Freshwater Bivalve Ecotoxoicology - Chapter 10 potx

Although there are about 1,000 species of unionids worldwide, and about 300 species in North America, only about 11 species have been reported in the peer-reviewed literature to assessth

10 Biomarker Responses of Unionid

Mussels to Environmental Contaminants

Teresa J Newton and W Gregory Cope

INTRODUCTION

Unionid mussels are ecologically and economically important in aquatic ecosystems The biomass ofunionids can exceed the biomass of all other benthic organisms by an order of magnitude (Negus 1966;Layzer, Gordon, and Anderson 1993), and production (range, 1–20 g dry mass/m2/yr) can equal that byall other macrobenthos in many streams (Strayer et al 1994) Thus, unionids may play important roles

in particle processing, nutrient release, and sediment mixing (Vaughn and Hakenkamp 2001; Strayer

et al 2004) Mussels also serve as food for aquatic mammals, including raccoons, muskrats, and otters(Van der Schalie and Van der Schalie 1950) Historically, unionids provided a supplemental foodsource to indigenous peoples prior to European settlement, but large-scale commercial interest inunionids did not develop until the mid-1800s (Claassen 1994) Soon thereafter, unionids wereextensively harvested for the production of pearl buttons, and presently, unionid shells are used inthe multimillion dollar Asian cultured-pearl industry (Anthony and Downing 2001)

Overharvesting, widespread habitat destruction, pollution, land-use change, and invasivespecies introductions have caused many unionid populations to decline or disappear In NorthAmerica, most species are now extinct or imperiled, and unionids are widely recognized as one

of the most imperiled plants or animals on the continent (Master et al 2000) Although notsufficiently documented, exposure to toxic contaminants may also be contributing to these declines.There are few instances where chemical spills and other point sources of contaminants have causedlocalized mortality (Sheehan, Neves, and Kitchel 1989; Fleming, Augspurger, and Alderman1995); however, widespread decreases in density and diversity are more likely to result from thesubtle, pervasive effects of chronic, low-level contamination (Naimo 1995)

There is convincing evidence that unionids, and glochidial and juvenile life stages in particular,are sensitive to many contaminants relative to other aquatic species (e.g., Newton et al 2003,

Chapter 5 and Chapter 7) However, for a given chemical, toxicity can vary by an order ofmagnitude among life stage and species (Cherry et al 2002; Augspurger et al 2003) This is notsurprising given the diversity of life history adaptations present in this faunal group For example,differences in longevity (30–130 years), habitat requirements (silt to gravel), feeding strategies(filter-, deposit-, and pedal-feeding), and reproduction (hermaphrodites, dioecious) all contribute tothis diversity Although there are about 1,000 species of unionids worldwide, and about 300 species

in North America, only about 11 species have been reported in the peer-reviewed literature to assessthe effects of contaminants on biomarker responses in this imperiled faunal group Clearly,

257

ecotoxicological research on unionids needs to expand to encompass the breadth of life historiesfound in this group.

The development of physiological and biochemical tests or “biomarkers” of sublethal exposureare critical in assessing the condition of unionids Several textbooks and review articles have beenwritten on the use of biomarkers in a wide range of aquatic organisms (e.g., McCarthy and Shugart1990; Huggett et al 1992; Van der Oost, Beyer, and Vermeulen 2003)—many of these haveapplication in unionids

Freshwater bivalve ecotoxicology gained momentum in the early 1970s with potassium andcopper bioassays with unionids (Imlay 1971) Once initial studies revealed that unionids wereindeed sensitive to a variety of contaminants relative to other invertebrates and fishes, there hasbeen growing interest in using unionids to evaluate the toxicity of chemicals Most studies infreshwater bivalve ecotoxicology have either been short-term laboratory tests with single chemicals

(Chapter 5 and Chapter 7) or monitoring of biomarker responses in the field (Chapter 6 and

Chapter 9) Both of these approaches are necessary to understand basic biological responses.However, methods had to be, and are continuing to be, developed to culture and maintain adultsand juveniles in suitable physiological condition before and during testing and to evaluate suitablesublethal response endpoints (ASTM 2006) The field has made substantial progress in many ofthese areas in the past two decades, but it could further benefit from lessons learned in otherecotoxicological areas of study, such as with marine bivalves (ASTM 2002)

Although there are substantial differences in reproductive strategies between unionids andmarine bivalves that may confound certain direct comparisons, the field of unionid ecotoxicologycould benefit from some of the approaches commonly used in marine studies For example, aframework used in marine bivalve ecotoxicology recognized that this field needed to movebeyond simple acute tests to multidisciplinary studies that gather knowledge at several levels ofbiological organization, conduct persistent and systematic field studies, and provide iterationbetween experiment and field observations (Luoma 1996) Further, this framework stressed thatstandardized “simple” ecotoxicologic approaches lack power in explaining implications ofcontaminants in complicated circumstances Additional lessons from marine bivalves that should

be incorporated into a future framework for unionid ecotoxicology include the following (fromLuoma 1996):

1 Viewing contaminants as just one of the several influential physical, chemical, or logical variables in many aquatic systems

bio-2 Recognizing that contaminants are distributed among solution, suspended particles,sediments, pore waters, and food resources and that each species or life-history stagemay “sample” differently from this complex matrix

3 Toxicity databases derived from water-only exposures were developed to supportregulatory criteria, but these probably underestimate the exposures of bivalves inmany circumstances

4 Biological responses to contaminants in nature can be much more complex than theresponses observed in the laboratory

5 When conducted alone, simplistic toxicity tests, a single biomarker, whole organismanalysis, or studies that exclude variables other than contaminants will probably beinsensitive to all but the most extreme influences of contaminants

Contaminants can influence unionids at many different levels of biological organization

(Figure 10.1).Usually, the most robust approach is the use of a number of different indices thatspan the levels of biological complexity, such as biochemical, cytological, physiological, andautecological In theory, a contaminant first exerts its effects at the molecular level, and thechanges at this level lead to changes in organelles and cellular structures, and so forth However,

in nature, things are rarely this simple because at each level of biological organization, thereprobably exists homeostatic mechanisms to counteract these disruptive effects Also, becauseorganisms are rarely exposed to a single contaminant at a time, certain combinations of contami-nants may act in an additive or synergistic manner to amplify the effects of one another Lastly, thisapproach implies a linear response to contaminants that may not include a detectable thresholdeffect Certain contaminants, like essential metals, may be required for basic biological functions

at low concentrations, whereas above a threshold level, they may become toxic This conceptshould not be overlooked in studies of the effects of contaminants on unionids

This review of biomarker responses in unionids exposed to environmental contaminantsfocuses on studies that (1) reported measured contaminant concentrations; (2) had robust experi-mental designs, including the replication of control and contaminant treatments; and (3) werepublished in the peer-reviewed literature These criteria effectively removed about 50% of thepapers published in this area, primarily because many authors reported only nominal concen-trations However, we believe these criteria are critical for the objective evaluation of the effects

of contaminants on unionids For example, if only nominal concentrations are reported, the actualconcentration of a given contaminant that is available for uptake is basically unknown In somestudies that only report nominal concentrations, the exposure concentrations are so high that theyexceed the known solubility of the contaminant In these instances, the actual amount of thecontaminant that is available for uptake may be a small fraction of the nominal concentration,which could seriously underestimate toxicity Also, many contaminants may adhere to the walls ofexposure chambers or can be lost to volatilization—both of which may effectively reduce the actualexposure concentration Similarly, treatments need to be replicated to get an estimate of thevariation associated with the exposure Future studies should strive to report measured contaminantconcentrations (whenever possible) and replicate contaminant and control treatments to ensure

a robust design and analysis

Population

Organelle

Cellular

Tissue Organism

FIGURE 10.1 A generalized approach of biological organization that illustrates how a given contaminant mayexert its influence on unionids (Modified from Stebbing A R D., The Effects of Stress and Pollution on MarineAnimals, Praeger Scientific, New York, 1985 With permission.)

chemicals (Peakall 1994) As defined, biomarkers can span several levels of biological organization

(Figure 10.1),but biomarkers that have been investigated most extensively have been enzymesinvolved in the detoxification of xenobiotics and their metabolites (biotransformation and antiox-idant enzymes) Biomarkers are generally classified into those that indicate exposure, effects, orsusceptibility Although connections must be established between exposure to contaminants andeffects on biota, biomarkers show promise as indicators demonstrating that contaminants haveentered organisms, have been distributed among tissues, and are eliciting toxic effects at criticaltargets (McCarthy and Shugart 1990) In a recent review of biomarkers in the fisheries literature,Van der Oost, Beyer, and Vermeulen (2003) proposed six criteria that should be established foreach candidate biomarker:

1 The assay should be reliable, relatively inexpensive, and easy to perform

2 The response should be sensitive to pollutant exposure and/or effects in order to serve as

an early warning parameter

3 Baseline data of the biomarker should be well defined in order to distinguish betweennatural variability and contaminant-induced stress

4 The impacts of the confounding factors should be well established

5 The underlying mechanism of the relations between the response and pollutant exposureshould be established

6 The relation between the biomarker response and its long-term impact to the organismshould be established

Although there are numerous classification systems for biomarkers, we will follow the onerecently used by Van der Oost, Beyer, and Vermeulen (2003) that groups biomarkers into 1 of 10categories (Table 10.1).Many of these biomarker types have already been applied to unionids,but additional studies with multiple species are needed The available data suggest that this is

a promising avenue for future research

BIOTRANSFORMATION ENZYMES

To our knowledge, there have only been two studies that have examined the effects of contaminants

on enzymes associated with Phase I of biotransformation This first phase of metabolism involvesthe exposing or adding of reactive functional groups, through oxidation, reduction, or hydrolysis.The activity of 7-ethoxyresorufin O-deethylase (EROD) has been used as a biomarker in fish, andthese data suggest that EROD activity may not only indicate chemical exposure (primarily toorganic contaminants such as polycyclic aromatic hydrocarbons [PAHs], polychlorinated biphenyls[PCBs], polychlorinated dibenzo-p-dioxins [PCDDs] and polychlorinated dibenzofurans), but mayalso precede effects at various levels of biological organization (Whyte et al 2000; Van der Oost,Beyer, and Vermeulen 2003) In unionids, a 1.5-fold increase in EROD activity was observed in thedigestive gland of Elliptio complanata deployed for 62 days downstream of a municipal wasteeffluent, relative to mussels deployed upstream (Gagne´ et al 2002) These findings are consistentwith elevated levels of PAHs—a common constituent in sewage effluent and urban runoff Theseinitial data suggest that EROD activity may be a valuable indicator of exposure to organic contami-nants in unionids (Table 10.1) More recently, exposure of Unio tumidus to diethylhexylphthalateinduced the expression of CYP4 (a cytochrome P450 enzyme), however, no CYP1A sequence wasamplified in Aroclor-treated mussels (Chaty, Rodius, and Vasseur 2004)

The bulk of the published data on biotransformation enzymes in unionids involves thoseenzymes associated with the second phase of biotransformation This phase involves theconjugation (the addition of large and often polar chemical groups) of the xenobiotic parentcompound or its metabolite with an endogenous ligand The addition of more polar groups

TABLE 10.1

Categories of Biomarkers with Known or Potential Application to Unionids

Category of Biomarker Examples Reference in Unionids

Biotransformation enzymes

A Phase I Cytochrome P450 Chaty, Rodius, and Vasseur (2004)

Ethoxyresorufin O-deethylase (EROD) Gagne´ et al (2002) Aryl hydrocarbon hydroxylase (AHH) NA a

B Phase II Reduced (GSH) and oxidized (GSSG)

Catalase (CAT) Cossu et al (1997), Doyotte et al (1997) Glutathione peroxidase (GPOX) Cossu et al (1997, 2000), Doyotte et al.

(1997) Glutathione reductase (GRED) Cossu et al (1997, 2000), Doyotte et al.

(1997) Lipid peroxidation (LPOX) Cossu et al (1997, 2000), Doyotte et al.

(1997) Biotransformation products Polyaromatic hydrocarbon metabolites

a

Amino acids and proteins Amino acids Gardner, Miller, and Imlay (1981), Day,

Metcalfe, and Batchelor (1990) Stress proteins NA a

Metallothioneins (MT) Holwerda (1991), Couillard et al (1993,

1995a, 1995b), Malley et al (1993), Wang et al (1999), Gagne´ et al (2002), Perceval et al (2002)

Hematological Serum transaminases NA a

Alterations in the heme pathway Chamberland et al (1995) Immunological Cell- and humoral-mediated immunity NA a

Phagocytosis Blaise et al (2002) Lysosomal activity NA a

Reproductive and endocrine Imposex NA a

Vitellogenin Gagne´ et al (2001a, 2001b, 2001c),

Riffeser and Hock (2002), Blaise et al (2003)

Sexual competence NA a Neuromuscular Cholinesterases Doran et al (2001)

Genotoxic DNA damage Black et al (1996), Gagne´ et al (2002),

Rodius, Hammer, and Vasseur (2002) Irreversible genotoxic events NA a

Physiological and morphological Histopathology Lasee (1991)

Osmotic and ion regulation Malley, Huebner, and Donkersloot

(1988), Hemelraad et al (1990) Digestive processes Ma¨kela¨, Lindstro¨m-Seppa¨, and Oikari

(1992), Naimo, Atchison, and Bartels (1992), Milam and Farris (1998)

Holland-(continued)

generally facilitates the excretion of these chemicals by biota Many biotransformation enzymescan be induced or inhibited upon exposure to contaminants Enzyme induction is an increase in theamount or activity of these enzymes (or both), while inhibition refers to the blocking of enzymaticactivity, usually because of binding or complex formation with the inhibitor (Van der Oost, Beyer,and Vermeulen 2003) The primary Phase II enzymes that have been examined in unionids includereduced (GSH) and oxidized (GSSG) glutathione and glutathione S-transferase (GST,Table 10.1).

Reduced glutathione is a tripeptide whose major functions are to conjugate electrophilic mediates and to serve as an antioxidant This enzyme ensures the reduction of oxidants, thequenching of free radicals, the neutralization of organic peroxides, and the elimination of hydro-carbons by conjugation (Cossu et al 1997) Most of the research on the utility of Phase II enzymes

inter-as potential biomarkers in unionids comes from research on U tumidus in which reduced levels ofGSH in cytosolic and particulate fractions of the gills and digestive gland were reported after 15-and 30-day deployment at sites near and downstream from the outfall of a cokery (primarilycontaminated with PCBs and PAHs; Cossu et al 1997; Doyotte et al 1997) For example, GSHconcentrations were reduced in the cytosolic fraction by 79% in the gills and by 59% in thedigestive gland at the most polluted site (Cossu et al 1997) These decreases paralleled lipidperoxidation in the gills (see Oxidative Stress section), which reflected cell injury and toxiceffects in this tissue Similar results were found when U tumidus were transplanted to otherareas contaminated by effluents from a laundry and a foundry (Cossu et al 2000)

One of the detoxification enzymes that has been assayed in unionids is GST This is a family ofenzymes that catalyze the initial step of mercapturic acid synthesis—the conjugation of GSH with

TABLE 10.1 (Continued)

Category of Biomarker Examples Reference in Unionids

Condition indices Ma¨kela¨, Lindstro¨m-Seppa¨, and Oikari

(1992), Naimo, Atchison, and Bartels (1992), Couillard et al (1995a, 1995b), Hickey, Roper, and Buckland (1995), Hyo¨tyla¨inen, Karels, and Oikari (2002), Blaise et al (2003) Energetics Ma¨kela¨, Lindstro¨m-Seppa¨, and Oikari

(1992), Naimo, Atchison, and Bartels (1992), Hickey, Roper, and Buckland (1995), Gagne´ et al (2002, 2001b), Hyo¨tyla¨inen, Karels, and Oikari (2002)

Holland-Valve activity Balogh and Salanki (1984); Huebner and

Pynno¨nen (1992), Englund and Heino (1996), Ka´da´r et al (2001), Markich, (2003)

Growth Manly and George (1977); Foster and

Bates (1978), Muncaster, Hebert, and Lazar (1990), Lasee (1991); Couillard

et al (1995a), Beckvar et al (2000), Gagne´ et al (2001b), Bartsch et al (2003), Newton et al (2003)

a Not available.

Source: Modified from Van der Oost, R., Beyer, J., and Vermeulen, N P E., Environ Toxicol Pharmacol, 13, 57–149, 2003 With permission.

xenobiotics and their metabolites Their primary roles are defense against oxidative damage andperoxidative products of DNA and lipids GST activity was significantly reduced (~20%) in thedigestive gland of Anodonta anatina deployed for 4 months at a site 20 km downstream of ableached kraft pulp and paper mill, but not at a site 5 km downstream, compared to control sites(Figure 10.2; Ma¨kela¨, Lindstro¨m-Seppa¨, and Oikari 1992) Given that the spatial extent of thiseffect was limited, the authors concluded that organically bound chlorine (a major constituent of theeffluent from bleached kraft pulp and paper mills) did not consistently induce GST activity

in unionids

OXIDATIVE STRESS

Many environmental contaminants have been shown to exert toxic effects through oxidative stress.Antioxidant defenses include antioxidant enzymes (superoxide dismutase [SOD], catalase [CAT],glutathione peroxidase [GPOX], and glutathione reductase [GRED]) and free radical scavengers(vitamins C and E, carotenoids, and glutathione), whose function is to remove reactive oxygenspecies thus protecting organisms from oxidative stress (Doyotte et al 1997) SOD catalyzes theconversion of reactive superoxide anions into hydrogen peroxide, which in turn is detoxified byCAT Hydrogen peroxide and hydroperoxides are destroyed by GPOXs in the presence of GSH.Glutathione is regenerated by GRED When antioxidant systems are impaired, oxidative stress mayalso produce lipid peroxidation (LPOX), or the oxidation of polyunsaturated fatty acids

Catalase is often one of the earliest antioxidant enzymes to be induced and has been shown to beinduced in Mytilus sp exposed to organic pollution (Porte et al 1991) Catalase and SOD activitywere measured in U tumidus after in situ deployment at sites upstream and downstream of effluentfrom a cokery In one study, CAT and SOD(Figure 10.3)were significantly reduced at the mostpolluted site, relative to mussels deployed upstream (Cossu et al 1997) In the second study, SOD andCAT were markedly unchanged upon exposure to the cokery effluent, suggesting that these enzymeswere not sensitive to short-term exposure to the chemicals contained therein (Doyotte et al 1997).Interestingly, the experimental design for these two studies were nearly identical, with the exceptionthat mussels in the latter study were deployed for 7 days, whereas mussels in the Cossu et al

Distance (km) from pulp mill 0

100 50

150 200 250

deploy-(1997) study were deployed for 15 and 30 days These data suggest that the temporal variation

in these oxidative enzymes may make it difficult to detect treatment effects at shorterexposure durations

Selenium-dependent GPOX, GRED, and GSH levels appear to be early biomarkers of exposure

to pollutants in unionids For example, deployment of U tumidus to sites upstream and downstream

of effluent from a cokery resulted in significant decreases in Se-dependent GPOX and GREDactivities (Cossu et al 1997; Doyotte et al 1997) Similar exposure of U tumidus to effluentsfrom other sources (a laundry and a foundry) also resulted in significant reductions in Se-dependentGPOX and GRED activities (by 70 and 80%, respectively) (Cossu et al 2000) However, in thelatter study, a relation among the antioxidant response and the degree and type of contamination insediments was not consistently observed, suggesting that these effects could result from unidenti-fied contaminants and/or issues associated with contaminant bioavailability

Lipid peroxidation is an important outcome of oxidative stress because it can demonstratethe ability of a single radical species to propagate a number of adverse biochemical reactions(Van der Oost, Beyer, and Vermeulen 2003) Studies suggest that LPOX has considerablepotential as a biomarker for environmental risk assessment (Stegeman et al 1992), although itcan result as a consequence of cellular damage because of a variety of stressors other thanexposure to contaminants Numerous studies in the fisheries literature have demonstrated anenhancement in LPOX as a function of contaminant exposure (see references in Van der Oost,Beyer, and Vermeulen 2003), and the preliminary data on unionids are promising In the first ofthree studies on antioxidant enzymes in U tumidus, LPOX (as expressed by malonaldehydecontent, MDA) did not differ between reference unionids and those exposed to a complexindustrial effluent (Doyotte et al 1997) It is possible that the lack of an effect in this studywas a function of the 7-day deployment period; in subsequent studies, unionids were deployed for

at least 15 days In a follow-up study, decreases in Se-dependent GPOX and GRED activities andGSH levels were associated with a three-fold increase in MDA content in the gills and with ahigh level of contamination of sediments by PAHs and PCBs (Cossu et al 1997) Similarly,

FIGURE 10.3 Superoxide dismutase activity in the cytosolic fraction of the gills of Unio tumidus deployed for

15 or 30 days to four sites along the Fensch River, France Site A was a reference, Site B was upstream of acomplex cokery effluent, Site C was downstream near the outfall, and Site D was about 2.5 km downstream.Asterisk indicates a significant difference from control (Adapted from Cossu, C., Doyotte, A., Jacquine, M C.,Babut, M., Exinger, A., and Vasseur, P., Ecotoxicol Environ Saf., 38, 122–131, 1997.)

Cossu et al (2000) found elevated levels of MDA in unionids deployed in four rivers in Francewith various pollution sources In particular, in one river primarily contaminated by metals,deficiencies in antioxidant defenses resulted in dramatic lipid peroxidation—MDA concen-trations ranged from 29 to 85 ng/mg protein downstream of the source, while the controls didnot exceed 5 ng/mg protein The elevated levels of MDA, coupled with decreases in Se-depen-dent GPOX and GRED activities and GSH levels, suggest that these antioxidant parameters may

be useful biomarkers of exposure

AMINO ACIDS AND PROTEINS

Changes in the concentrations of free amino acids in the gill, mantle, and adductor muscle in marinebivalves have been used as biomarkers of exposure to contaminated environments (Livingstone

1985 and references therein) For example, the ratio of taurine to glycine has been used in marinebivalves to indicate biochemical responses to hydrocarbons (Widdows et al 1982) To our knowl-edge, two studies have measured free amino acid concentrations in unionids and neither measuredthis ratio; thus, the utility of this ratio in unionids is unclear but deserves future study Both unionidstudies suggest that increases or decreases in total-free amino acids in some tissues (especially themantle and adductor muscle) may be indicative of generalized stress induced by a variety ofenvironmental factors (e.g., starvation or increased temperature) and may be useful as an in situbiochemical index of toxicity (Gardner, Miller, and Imlay 1981; Day, Metcalfe, and Batchelor1990) As with many aspects of unionid biochemistry and physiology, the concentrations of mostfree amino acids vary seasonally Thus, the baseline seasonal variation needs to be characterized

in a given population prior to attributing changes in amino acid concentrations to induced effects (biomarker criterion 3 in Biomarker Concept)

contaminant-The molecules that probably offer the greatest potential for monitoring biological effects andhave attracted the most attention are enzymes and other functional proteins such as metallothio-neins (MT) (Livingstone 1985) Freshwater and marine molluscs are known to accumulate metalsfrom their environment The tolerance of the resulting body burden has been attributed to theexistence of an effective detoxification system (Viarengo 1989) Metallothioneins are low molecu-lar weight, cysteine-rich metal binding proteins that function as a detoxification mechanism

by sequestering divalent metals through specific ligands present in the cytosol They are thought

to provide one of two protective functions (1) interception and binding of free metal ions that areinitially taken up by the cell and (2) removal of metals from non-thionein ligands that includecellular targets of toxicity (Van der Oost, Beyer, and Vermeulen 2003)

One of the first studies to examine the role of metal-binding proteins in unionids (A cygnea)found that Cd and Cu were bound to different fractions upon laboratory exposure For example,

Cu was generally bound to the high molecular weight fraction, whereas Cd was mainly bound

to a specific metal-binding, carbohydrate-containing protein fraction of Mr~11,000 (Holwerda1991) The Cd-binding protein was similar to the metal-binding proteins observed in Mytilus(Roesijadi and Hall 1981) and Crassostrea (Ridlington and Fowler 1979) In one of the firstreports of MT-induction in unionids, Malley et al (1993) showed that a 22-day exposure of

A grandis grandis to waterborne Cd induced MT in the gills Although MT was measured inthe mantle, gills, foot, kidney, visceral mass, and whole body, only MT in the gills increasedsignificantly with increasing exposure to Cd More recently, the time course for MT inductionhas been examined Before induction of MT (~14 d) in Pyganodon grandis, Cd was primarilybound to the high molecular weight fraction, and after induction (~90 d) all the Cd had apparentlyshifted to fractions of moderate molecular size (15 to 3 kDa)—which corresponds to the expectedMT-fraction (Couillard et al 1995a)

Field studies support experimental data that MTs sequester heavy metals and that MT levelscorrelate with tissue levels of heavy metals Induction of MT in unionids has been observed in field

locations where mussels were exposed to Cd (Malley et al 1993; Couillard et al 1995b) and urbaneffluents (Gagne´ et al 2002) Concentrations of MT generally correlate with tissue Cd concen-trations, but not with tissue concentrations of Cu or Zn (Figure 10.4; Couillard, Campbell, andTessier 1993; Wang et al 1999) Variation in MT concentrations among field locations is stronglycorrelated with free Cd2C (Couillard, Campbell, and Tessier 1993; Wang et al 1999) Mostrecently, Perceval et al (2002) examined the relative influence of limnological and geochemicalconfounding factors on MT synthesis in natural populations of P grandis in lakes along a Cdconcentration gradient (biomarker criterion 4 in Biomarker Concept) Predictive models found thatdissolved Ca (K) and free Cd2C(C) explained 62% of the variation in MT These data can be used

in monitoring programs to select field sites to reduce the relative influence of factors that confound

MT concentrations (Perceval et al 2004)

0 500 1000 1500 2000 2500 0

100 200 300

400 Cadmium

p < 0.01

0 1000 2000 3000 4000 5000 6000 0

100 200 300 400

Metal concentration in gills (nmol/g dry wt)

Copper

p > 0.05

Zinc

p > 0.05

FIGURE 10.4 Concentration of metallothionein as a function of Cd, Cu, and Zn in the gills of A grandis from

11 lacustrine sites along a geochemical gradient of pH, Cd, Cu, and Zn in an area influenced by mining andsmelting (Adapted from Couillard, Y., Campbell, P G C., and Tessier, A., Limnol Oceanogr., 38, 299–313,1993.)

The heme biosynthesis pathway is sensitive to organic and inorganic contaminants, and porphyrinintermediates of this pathway have been used as biomarkers of chemical stress in mammals.Chamberland et al (1995) identified and quantified seven porphyrin profiles in A g grandis intwo reference lakes and in a lake in which Cd (57–177 ng/L) had been added for 5 years Abnormalporphyrin profiles (especially hepatocarboxyporphyrin) were apparent in mussels from theCd-treated lake, relative to mussels in the reference lakes, and these changes were consistentwith a Cd-induced biochemical response These data suggest that a mussel’s porphyrin profilehas potential as an ecotoxicological biomarker for Cd and possibly other metals

IMMUNOLOGICAL

The highly regulated nature of the immune system renders it vulnerable to exposure to mental contaminants In marine bivalves, hemocytes circulating in the hemolymph are the maincomponents of the immune system—these cells are capable of several immune functions, includingphagocytosis and cytotoxicity through the generation of reactive oxygen intermediates (Cheng1977) Contaminants that adversely affect the immune system can ultimately make unionidsmore susceptible to infectious diseases Blaise et al (2002) developed a miniaturized immunocom-petence assay that examines the effects of contaminants on three critical targets of the immunesystem: capacity to ingest bacteria (phagocytosis activity), cell viability, and hemolymph bacterialevels In the laboratory portion of that study, exposure of E complanata to a municipal effluent led

environ-to immunosuppression (phagocyenviron-tosis activity and cell viability were reduced) In the field,

E complanata deployed upstream and downstream of the effluent for 62 days resulted in bothimmunosuppressive and immunostimulative effects; low concentrations of heavy metals and micro-organisms presumably caused immunostimulation, whereas immunosuppressive effects appearedafter exposure to higher levels of contaminants and microorganisms This study suggests thatcontaminants may compromise the immune system of unionids Future studies should determinethe contaminant concentrations that contribute to irreversible damage to the immune systemand how these effects may be translated into population-level effects (biomarker criterion 6 inBiomarker Concept)

REPRODUCTIVE AND ENDOCRINE

Since the early 1990s, considerable effort has been directed towards understanding the effects ofendocrine-disrupting chemicals on aquatic biota For example, high levels of serum vitellogenin(an estrogen-inducible egg protein precursor) were found in male carp in a river contaminated by

a municipal effluent (Folmar et al 1996) Much of this research has focused on surface watersdownstream from municipal effluents and agricultural fields because the concentrations of potentialxenoestrogens in these environments may be sufficient to produce biological effects, such asalterations in sexual differentiation, gonad development, and oocyte growth In certain invert-ebrates, the presence of estrogens trigger the synthesis of vitellogenin In marine gastropods, lowlevels of organotin compounds (e.g., 0.02 mg/L tin) can induce imposex—the manifestation of malemorphological sex characters in females (Bryan et al 1986) Comparatively less is known about theeffects of endocrine-disrupting chemicals on unionids, although advances in this area are ongoingand seem promising

Most of our knowledge on the effects of contaminants on reproductive-related processes inunionids comes from several studies evaluating the effects of a municipal effluent on

E complanata In one study, extracts from surface waters upstream and downstream of a municipaleffluent were injected into mussels and the effluent emitted relatively high levels of estradiolcompetitors that were able to compete with estradiol-binding sites in gonad cytosols and induced

vitellogenin (Gagne´ et al 2001a) Moreover, the estrogen-competing potential of the extracts wassignificantly correlated with total and fecal coliform bacteria in the effluent and with levels ofvitellogenin in the hemolymph In a combined laboratory and field study, Gagne´ et al (2001b)found that injections of 17b-estradiol and p-nonylphenol and exposure to the effluent increasedvitellogenin-like proteins in the hemolymph of male and female E complanata Interestingly,females responded more readily to effluent treatment than males, which suggests that continuousexposure of mussel populations to urban effluents not only has the potential to elicit estrogen-mediated effects that include vitellogenin synthesis, but also may cause a shift in the male-to-fe-male sex ratio Recently, contaminants in the effluent plume were reported to lead to feminization

of E complanata populations (Blaise et al 2003) Adults deployed for 1 year at sites 8 and 10 kmdownstream from the effluent plume had an elevated gonado-somatic index, increased conditionfactor, and significantly elevated levels of vitellogenin-related proteins in gonads, relative tomussels deployed upstream Also, the proportion of females increased from 41% at the upstreamsite to 66% at the furthest downstream site Environmental chemicals with the potential to interferewith the endocrine system can mimic estrogen or inhibit androgen Most of the current research is

on estrogen; future research should evaluate both estrogenic and androgenic effects on unionids.One compound of concern in municipal effluents is coprostanol (5b [H]-cholestan-3b ol)—areduced metabolite of cholesterol that is produced by microorganisms in the intestinal tract ofmammals Coprostanol was found in significantly elevated concentrations in adult E complanataafter a 62-day deployment downstream of a municipal effluent (Gagne´ et al 2001c) In the labora-tory, coprostanol was shown to have a relatively weak binding potential to estradiol-binding sites,although its injection into the adductor muscle led to increased levels of vitellogenin-like proteins

in mussel gonads and hemolymph This implies that coprostanol may be biologically transformedinto a more potent estrogenic metabolite and may be metabolized in the gonads of unionids Thesestudies suggest that continuous exposure to the numerous contaminants and microorganisms inmunicipal effluents may have deleterious effects on reproductive processes in unionids that mayultimately be manifested in population-level effects

In contrast to these findings, no changes in protein patterns were observed in A cygnea inresponse to exposure to 17b-estradiol or effluents from a wastewater treatment plant (Riffeser andHock 2002) These authors suggested that hemolymph may not be the carrier of major egg yolkproteins in bivalves, rather the presence of vitellogenin-like proteins in hemolymph may originatefrom break-down products of gonadal tissue Future mechanistic-based research with unionidsshould help clarify this issue

Female unionids are generally classified into one of two reproductive categories (1) those thatcarry glochidia over winter and release them in the spring or early summer (long-term brooders)and (2) those that carry glochidia for a short duration in the spring and release them in summer(short-term brooders; Chapter 3).Thus, it is often possible to choose a species for study that isgravid during the time frame of interest Because females of certain species use either the inner orouter gills, or both, to hold the developing glochidia, a gill index (mass of outer gill divided by themass of the inner gill) has been proposed as an index of reproductive potential in those species whohold glochidia in their outer gills (Ma¨kela¨, Lindstro¨m-Seppa¨, and Oikari 1992) However, in situexposure of A anatina (a long-term brooder) did not adversely affect the gill index among indi-viduals deployed upstream and downstream of a pulp mill effluent, nor were any gross histologicalchanges in gonad development observed (Ma¨kela¨, Lindstro¨m-Seppa¨, and Oikari 1992) This indexrequires additional study in other unionid species

NEUROMUSCULAR

Cholinesterases are a group of enzymes whose function is to transmit nerve impulses Inhibition ofcholinesterase prevents the metabolic breakdown of choline neurotransmitters and prolongs signal

transmissions through the synapse, which effectively causes paralysis through over-stimulation ofthe nervous system In the field, organophosphate and carbamate insecticides have been implicated

in the mortality of ~1000 Elliptio spp in North Carolina, presumably by inhibiting the mussels’cholinesterases (Fleming, Augspurger, and Alderman 1995) In the laboratory, exposure ofAmblema plicata to chlorpyrifos, an organophosphorus insecticide, resulted in a significantreduction in acetylcholinesterase activity in the adductor muscle after 96 hours (Figure 10.5), butthe exposure did not follow a concentration-response relation (Doran et al 2001) The authors notedthat acetylcholinesterase activity was highly variable among individuals and suggested that eithersample sizes need to be increased or different tissues should be assayed This is a promisingbiomarker of exposure, and additional research with cholinesterase inhibition in unionids is needed

GENOTOXIC

In marine bivalves, the Comet assay (alkaline single-cell gel electrophoresis) has been a frequentlyused biomarker to assess the effects of contaminants on DNA strand breakage (Mitchelmore et al.1998; Shaw et al 2000) Although substantially less research has been done with freshwatermolluscs, recent genotoxicity studies show that contaminants can indeed adversely effectgenomic DNA For example, exposure of A grandis to 50mg Pb/L in the laboratory showedsignificant DNA strand breakage and a reduction in strand length after 28 days (Black et al.1996) However, no damage was observed upon exposure to higher Pb concentrations (500 and5,000 mg/L), suggesting a threshold effect for DNA damage and repair In contrast, Gagne´ et al.(2002) observed that levels of DNA damage (measured as DNA strand breaks) in the digestivegland of E complanata were significantly reduced in individuals exposed for 62 days to a municipaleffluent plume compared to mussels deployed upstream Recently, a technique termed RNA arbi-trarily primed polymerase chain reaction was developed to detect DNA damage and variations ingene expression in U tumidus in response to exposure to toxic contaminants (Rodius, Hammer, andVasseur 2002) These authors report variations corresponding to the loss of polymerase chainreaction products in some individuals deployed in situ for 14 days in a river polluted by PAHs,PCBs, and metals, relative to controls Although sample sizes were small, this method showspromise as a new technique to explore the genotoxic effects of contaminants on unionids More

Chlorpyrifos concentration (mg/L) 0.0 0.2 0.4 0.6 0.8 1.0 1.2

1200

Amblema plicata

FIGURE 10.5 Acetylcholinesterase activity in Amblema plicata after 96-hour exposure to chloropyrifos, anorganophosphorus insecticide (Adapted from Doran, W J., Cope, W G., Rada, R G., and Sandheinrich M B.,Ecotoxicol Environ Saf., 49, 91–98, 2001.)

methods development, concentration-response studies with model chemicals, and documentationthat genotoxic effects on individuals ultimately influence populations are still required (biomarkercriterion 6 in Biomarker Concept).

PHYSIOLOGICAL AND MORPHOLOGICAL

From the mid-1970s, aquatic ecotoxicology has increasingly used the tools of physiologists.Although this was partly done to understand why an organism is physiologically stressed, it wasalso because of the realization that many sublethal effects were apparent before death of theindividual (biomarker criterion 2 in Biomarker Concept) Widdows (1985) suggested that physio-logical responses have three important attributes in providing an assessment of an individual’scondition: (1) they represent an integration of the many cellular and biochemical processes that canalter in response to changes in the environment, (2) they represent non-specific responses to the sum

of environmental stimuli that are complementary to more specific responses at the biochemicallevels, and (3) they are capable of reflecting deterioration in the environment before effects mani-fest themselves in the population or the community In this section, we review the responses ofunionids to contaminants on histopathology, osmotic and ion regulation, digestive processes, con-dition indices, energetics, valve activity, and growth

HISTOPATHOLOGY

We are aware of only one study that documented the histopathological effects of contaminants onunionids In that study, Lasee (1991) evaluated the effects of Cd on the histopathology of 0-day-oldjuvenile Lampsilis ventricosa in 7-day laboratory tests Exposures as low as 30 mg Cd/L resulted

in the dissolution and loss of the crystalline style; vacuolization, necrosis, and tissue separations inmantle, ganglia, and digestive glands; altered lipid catabolism; hypersecretion of mucus; andreduced feeding While this study suggested that Cd may adversely affect several tissueand organ features, it is unclear how, or if, these changes translate into individual- or popu-lation-level effects Substantial research needs to be directed at closing the gap betweenhistological events, such as the loss of the crystalline style, and its influence on feeding andultimately growth

OSMOTIC ANDIONREGULATION

Because unionids inhabit freshwater systems with low ionic strength, they must accumulate andretain solutes (i.e., NaCand ClK) To preserve their water balance, they need to excrete a volume ofwater equivalent to that taken up osmotically in order to maintain ionic homeostasis Under normalconditions, unionids are able to obtain ions from food and through epithelial ion transportsystems, thereby compensating for ion losses (Murphy and Dietz 1976) Exposure of A cygnea

to 50 mg Cd/L for 12 weeks resulted in substantial changes in the composition of 27 elements(Hemelraad et al 1990) In particular, NaCconcentrations in hemolymph declined 55% from 20

to 9 mM The disruption of NaCtransport may have additional consequences because the transport

of salts (including NaC) has been intimately linked to acid–base balance in unionids (Dietz 1985).The 3-day addition of aluminum sulfate (alum) to an experimental lake resulted in a substantialincrease in Ca2C(from 8 to 15 meq/L) but decreases in NaCand ClKin the hemolymph of A g.grandis; most values returned to near normal after 21 days (Malley, Huebner, and Donkersloot1988) The increase in blood Ca2Cwas probably a result of the dissolution of Ca2Cstores (as aresult of the lower pH associated with the alum addition) in the shell or mantle These data suggestthat contaminants can disrupt the osmotic balance in unionids, at least for some period of time.Additionally, because ion balance in unionids is probably coupled to acid–base balance,