increased DOC) reference water (RW) and 45percent process water effluent (PWE) treatments on day 21 of the exposure period.Means that do not share letters are statistically different from one another (TukeyHSD posthoc test, po0.05).

J.D. Ouellet et al. / Ecotoxicology and Environmental Safety 95 (2013) 104–112 109

(percent), or condition factor for either males or females amongthe six different treatments (ANOVA; p40.05) (data not shown).There were also no statistically significant differences observed inhatch rate, larval deformities, or egg sizes among any of the sixdifferent treatments (ANOVA; p40.05 for each) (data not shown).

Fathead minnow egg production, measured as cumulativemean daily egg production, was variable among the treatmentsover the 21-day exposure period (Fig. 3a). Fish in the RW+increased alkalinity and RW+increased DOC treatments pro-duced significantly more eggs (+496.0 eggs/pair and +588.6eggs/pair, respectively) relative to fish in the RW treatment(two-sample Kolmogorov–Smirnov test; p¼0.002 and p¼0.001,respectively). In contrast, fish in the 45 percent PWE treatmentproduced significantly fewer eggs than fish in the RW treatment(−301.8 eggs/pair) (two-sample Kolmogorov–Smirnov test;p¼0.002). Interestingly, similar to our observation in the RWtreatments, fish exposed to 45 percent PWE either with elevatedalkalinity or elevated DOC level produced significantly more eggsthan fish in the 45 percent PWE treatment (+569.8 eggs/pair and+261.5 eggs/pair, respectively) (two-sample Kolmogorov–Smirnovtest; p≤0.001 for both). Consistent to this observation, we alsorecorded a trend towards greater brood sizes in fish in themodified treatments relative to the unmodified treatments irre-spective of 45 percent PWE exposure, although this difference wasnot statistically significant (Kruskal–Wallis; p¼0.102) (Fig. 3b).

3.5. Densities of C. dilutus

Densities of C. dilutus measured on day 21 of the exposure periodare presented in Fig. 4. The highest density of C. dilutus larvae(1.35 larvae/cm2) was observed in the RW treatment, followedclosely by the RW+increased DOC treatment (1.11 larvae/cm2). TheRW and RW+increased DOC treatments were the only two treat-ments that contained approximately the desired densities for satia-tion (1.48 larvae/cm2, equivalent to 1 g/pair/day), and the densitieswere not significantly different between these two treatments(Tukey’s HSD posthoc test; p¼0.823). The RW+increased alkalinitytreatment contained significantly fewer C. dilutus larvae than theunmodified RW treatment (−0.60 larvae/cm2) (Tukey’s HSD posthoctest; p¼0.044). The unmodified 45 percent PWE treatment con-tained significantly fewer larvae than the unmodified RW treatment(−0.71 larvae/cm2) (Tukey’s HSD posthoc test; p¼0.011). Larvaldensities in the 45 percent PWE+increased alkalinity and 45 percentPWE+increased DOC treatments were not significantly different

from the unmodified 45 percent PWE treatment (Tukey's HSDposthoc test; p¼0.993).

4. Discussion

4.1. Reproductive impacts

The findings of our study suggest that elevated alkalinity(achieved by adding NaHCO3) or DOC (achieved by adding commer-cial humic acid) can improve fathead minnow fecundity duringchronic exposure to 45 percent PWE. Fish that were exposed tounmodified 45 percent PWE produced significantly fewer eggs overthe 21-day exposure period relative to fish in the unmodified RWtreatment—a typical response recorded in several previous studieswith 45 percent PWE (Ouellet et al., 2013a, 2013b; Rickwood et al.,2006; Rozon-Ramilo et al., 2011a, 2011b). Fish fecundity improvedmarkedly once 45 percent PWE treatments were modified toincrease the alkalinity or DOC level in the exposure water. However,fish fecundity in both of the modified 45 percent PWE treatmentswas significantly less than the corresponding modified RW treat-ments (i.e., 45 percent PWE+increased alkalinity vs. RW+increasedalkalinity, and 45 percent PWE+increased DOC vs. RW+increasedDOC). This might be due to the additional energetic cost of metal

J.D. Ouellet et al. / Ecotoxicology and Environmental Safety 95 (2013) 104–112110

handling and metabolism in 45 percent PWE treatments fish, sincetissue-specific metal burden was significantly greater in these fishrelative to RW treatments fish, irrespective of water chemistrymodifications (discussed below).

It appears that sodium bicarbonate or humic acid additionimproves egg production in fathead minnows in general, asopposed to only ameliorating the effects of 45 percent PWE. Inagreement with our observation, Mager et al. (2010) recordedimproved egg production in fathead minnows when bicarbonate(pH 8.3) was added in the exposure water. Interestingly however,they reported that this effect was present only when bicarbonatewas added into the control water but not in the treatment withmetal (lead). Mager et al. (2010) also observed moderate improve-ments to fish fecundity when humic acid was added (4 mg/L) tothe exposure water, with or without lead. In addition, greater eggproduction in fathead minnows has been reported at pH 7.5 com-pared to lower pH values (e.g., small decrease in eggs at pH 6.6 anddramatic decrease below pH 5.9) in the exposure water (Mount,1973). Although pH likely had a minor influence on fish eggproduction in our study, since the decrease in pH in the 45 percentPWE treatment relative to the RW treatment was relativelymodest (∼0.5 unit). Furthermore, pH in the 45 percent PWEtreatment (∼7.2) was within the acceptable range for the fishreproductive bioassay employed in this study (Ankley et al., 2001).Overall, these observations suggest that water chemistry itself maybe an important factor in influencing reproductive capacity in fish.

At present, it is difficult to explain the exact reasons forimprovements in fish fecundity due to changes in water chemistry.Mager et al. (2010) suggested that humic acid, because of its abilityto bind biological surfaces, might contribute to better egg attach-ment on the breeding tiles. Sodium bicarbonate could perhapsimprove the osmotic balance between the treatment waters andeggs, resulting in improved egg hardness and attachment to thebreeding tiles, as well. Improvements in egg attachment to thebreeding tiles would likely be indicated by greater mean broodsizes from fish in water chemistry modified treatments, which wasobserved in the present study. It is also possible that greatersodium concentration in elevated alkalinity treatments might haveameliorated the energetic cost of osmoregulation, allowing fish toallocate more energy towards reproduction, which eventually ledto improved fecundity. In the present study, the addition ofsodium bicarbonate in alkalinity treatments increased the sodiumconcentrations from approximately 10 to 32 mg/L in the RW andfrom approximately 65 to 113 mg/L in the 45 percent PWE. It hasbeen recently reported that egg production in fathead minnowsgenerally improves with increasing sodium level in the water, witha pronounced effect at ≥25 mg/L (Squires, 2011, Squires et al.,2013), which is comparable to the sodium level in our RW+increased alkalinity treatment. However, the 45 percent PWEtreatments in the present study already contained high sodiumlevels in the water (460 mg/L), which suggests that sodium maynot be the only factor that influenced egg production in these fish.

In contrast, the addition of humic acid (as a sodium salt) in theRW and 45 percent PWE treatments resulted in a small andinsignificant increase in dissolved sodium levels compared to thatin the respective unmodified treatments, nevertheless resulted inincreased egg production in fish, as well. This might have occurreddue to the direct physiological effects of humic acid since it hasbeen found to stimulate branchial sodium uptake and reduceparacellular permeability of the gill in fish (Matsuo et al., 2004;Wood et al., 2003). Both of these effects are expected to alleviatethe osmoregulatory stress that freshwater fish experience due tothe continuous loss of essential ions from the body through simplediffusion, and thus may facilitate egg production by fish. Inaddition, these effects may also ameliorate the disruption ofessential ion homeostasis in fish induced by exposure to metals

(e.g., copper) (Wood et al., 2011). It is therefore reasonable tosuggest that humic acid influenced egg production in fatheadminnows exposed to RWor 45 percent PWE possibly because of itsbeneficial physiological effects, rather than by reducing metalbioavailability through increased complexation of dissolved metals(discussed below). To the best of our knowledge, this is one of thefew studies to report that humic acid exposure can augment eggproduction in fish, and further investigations are required tounderstand the mechanistic underpinnings of this effect. It is alsoimportant to note here that sulfate levels were greater than 20-fold higher in the 45 percent PWE treatments relative to the RWtreatments in the present study. Although the egg production infathead minnows have been found to decrease with increasingdissolved sulfate level (Squires et al., 2013), it is unlikely to be thecase in the present study since we have previously recorded nodecrease in egg production when fish were exposed to singlemetals (copper, nickel or selenium) along with almost identicallevels of dissolved sulfate, as recorded in our 45 percent PWEtreatments (Ouellet et al., 2013a).

4.2. Metal bioaccumulation

In addition to the amelioration of reproductive impairment,one of the other hypotheses of the present study was that elevatedalkalinity or DOC in the 45 percent PWE treatment would reducewaterborne metal bioavailability and tissue-specific metal accu-mulation in fathead minnows due to increased complexation offree metal ions. Chronic exposure to 45 percent PWE resulted inincreased accumulation of multiple metals in different tissues offathead minnows, however contrary to our hypothesis we did notobserve any protective effect of increased alkalinity or DOC on themetal(s) accumulation in any fish tissues. One of the possiblereasons for the lack of protective effect might be that the increasein alkalinity or DOC produced a modest decrease in the free ionconcentrations of most metals present in the 45 percent PWEexposure. For example, the four-fold increase in alkalinity in the 45percent PWE treatment only produced 7–20 percent decrease inthe free ion concentration of the major metals with the exceptionof copper (490 percent decrease). Similarly, the addition of humicacid (4 mg/L) into the 45 percent PWE exposure did not affect thefree ion activity of most metals except cobalt and copper (7 and 70percent decrease, respectively). This likely occurred because oftwo factors: (i) the actual increase of DOC in the exposure wasmarginal (only by 25 percent), and (ii) the complexation effect ofDOC on metal speciation is metal-specific and depends on therelative affinity of binding of free metal ion(s) to DOC, and copperbinds more strongly to DOC compared to any other metals presentin 45 percent PWE [see Niyogi and Wood (2004) for review].Perhaps, an unusually high sulfate level in the 45 percent PWE alsoinfluenced the relative complexation of different metals by bicar-bonate or humic acid. Moreover, it is important to note that thegeochemical speciation models like MINTEQ, which was used inthe present study for the estimations of free metal ion activities,are based on short-term metal–ligand binding under equilibriumcondition, and whether such equilibrium assumptions are validunder prolonged exposure remains an open question.

Although the protective effects of DOC and/or high pH/alkali-nity on metal accumulation and toxicity to fish during acuteexposures to metal(s) or metal-mixtures are well-documented(Burnison et al., 2006; Klinck et al., 2005; Komjarova and Blust,2009; Niyogi et al., 2008; Pyle et al., 2002; Richards et al., 2001),recent evidence indicate that these water chemistry parametersoften provide much less or no protection during chronic metalexposures. Mager et al. (2010) observed no significant effect ofincreased alkalinity (added as 500 mm NaHCO3) on lead accumula-tion in fathead minnows during chronic waterborne exposure to

J.D. Ouellet et al. / Ecotoxicology and Environmental Safety 95 (2013) 104–112 111

lead, despite a �30 percent decrease in free lead ion concentra-tion in the exposure water. Interestingly though, they reportedamelioration of lead accumulation and toxicity in fish when DOClevel was elevated in the exposure (added as 4 mg/L humic acid),which also resulted in a �30 percent decrease in free lead ionlevel. On the other hand, Brauner and Wood (2002a) reported thatincreased DOC (added as 12 mg/L humic acid) decreased mortalitybut did not ameliorate the adverse physiological response inrainbow trout chronically exposed to waterborne silver. Currently,very little is known about the effects of alkalinity or DOC on thebioavailability and toxicity of metals to fish during chronicexposures to metal mixtures. Recently, Kamunde and MacPhail(2011) examined the effect of increased DOC (added as 5 mg/Lhumic acid) during chronic exposure to a waterborne mixture ofcopper, cadmium and zinc to rainbow trout. They reported that theprotective effect of DOC against metal(s) accumulation was vari-able (absent or partial/complete) depending on tissue and expo-sure duration. More importantly, they also concluded that themodeled free metal ion activities in the exposure could onlyexplain the cadmium, but not copper and zinc, accumulation infish. Overall, our findings on the effects of increased alkalinity andDOC on metal(s) accumulation in fish during chronic exposure tocomplex metal-mixture are more or less consistent with previousobservations, and provide further evidence that the free ionactivity alone is not a reliable predictor of chronic metal accumu-lation or toxicity in fish.

4.3. Other biological endpoints in fish

Apart from cumulative egg production and tissue-specificmetal accumulation, none of the other biological endpointsexamined in the present study were influenced by 45 percentPWE, as well as increased alkalinity or DOC. Fathead minnowsexposed to 45 percent PWE typically do not show changes tobiological endpoints such as GSI, LSI, or condition factor (Ouelletet al., 2013a; Rickwood et al., 2006; Rozon-Ramilo et al., 2011a).The effects on hatch rate, larval deformities, and egg size havebeen variable in fish chronically exposed to 45 percent PWE.Rickwood et al. (2006) observed a decrease in hatching successand an increase in larval deformities when fathead minnows wereexposed to 45 percent PWE simultaneously via water and diet,however no effects when fathead minnows were exposed viawater only. In subsequent studies, Rozon-Ramilo et al. (2011a) andOuellet et al. (2013a, 2013b), however, did not observe anysignificant differences in hatching success or larval deformitiesduring chronic multi-trophic exposure to 45 percent PWE. Inaddition, a decrease in egg size was previously recorded in fishduring chronic exposure to 45 percent PWE, although the effectwas not always consistent across similar experiments (Ouelletet al., 2013a). It is possible that egg sizes might have been affectedby 45 percent PWE, instead of decreases in egg production, insome experiments. Consequently, fathead minnows produced asimilar number of eggs but in smaller sizes, instead of producingfewer eggs. Regardless, the egg size was not affected by any of thetreatments carried out in the present study.

4.4. The role of diet (C. dilutus)

The presence of live C. dilutus larvae as a food source in multi-trophic MME exposures have been shown to be an importantfactor that contributes to metal mixture accumulation and toxicresponses in fish (Rickwood et al., 2006; Rozon-Ramilo et al.,2011b). The present study indicates that the metal bioavailabilityvia diet during chronic 45 percent PWE exposure was not alteredby increased alkalinity or DOC, since the C. dilutus metal burdenwas generally similar across all three 45 percent PWE treatments.

However, we recorded differences in densities of C. dilutus larvaeamong treatments, which were measured at the end of theexposure period. A decrease in densities of C. dilutus larvae dueto 45 percent PWE exposure have also been reported in previousstudies (Hruska and Dubé, 2004; Rozon-Ramilo et al., 2011a), andit has been implicated for reduced egg production in fatheadminnows (Ouellet et al., 2013b). In the present study, egg produc-tion in fish might also have been affected by the decrease in foodavailability in the unmodified 45 percent PWE treatment. How-ever, a difference in food availability alone does not explain thedecrease in fish fecundity induced by 45 percent PWE, since fish inboth of the modified 45 percent PWE treatments produced asmany eggs as the fish in unmodified RW treatment, despiteencountering a similar decrease in C. dilutus density. This furtheremphasizes the point that sodium bicarbonate and humic acidprobably facilitated fish egg production by directly influencing thephysiology of fish.

5. Conclusions

Overall, the findings of the present study corroborate thatenvironmentally relevant chronic exposure to MME can lead tosignificant tissue-specific accumulation of metals and reduce fishfecundity. Modifications of the water chemistry through theaddition of sodium bicarbonate (increased alkalinity) or humicacid (increased DOC) restored or improved the reproductive out-put, but did not alter the tissue-specific metal accumulationprofile, in fish exposed to MME. Interestingly, the addition ofsodium bicarbonate and humic acid also resulted in significantlygreater fecundity in fish exposed to reference water. Fish in theMME exposures, irrespective of water chemistry modifications,encountered reduced food availability (C. dilutus density). Collec-tively, these results indicate that reproductive impairment in fishinduced by chronic MME exposure was not solely mediated by theaccumulation of metals and indirect effects such as reduced foodavailability probably contributed to the effect. However, it appearsthat sodium bicarbonate or humic acid addition enabled fish tocompensate adverse reproductive consequences of MME exposure,possibly through their direct beneficial physiological effects.Further studies are required to examine the potential usefulnessof water chemistry modifications in reducing the toxicity of MMEto biota in the receiving environment.

Acknowledgments

We would like to thank Allison Merla and Christine Brereton atVale-Canada for support on this project, as well as Sara Pryce, LisaRozon-Ramilo, Allison Squires, and Melissa Driessnack for theirassistance in the laboratory. Funding was provided by Vale-Canada, Natural Sciences and Engineering Research Council ofCanada, and the University of Saskatchewan.

References

Ankley, G.T., Jensen, K.M., Kahl, M.D., Korte, J.J., Makynen, E.A., 2001. Descriptionand evaluation of a short‐term reproduction test with the fathead minnow(Pimephales promelas). Environ. Toxicol. Chem. 20, 1276–1290.

Bennett, P.M., Janz, D.M., 2007. Bioenergetics and growth of young-of the-yearnorthern pike (Esox lucius) and burbot (Lota lota) exposed to metal miningeffluent. Ecotoxicol. Environ. Saf. 68, 1–12.

Borgmann, U., Norwood, W., Dixon, D., 2008. Modelling bioaccumulation andtoxicity of metal mixtures. Hum. Ecol. Risk Assess. 14, 266–289.

Brauner, C.J., Wood, C.M., 2002a. Effect of long-term silver exposure on survival andionoregulatory development in rainbow trout (Oncorhynchus mykiss) embryosand larvae, in the presence and absence of added dissolved organic matter.Comp. Biochem. Physiol. C. Toxicol. Pharmacol, 133, 161–173.

Brauner, C.J., Wood, C.M., 2002b. Ionoregulatory development and the effect ofchronic silver exposure on growth, survival, and sublethal indicators of toxicity

J.D. Ouellet et al. / Ecotoxicology and Environmental Safety 95 (2013) 104–112112

in early life stages of rainbow trout (Oncorhynchus mykiss). J. Comp. Physiol. B172, 153–162.

Burnison, B.K., Meinelt, T., Playle, R., Pietrock, M., Wienke, A., Steinberg, C.E.W.,2006. Cadmium accumulation in zebrafish (Danio rerio) eggs is modulated bydissolved organic matter (DOM). Aquat. Toxicol. 79, 185–191.

De Schamphelaere, K.A.C., Janssen, C.R., 2004a. Development and field validation ofa biotic ligand model predicting chronic copper toxicity to Daphnia magna.Environ. Toxicol. Chem. 23, 1365–1375.

De Schamphelaere, K.A.C., Janssen, C.R., 2004b. Effects of dissolved organic carbonconcentration and source, pH, and water hardness on chronic toxicity of copperto Daphnia magna. Environ. Toxicol. Chem. 23, 1115–1122.

Di Toro, D.M., Allen, H.E., Bergman, H.L., Meyer, J.S., Paquin, P.R., Santore, R.C., 2001.Biotic ligand model of the acute toxicity of metals. 1. Technical basis. Environ.Toxicol. Chem. 20, 2383–2396.

Driessnack, M.K., Dubé, M.G., Rozon-Ramilo, L.D., Jones, P.D., Wiramanaden, C.I.E.,Pickering, I.J., 2011. The use of field-based mesocosm systems to assess theeffects of uranium milling effluent on fathead minnow (Pimephales promelas)reproduction. Ecotoxicology 20, 1209–1224.

Dubé, M.G., MacLatchy, D.L., Hruska, K.A., Glozier, N.E., 2006. Assessing theresponses of creek chub (Semotilus atromaculatus) and pearl dace (Semotilusmargarita) to metal mine effluents using in situ artificial streams in Sudbury,Ontario, Canada. Environ. Toxicol. Chem. 25, 18–28.

Dubé, M.G., MacLatchy, D.L., Kieffer, J.D., Glozier, N.E., Culp, J.M., Cash, K.J., 2005.Effects of metal mining effluent on Atlantic salmon (Salmo salar) and slimysculpin (Cottus cognatus): using artificial streams to assess existing effects andpredict future consequences. Sci. Total Environ. 343, 135–154.

Franssen, C.M., 2009. The effects of heavy metal mine drainage on population sizestructure, reproduction, and condition of western mosquitofish, Gambusiaaffinis. Arch. Environ. Contam. Toxicol. 57, 145–156.

Gerhardt, A., 1998. Whole effluent toxicity testing with Oncorhynchus mykiss(Walbaum 1792): survival and behavioral responses to a dilution series of amining effluent in South Africa. Arch. Environ. Contam. Toxicol. 35, 309–316.

Hruska, K.A., Dubé, M.G., 2005. Comparison of a partial life‐cycle bioassay inartificial streams to a standard beaker bioassay to assess effects of metal mineeffluent on Chironomus tentans. Environ. Toxicol. Chem. 24, 2325–2335.

Hruska, K.A., Dubé, M.G., 2004. Using artificial streams to assess the effects ofmetal‐mining effluent on the life cycle of the freshwater midge (Chironomustentans) in situ. Environ. Toxicol. Chem. 23, 2709–2718.

Jezierska, B., Ługowska, K., Witeska, M., 2009. The effects of heavy metals onembryonic development of fish (a review). Fish Physiol. Biochem. 35, 625–640.

Kamunde, C., MacPhail, R., 2011. Effect of humic acid during concurrent chronicwaterborne exposure of rainbow trout (Oncorhynchus mykiss) to copper,cadmium and zinc. Ecotoxicol. Environ. Saf. 74, 259–269.

Klinck, J., Dunbar, M., Brown, S., Nichols, J., Winter, A., Hughes, C., Playle, R.C., 2005.Influence of water chemistry and natural organic matter on active and passiveuptake of inorganic mercury by gills of rainbow trout (Oncorhynchus mykiss).Aquat. Toxicol. 72, 161–175.

Komjarova, I., Blust, R., 2009. Effects of Na, Ca, and pH on the simultaneous uptakeof Cd, Cu, Ni, Pb, and Zn in the zebrafish Danio rerio: a stable isotopeexperiment. Environ. Sci. Technol. 43, 7958–7963.

Mager, E.M., Brix, K.V., Grosell, M., 2010. Influence of bicarbonate and humic acid oneffects of chronic waterborne lead exposure to the fathead minnow (Pimephalespromelas). Aquat. Toxicol. 96, 135–144.

Matsuo, A.Y., Playle, R.C., Val, A.L., Wood, C.M., 2004. Physiological action ofdissolved organic matter in rainbow trout in the presence and absence ofcopper: sodium uptake kinetics and unidirectional flux rates in hard andsoftwater. Aquat. Toxicol. 70, 63–81.

McPherson, T.D., Mirza, R.S., Pyle, G.G., 2004. Responses of wild fishes to alarmchemicals in pristine and metal-contaminated lakes. Can. J. Zool. 82, 694–700.

Mount, D.I., 1973. Chronic effect of low pH on fathead minnow survival, growth andreproduction. Water Res. 7, 987–993.

Niyogi, S., Kent, R., Wood, C.M., 2008. Effects of water chemistry variables on gillbinding and acute toxicity of cadmium in rainbow trout (Oncorhynchus mykiss):a biotic ligand model (BLM) approach. Comp. Biochem. Physiol. C. Toxicol.Pharmacol. 148, 305–314.

Niyogi, S., Wood, C.M., 2004. Biotic ligand model, a flexible tool for developingsite-specific water quality guidelines for metals. Environ. Sci. Technol. 38,6177–6192.

Ouellet, J.D., Dubé, M.G., Niyogi, S., 2013a. A single metal, metal mixture, andwhole-effluent approach to investigate causes of metal mine effluent effects onfathead minnows (Pimephales promelas). Water Air Soil Pollut. 224, 1462.

Ouellet, J.D., Dubé, M.G., Niyogi, S., 2013b. The influence of food quantity on metalbioaccumulation and reproduction in fathead minnows (Pimephales promelas)during chronic exposures to a metal mine effluent. Ecotoxicol. Environ. Saf. 91,188–197.

Pagenkopf, G.K., Russo, R.C., Thurston, R.V., 1974. Effect of complexation on toxicityof copper to fishes. J. Fish. Res. Board Can. 31, 462–465.

Paquin, P.R., Santore, R.C., Wu, K.B., Kavvadas, C.D., Di Toro, D.M., 2000. The bioticligand model: a model of the acute toxicity of metals to aquatic life. Environ.Sci. Policy 3, 175–182.

Playle, R.C., Dixon, D.G., Burnison, K., 1993. Copper and cadmium binding to fishgills: modification by dissolved organic carbon and synthetic ligands. Can. J.Fish. Aquat. Sci. 50, 2667–2677.

Pyle, G.G., Rajotte, J.W., Couture, P., 2005. Effects of industrial metals on wild fishpopulations along a metal contamination gradient. Ecotoxicol. Environ. Saf. 61,287–312.

Pyle, G.G., Swanson, S.M., Lehmkuhl, D.M., 2002. The influence of water hardness,pH, and suspended solids on nickel toxicity to larval fathead minnows(Pimephales promelas). Water Air Soil Pollut. 133, 215–226.

Richards, J.G., Curtis, P.J., Burnison, B.K., Playle, R.C., 2001. Effects of natural organicmatter source on reducing metal toxicity to rainbow trout (Oncorhynchusmykiss) and on metal binding to their gills. Environ. Toxicol. Chem. 20,1159–1166.

Rickwood, C.J., Dubé, M.G., Weber, L.P., Driedger, K.L., Janz, D.M., 2006. Assessingeffects of metal mining effluent on fathead minnow (Pimephales promelas)reproduction in a trophic-transfer exposure system. Environ. Sci. Technol. 40,6489–6497.

Rozon-Ramilo, L.D., Dubé, M.G., Rickwood, C.J., Niyogi, S., 2011a. Examining theeffects of metal mining mixtures on fathead minnow (Pimephales promelas)using field-based multi-trophic artificial streams. Ecotoxicol. Environ. Saf. 74,1536–1547.

Rozon-Ramilo, L.D., Dubé, M.G., Squires, A.J., Niyogi, S., 2011b. Examining water-borne and dietborne routes of exposure and their contribution to biologicalresponse patterns in fathead minnow (Pimephales promelas). Aquat. Toxicol.105, 466–481.

Santore, R.C., Di Toro, D.M., Paquin, P.R., Allen, H.E., Meyer, J.S., 2001. Biotic ligandmodel of the acute toxicity of metals. 2. Application to acute copper toxicity infreshwater fish and Daphnia. Environ. Toxicol. Chem. 20, 2397–2402.

Schwartz, M.L., Curtis, P.J., Playle, R.C., 2004. Influence of natural organic mattersource on acute copper, lead, and cadmium toxicity to rainbow trout (Oncor-hynchus mykiss). Environ. Toxicol. Chem. 23, 2889–2899.

Spencer, P., Bowman, M.F., Dubé, M.G., 2008. A multitrophic approach to monitor-ing the effects of metal mining in otherwise pristine and ecologically sensitiverivers in Northern Canada. Integr. Environ. Assess Manage. 4, 327–343.

Squires, A.J., 2011. Development and implementation of a watershed-scale cumu-lative effects framework for the Athabasca River Basin, Graduate Studies andResearch. University of Saskatchewan, Saskatoon, SK, Canada p. 175.

Squires, A.J., Dubé, M.G., Rozon‐Ramilo, L.D., 2013. Assessing the sublethal effects ofin‐river concentrations of parameters contributing to cumulative effects in theathabasca river basin using a fathead minnow bioassay. Environ. Toxicol. Chem.32, 662–672.

Wang, W.-X., 2011. Incorporating exposure into aquatic toxicological studies: animperative. Aquat. Toxicol. 105, 9–15.

Wood, C.M., Al-Reasi, H., Smith, D.S., 2011. The two faces of DOC. Aquat. Toxicol.105S, 3–8.

Wood, C.M., Matsuo, A.Y., Wilson, R.W., Gonzalez, R., Patrick, M.L., Playle, R.C., Val,A.L., 2003. Protection by natural blackwater against disturbances in ion fluxescaused by low pH exposure in freshwater stingrays endemic to the Rio Negro.Physiol. Biochem. Zool. 76, 12–27.