A

amwib

dfTaet©

K

1

st0ni2t2

0d

Aquatic Toxicology 86 (2008) 370–378

Effects of waterborne uranium on survival, growth, reproduction andphysiological processes of the freshwater cladoceran Daphnia magna

Florence Anna Zeman a, Rodolphe Gilbin a,∗, Frederic Alonzo b,Catherine Lecomte-Pradines a, Jacqueline Garnier-Laplace a, Catherine Aliaume c

a Laboratoire de Radioecologie et Ecotoxicologie, Institut de Radioprotection et Surete Nucleaire, Cadarache, Bat 186,BP 3, 13115 Saint-Paul-lez-Durance Cedex, France

b Laboratoire de Modelisation Environnementale, Institut de Radioprotection et Surete Nucleaire, Cadarache, Bat 159,BP 3, 13115 Saint-Paul-lez-Durance Cedex, France

c UMR CNRS-UMII 5119 Ecosystemes Lagunaires, Universite Montpellier II, 34095 Montpellier Cedex 5, France

Received 14 May 2007; received in revised form 26 November 2007; accepted 27 November 2007

bstract

Acute uranium toxicity (48 h immobilisation test) for Daphnia magna was determined in two different exposure media, differing in pH andlkalinity. LC50 varied strongly between media, from 390 ± 40 �g L−1 U at pH 7 to 7.8 ± 3.2 mg L−1 U at pH 8. According to the free ion activityodel uranium toxicity varies as a function of free uranyl concentration. This assumption was examined by calculating uranium speciation in ourater conditions and in those reported in the literature. Predicted changes in free uranyl concentration could not solely explain observed differences

n toxicity, which might be due to a competition or a non-competitive inhibition of H+ for uranium transport and/or the involvement of otherioavailable chemical species of uranium.

Chronic effects of uranium at pH 7 on mortality, ingestion and respiration, fecundity and dry mass of females, eggs and neonates were investigateduring 21-day exposure experiments. A mortality of 10% was observed at 100 �g L−1 U and EC10 for reproduction was 14 ± 7 �g L−1 U. Scopeor growth was affected through a reduction in feeding activity and an increase in oxygen consumption at 25 �g L−1 U after 7 days of exposure.

−1

his had strong consequences for somatic growth and reproduction, which decreased, respectively, by 50% and 65% at 50 �g L U after 7 daysnd at 25 �g L−1 U after 21 days. Uranium bioaccumulation was quantified and associated internal alpha dose rates from 2.1 to 13 �Gy h−1 werestimated. Compared to the toxicity of other alpha-emitting radionuclides and stable trace metals, our results confirmed the general assumptionhat uranium chemical toxicity predominates over its radiotoxicity.

2007 Elsevier B.V. All rights reserved.

Ener

raainFma

eywords: Daphnia magna; Uranium; Chemical speciation; Scope for growth;

. Introduction

Uranium is a naturally occurring metal from the actinideeries and is composed of three alpha-emitting radioactive iso-opes, 238U, 235U and 234U, respectively, contributing 99.27%,.72% and 0.0055% of mass (Colle et al., 2001). Its behaviour inatural ecosystems has been extensively studied and describedn several reviews (Colle et al., 2001; Ragnarsdottir and Charlet,

000). Uranium is ubiquitous in natural waters at trace concen-rations, ranging from 0.02 to 6 �g L−1 U (Bonin and Blanc,001). Locally, higher concentrations may reach 2 mg L−1 U,

∗ Corresponding author. Tel.: +33 4 42 19 95 37; fax: +33 4 42 19 91 51.E-mail address: rodolphe.gilbin@irsn.fr (R. Gilbin).

cgpeil

166-445X/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.aquatox.2007.11.018

gy allocation

eflecting mainly the composition of surrounding rocks (Boninnd Blanc, 2001; WHO, 2001). Uranium concentration maylso increase in some ecosystems due to anthropogenic activ-ties such as mining, extraction and processing of uranium foruclear fuel and weapons, as well as spent fuel reprocessing.or two decades, the chemical toxicity of uranium, as a heavyetal, has become of increasing concern (Environment-Canada

nd Health-Canada, 2000; Sheppard et al., 2005).In freshwater organisms, ecotoxicological data for acute and

hronic exposure concern a wide range of endpoints and showreat variability, notably due to differences in the chemical com-

osition of the exposure medium. In a recent review, Sheppardt al. (2005) showed that the sensitivity of organisms to uraniums dependent on several environmental parameters such as alka-inity (due to complexation of uranyl ion UO2

2+ with soluble

Toxic

can41Lclc

nei2(tmnat3o

twnaswedcdssopCa

lpcoarrm

2

2

Hrm

hp0SSCac(

rd(ratandmb

2

Facp1toiaaist2lryaanw

2

mnw

F.A. Zeman et al. / Aquatic

arbonates), and hardness (due to its competition with calciumnd magnesium). For example, in Daphnia magna water hard-ess and alkalinity reduce the acute toxicity of uranium, with8 h LC50 increasing from 6.4 to 51.9 mg L−1 U (Poston et al.,984). Barata et al. (1998) reported a similar value of acuteC50, ranging from 8.3 to 22.4 mg L−1 U dependent on daphnidlone and water hardness. However, as pH, hardness and alka-inity varied concomitantly, the effects of complexation and/orompetition on uranium toxicity are difficult to differentiate.

Discrepancies between ecotoxicity data on uranium origi-ate also from the diversity in tested endpoints and time ofxposure. In D. magna, standard ecotoxicological tests for thedentification of chronic effects are based on the outcome of1-day reproduction. As previously studied by Poston et al.1984), reproduction inhibition was shown at uranium concen-rations from 0.5 to 3.5 mg L−1 U. Reproduction was shown to beore sensitive for another freshwater cladoceran, Moinodaph-

ia macleayi with a 6-day LOEC of 20–49 �g L−1 U (Semaan etl., 2001). In Ceriodaphnia dubia, reported effects on reproduc-ion strongly differed between studies, with a 7-day LOEC of.91 mg L−1 U (Kuhne et al., 2002) and a chronic (7-day) EC25f 3 �g L−1 U (Pickett et al., 1993).

In the literature, effects of uranium are commonly reportedo concentrations, as it is done for trace metals, in accordanceith the assumption that chemotoxicity of uranium predomi-ates over its radiotoxicity. A small part of toxicity of uranium,s a radioelement, might be imputable to the radiological expo-ure. Sublethal effects of chronic alpha and gamma irradiationere recently reported in D. magna (Alonzo et al., 2006; Gilbin

t al., in press). A comparison with these works requires thatose rates are quantified for uranium, based on accumulatedoncentration in tissues. At dose rates ≥0.9 mGy h−1, alpha irra-iation induced a reduction in somatic growth with potentialtrong consequences for energy allocation in organisms. Suchtudies of individual energy budgets have increasingly been usedver the last decade to link effects of pollutants on physiologicalrocesses to growth and reproduction (Calow and Sibly, 1990;alow, 1991; Kooijman, 2000; Knops et al., 2001; Baillieul etl., 2005).

The objectives of this work are: (1) to modulate pH and alka-inity of exposure medium within the tolerance range of daphnidhysiology, in order to increase bioavailability and acute toxi-ity of uranium to D. magna; (2) to determine chronic effectsf uranium on D. magna survival, reproduction, somatic growthnd individual endpoints governing energy budget (ingestion,espiration); (3) to quantify uranium uptake and associated doseate in daphnids and compare toxicity with those of stable traceetals and radiological stressors.

. Materials and methods

.1. D. magna culture

D. magna cultures (clone obtained from INERIS Verneuil enalatte, France) were maintained in continuous parthenogenic

eproduction in artificial freshwater at pH 8 (Elendt, 1990; M4edium, hereafter ‘M4-pH8’) and pH 7 (modified M4 medium,

(fwC

ology 86 (2008) 370–378 371

ereafter ‘M4-pH7’) renewed twice a week. Composition of M4-H8 was: 2 mM Ca, 0.5 mM Mg, 0.87 mM Na, 0.081 mM K,.51 mM SO4, 3.2 �M NO3, 2.1 �M PO4, 4.9 nM NH4, 35 �MiO3, 46 �M B, 1.8 �M Mn, 7.2 �M Li, 0.59 �M Rb, 0.57 �Mr, 0.16 �M Br, 0.31 �M Mo, 98 nM Cu, 50 nM Zn, 43 nMo, 19 nM I, 11 nM Se, 4.9 nM V, 7.2 �M Fe, 13.4 �M EDTAnd 4.1 mM Cl. Composition of M4-pH7 differed only in a Cloncentration of 4.8 mM. Media were at equilibrium with airpCO2 = 3.16 × 10−4 atm).

Daphnids were fed daily with green algae Chlamydomonaseinhardtii (Dangeard, strain 11/32B from CCAP, United King-om). Algae under exponential growth phase were centrifuged15 min, 1000 × g) and resuspended in M4 to achieve a dailyation of 100 �g carbon per daphnid. Neonates were removednd the medium was renewed twice a week. Cultures were main-ained at 20 ◦C (±1 ◦C), a photoperiod of 16 h-light:8 h-darknd a light intensity of 30 �E m−2 s−1. Reproductive rates >60eonates per adult over 21 days were confirmed in the two mediauring 1 year, as a criterion for population health. All experi-ents were started with juveniles (<24-h old) from the fourth

rood.

.2. Exposure conditions

Uranium was obtained from Sigma–Aldrich (Saint-Quentinallavier, France) as uranyl nitrate hexahydrate in 0.2% nitriccid solution (1 g L−1 U). Uranium acute toxicity was studied atoncentrations ranging from 100 �g L−1 to 100 mg L−1 in M4-H8 and M4-pH7. Chronic toxicity was examined from 10 to00 �g L−1 U in M4-pH7. For all test conditions including con-rols, nitrate concentration was adjusted to 18 �M in M4-pH7r 1 mM in M4-pH8. This was done to eliminate differencesn NO3

− concentration associated with uranium spikes (addeds uranyl nitrate hexahydrate). This addition of nitrate did notffect survival of daphnids in the control. Uranium and majoronic concentrations were quantified prior and after 48 h expo-ure in acute condition or daily in freshly renewed medium andwice weekly after 24 h exposure in chronic conditions (filtration�m), by ICP-AES (Optima 4300DV, PerkinElmer—detection

imit = 10 �g L−1 and 0.5 mg L−1, for U and major cations,espectively) and ionic chromatography (Dionex DX-120, Sun-vale, CA, USA—quantification limit = 100 �g L−1 for majornions). All water samples were stored at 4 ◦C in darkness beforenalysis. All concentrations remained within 10% of nomi-al concentrations. pH was similarly monitored and remainedithin 0.1 unit of nominal pH.

.3. Modeling of aqueous speciation

Because of a lack of practicable techniques to directlyeasure individual uranium chemical species in solution, ura-

ium speciation in water conditions of the exposure mediumas predicted using the geochemical speciation code J-Chess

Java Chemical Speciation Equilibrium Speciation with Sur-aces, Van der Lee, 1998). A consistent thermodynamic databaseas compiled of the OECD-NEA (Organization for Economicooperation and Development-Nuclear Energy Agency, 1996)

3 Toxic

a2pJatuad

2

bOt6p7ebavg

2

2

cn2remTpraw

2

aopZtVscm(

C

wk

a

Ie

C

waicm

2

umpotuauaOg2R

R

wsc(o

[

ta

2

dpbcmwdcG

72 F.A. Zeman et al. / Aquatic

nd updated in a recent review (Denison and Garnier-Laplace,005). The input parameters for J-Chess were based on measuredhysiochemical data (temperature, pH and ion concentrations.-Chess calculations were constrained to equilibrium with thetmosphere (pCO2 = 3.16 × 10−4 atm). Calculations estimatedhe solubility limit and concentrations of complexes and freeranyl ion at the 48 h LC50 uranium concentration in M4-pH8nd M4-pH7. Speciation was similarly predicted for water con-itions reported by Poston et al. (1984) and Barata et al. (1998).

.4. Acute tests

Acute toxicity was evaluated in M4-pH7 and M4-pH8y determining the 48 h LC50 of uranium according to theECD methodology (OECD, 1998). Tested uranium concen-

rations were: 97 ± 4, 163 ± 2, 298 ± 3, 485 ± 13, 560 ± 10,71 ± 20, 757 ± 22, 864 ± 30 and 935 ± 30 �g L−1 in M4-H7; 10.1 ± 0.1, 16.4 ± 0.2, 39.7 ± 0.5, 47.9 ± 0.8, 56.5 ± 0.4,3.6 ± 1.6, 92.9 ± 2.6 and 109.0 ± 0.4 mg L−1 in M4-pH8. Forach condition, 10 groups of 5 animals were exposed in polycar-onate tubes containing 10 mL medium, without renewal. After24 and 48 h exposure, the number of surviving animals in eachial was counted, considering that animals without response toentle agitation after 15 s were dead.

.5. Chronic tests

.5.1. Experimental designChronic tests were performed in M4-pH7 at five uranium

oncentrations and a control, with medium renewed and daph-ids fed every day. Tested concentrations were: 10.1 ± 1.2,5.3 ± 3.8, 50.3 ± 5.1, 74.7 ± 7.5 and 101.7 ± 10.1 �g L−1. Foreproduction tests, 10 daphnids per condition were individuallyxposed for 23 days in polycarbonate bottles (50-mL). Survival,oulting and number of neonates produced were checked daily.hree batches of 10 daphnids per condition were exposed inolycarbonate bottles (500-mL) and sampled within 24 h ofelease of brood 1, 3 or 5 for the quantification of respirationnd dry masses. No significant difference in survival or fecundityas observed between 50 and 500 mL bottles.

.5.2. Ingestion rateFeeding activity was quantified at release of broods 1, 3

nd 5 in five individual 50-mL bottles per condition, basedn changes in algal density between t = 0 (when food was sup-lied) and 6 h. Algal densities were measured using a Coulter2 particle counter (1:10 dilution with ISOTON II isotonic solu-

ion, using a 100 �m orifice tube; Beckman Coulter France SA,illepinte, France). The algal diameter was checked to be con-

tant (∼5 �m). Particles between 3 and 10 �m diameter wereounted in order to get rid most of the cell debris. Exponentialodels were fitted to observed algal densities according to Frost

1972):

t = C0 ekt, Ct = C0 e(k−f )t

here C0 and Ct are algal densities at time 0 and t, respectively,(h−1) the growth constant measured in vials without daphnids

2

b

ology 86 (2008) 370–378

nd f (h−1) the daphnid grazing coefficient in 50-mL bottles.Mean algal density C (cell mL−1) and daphnid ingestion

(cell daphnid−1 day−1) were calculated using the followingquations:

= Ct − C0

(k − f )tand I = VfCt

here V is the experimental volume per individual (50 mL) andssuming a feeding time t = 10 h per day. Values were convertednto carbon uptake using a total organic carbon content of 25 pgarbon per cell, measured in an aliquot of algal culture (Shi-adzu TOC-5000A, Kyoto, Japan).

.5.3. Respiration rateAt release of broods 1, 3 and 5, respiration was measured

sing a Unisense microrespiration system (Unisense S/A, Den-ark). Six females per condition (from a 500-mL bottle) were

laced individually into respiration chambers containing 1 mLf test medium maintained at 20 ◦C. Decrease in oxygen par-ial pressure associated with respiration was recorded for 1 hsing a miniaturized Clarke-type oxygen sensor connected tohigh sensitivity picoammeter. Sensor signal was calibrated

sing vigorously bubbled M4-pH7 (100% of O2-saturation) andsolution of sodium ascorbate (0.1 M) in NaOH (0.1 M) (0% of2-saturation). Percentage of saturation was converted into oxy-en concentration using an equilibrium concentration in water of82.3 � mol O2 L−1 (20 ◦C, 1 atm). Oxygen consumption rates(� mol O2 daphnid−1 min−1) were calculated as

= [O2]0(1 − e−k�t)V

�t

here [O2]0 is the oxygen concentration (�mol mL−1) mea-ured at t = 0, V the volume (mL) of M4-pH7 in the respirationhamber, �t = 1 min and k is the consumption coefficientmin−1) obtained by fitting exponential models to observedxygen concentrations:

O2]t = [O2]0 e−kt

Mass specific respiration rates calculated in order to eliminatehe influence of differences in individual dry mass at differentges and between exposure conditions.

.5.4. Dry massAt release of broods 1, 3 and 5, body, moult, neonate and egg

ry masses were measured using 10 daphnids per condition (6reviously used for respirometry and 4 remaining from 500-mLottles). All samples were rinsed with UHQ-water. Eggs werearefully dissected out from the brood pouch under a binocularicroscope and counted. Daphnids, moults, neonates and eggsere transferred separately into pre-weighed aluminium pans,ried for 24 h at 60 ◦C and weighed on a microbalance afterooling (precision of 0.1 �g, ultra-microbalance SE2, Sartorius,oettingen, Germany).

.5.5. Scope for growthScope for growth (SFG) was calculated according to the

alance equation: SFG = EA − ER, where EA is the energy

Toxic

ap(hcberelS(

2

6uo1mltoeiiun(owaine

2

r2pvpmWcw((

tEt

y

wb(twaw

3

3

o1MmMawerfat(rsifcwrfMa(1984) and Barata et al. (1998). Moreover, calculated free uranylconcentration was correlated with pH (−log[UO2

2+] = 2.55pH − 8.84, R2 = 0.89; Fig. 1) when all values (Poston et al., 1984;Barata et al., 1998; this study) were taken into account.

F.A. Zeman et al. / Aquatic

ssimilated from food and ER is the energy metabolised (res-iration). Energy lost in excretion is often less than 10%Baillieul et al., 1996; Smolders et al., 2002) and was neglectedere. The conversion of ingestion to assimilated energy wasalculated assuming an assimilation efficiency of 80% for her-ivorous zooplankton (Mayzaud and Razouls, 2002) and annergy equivalent of 36.8 mJ per �g of assimilated C. Respi-ation was converted to energy consumption using an oxyjoulequivalent of 21 kJ L−1 O2 (Elliot and Davison, 1975). Calcu-ated SFG was reported to somatic dry mass (=mass specificFG, J mg−1 DW h−1) to take changes in mass into accountBaillieul et al., 2005).

.5.6. Bioaccumulation and calculation of dose ratesDry samples of daphnids were mineralised in 1 mL of HNO3

9% and 1 mL of H2O2 30% and heated on a sand bath (105 ◦C)ntil evaporation. Mineralised samples were taken up in 15 mLf HNO3 2% (v/v). Uranium was analysed by ICP-MS (Varian000) with a detection limit of ∼10 ng L−1 238U and bioaccu-ulation was expressed in ng 238U mg−1 DW. Daphnid body

ength (L, mm) were calculated using the linear allometric rela-ionship L = 0.005M + 2.44 (R2 = 0.81) obtained from binocularbservation. Biovolume was calculated assuming a constantllipsoid shape (axial ratio of 1:0.6:0.4). U concentrationsn daphnids (�g mL−1 U) was considered to be homogenousn organism and converted into volumic activities (Bq mL−1)sing a specific activity of 2.54 × 104 Bq g−1 for natural ura-ium (Delacroix et al., 2004). Dose rates delivered to daphnidsmGy h−1) were calculated using dose conversion coefficientf 2.2 × 10−3 mGy h−1/Bq mL−1 calculated by EDEN-v2 soft-are (Beaugelin-Seiller et al., 2006). Alpha particles from

ccumulated uranium contributed >99.9% to energy depositedn daphnids (beta- and gamma-emissions from accumulated ura-ium and external radiations from uranium in the water werestimated at <0.1% of total dose rate).

.6. Statistical analyses

Statistical analyses were made using the R language and envi-onment for statistical computing (R Development Core Team,006). Normality assumption was verified through normal qq-lots of residuals and Shapiro–Wilk tests. Homogeneity waserified graphically by plotting standardized residuals againstlots. When homogeneity assumption was rejected, the opti-al Box–Cox transformation (Box and Cox, 1964) was applied.hen one-way ANOVA was significant (p < 0.05), multiple

omparisons were made using Tukey’s post hoc test. Resultsere presented as mean ± standard error of the mean calculated

S.E.M.). Alpha levels were ≤0.05 (*), ≤0.01 (**) and ≤0.001***).

The add-on package drc (Ritz and Streibig, 2005) was usedo fit concentration–effect regressions for the estimation of theC and EC (50% and 10% effect concentrations, respec-

50 10

ively). Logistic models were tested:

= c + d − c

1 + exp[b(log(x) − log(e))]

F4(

ology 86 (2008) 370–378 373

here c is the fixed lower limit, d the maximal effect, e theackground-corrected EC50 according to the extra risk conceptOECD, 2003), b the slope, x uranium concentration and y ishe effect. Residual normality and homogeneity assumptionsere verified as described previously; when the homogeneity

ssumption was rejected, the optimal Box–Cox transformationas applied. Results were presented in mean ± 2σ.

. Results

.1. Acute toxicity and uranium speciation

Acute toxicity test yielded a 48 h EC50 mortality valuef 390 ± 40 �g L−1 U and a EC10 mortality value of70 ± 40 �g L−1 U in M4-pH7. Acute toxicity of uranium in4-pH8 was much lower than in M4-pH7 with a 48 h EC50ortality value of 7.8 ± 3.2 mg L−1 U. During acute test in4-pH8 a yellow precipitate was observed at concentrations

bove 10 mg L−1 U. These observations were in agreementith J-Chess calculation: saturation limit for uranium was

stimated at 0.8 and 1.8 mg L−1, in M4-pH7 and M4-pH8,espectively. Acute toxicity was compared between the dif-erent studies, on the basis of predicted uranium speciationt the 48 h LC50 (Table 1). Calculated free uranyl concentra-ions in the three different water conditions of Poston et al.1984) were 1.3 × 10−3, 4.5 × 10−4, 6.5 × 10−6 �g L−1 UO2

2+,espectively, at 70, 133 and 197 mg L−1 CaCO3. Dominantpecies were Ca2UO2(CO3)3(aq) and (UO2)2CO3(OH)3

−, withncreasing Ca2UO2(CO3)3(aq) at highest hardness. Calculatedree uranyl concentrations in moderate-hard and hard wateronditions reported by Barata et al. (1998) were consistentith those we calculated for this study using water conditions

eported by the authors (0.007 and 0.0015 �g L−1, respectively,or moderate-hard and hard water). Free uranyl concentration in

4-pH7 at 48 h LC50 level (∼1.7 × 10−1 �g L−1 UO22+) was

t least 25-fold higher than values calculated for Poston et al.

ig. 1. Calculated free uranyl concentration (−log �g L−1 UO22+) at uranium

8 h LC50 in relation to pH in water quality conditions, as reported in this studyM4-pH8 and M4-pH7) and in Barata et al. (1998) and Poston et al. (1984).

374 F.A. Zeman et al. / Aquatic Toxicology 86 (2008) 370–378

Table 1Uranium concentration (�g L−1 U), water quality and estimation of free uranyl concentration (�g L−1 UO2

2+ calculated using geochemical speciation simulations(J-Chess software (Van der Lee, 1998) on the basis of water quality conditions) at uranium 48 h LC50 for conditions reported by Barata et al. (1998), Poston et al.(1984) and conditions of this study (M4-pH8 and M4-pH7)

Synthetic water (this study) Synthetic water (Barata et al., 1998) Columbia river water (Poston et al., 1984)

M4-pH7 M4-pH8 Moderate-hard Hard water Soft Moderate-hard Hard

Total U concentration(mg L−1 U)

0.39 7.8 8.3 22.4 6.4 37.5 51.9

pH 7.0 8.0 7.7 8.1 7.9 8.3 8.6Alkalinity

(mg L−1 CaCO3)2.7 34 62 126 57 93 129

3

3

ho

mctda

ud

w

wgagAs

w

Fng

ww

12tidyDsma

bis

3

t

Hardness(mg L−1 CaCO3)

250 250 90

Free UO22+ (�g L−1 U) 1.7 × 10−1 1.7 × 10−4 7.0 × 10−3

.2. Chronic toxicity

.2.1. Survival, growth and moultingIn the chronic test, adult mortality occurred only at the

ighest concentration (100 �g L−1 U) where 10%-mortality wasbserved after 21 days (2 daphnids died/20 daphnids).

Daphnids exposed to uranium showed a reduced body dryass (Fig. 2). Somatic dry mass (without brood) was signifi-

antly lower after 7-day at uranium concentrations ≥50 �g L−1

han in the control. After 20 days of exposure, a significantecrease in daphnid body mass was observed at concentrationss low as 25 �g L−1 U.

Changes in dry mass w(i) with age i (in days) were describedsing an exponential model for the juvenile stage (age i ≤ 7ays):

(i) = w(1) expg(i−1)

ith w(1) = 9.4 �g daphnid−1, w(7) = 62.5 �g daphnid−1 and= 0.32 day−1 (in control and at concentrations of 10nd 25 �g L−1 U) and with w(7) = 51.5 �g daphnid−1 and= 0.28 day−1 (at concentrations of 50, 75 and 100 �g L−1 U).

s soon as daphnids started producing eggs (age i ≥ 7 days, adult

tage), dry mass w(i) followed a Von Bertalanffy model:

(i) = w(max) − [w(max) − w(7)] × expk(i−7)

ig. 2. Individual somatic dry mass (�g), in relation to age (days) and ura-ium concentration. Continuous line: adult growth in control. Dashed line: adultrowth at 100 �g L−1 U. ANOVA: ***p < 0.001. Errors bars = S.E.M.

os9

etpotoenft2

c1nwf

179 70 133 197

1.5 × 10−3 1.3 × 10−3 4.5 × 10−4 6.5 × 10−6

ith k = − 0.5 day−1 independent of exposure condition and(max) differing between exposure conditions: w(max) =80 �g daphnid−1 (in control and at concentrations of 10 and5 �g L−1 U, n = 90); w(max) = 116 �g daphnid−1 (at concen-rations of 50, 75 and 100 �g L−1 U, n = 90). Body massncreased strongly during the juvenile stage in exposure con-itions 0, 10 and 25 �g L−1 U (38% day−1) and then stabilizedielding an average growth rate ∼0% between 14 and 21 days.aphnids exposed to 50, 75 and 100 �g L−1 U were slightly

maller than individuals from the other conditions with an esti-ated mass specific growth rate of 33% during juvenile stage

nd ∼0% between 14 and 21 days.Exposure to uranium did not affect moulting frequency

etween exposed daphnids and the control. Mass of moultsncreased with the age of daphnids, following a linear relation-hip with body dry mass (y = 0.18x + 8.06, n = 12, r2 = 0.70).

.2.2. ReproductionReproduction was affected by uranium exposure for concen-

rations as low as 25 �g L−1. Total reproduction 21-d (numberf neonates produced per female over 21 days) was the mostensitive parameter: EC50 (Fig. 3A) and EC10 were estimated at1 ± 15 and at 14 ± 7 �g L−1 U (n = 10, α = 0.05) respectively.

Reproductive effects on broods 1, 3 and 5 were investigatedxamining two additional reproductive traits, namely the frac-ion of reproducing daphnids and fecundity (number of offspringer daphnid). The fraction of reproducing daphnids was affectednly for the first brood: 60% and 20% for total daphnids, respec-ively, at 75 �g L−1 U and 100 �g L−1 U. At low concentrationsf uranium (from 10 to 50 �g L−1 U) 100% daphnids producedggs (n = 10 per concentration tested) and 90% of the daph-ids produced eggs in the control. Compared to the control,ecundity was significantly reduced at the three highest concen-rations (from 50 to 100 �g L−1 U) for broods 1 and 3 and from5 �g L−1 U for brood 5 (Fig. 3B).

Egg dry mass increased significantly with female age in theontrol (Fig. 3C), from 4.3 ± 0.2 �g egg−1 on day 7 (brood

, n = 10, p < 0.05) to 8.5 ± 0.4 �g egg−1 on day 21 (brood 5,= 10, p < 0.05). Neonate dry mass also increased significantlyith female age in the control, from 5.5 ± 0.6 (n = 10, p < 0.05)

or brood 1 to 9.6 ± 0.8 (n = 10, p < 0.05) for brood 3, then

F.A. Zeman et al. / Aquatic Toxic

Fig. 3. Reproduction. (A) Total number of neonates produced per female over21 days in relation to uranium concentration. Errors bars = S.E.M. (B) Broods(c

sto(apn

3

bddcetenrhto

Musac

3

casiaosurements at 21 day (5th brood) were 0.02 ± 0.01, 2.1 ± 0.2,4.8 ± 0.3, 7.9 ± 0.5, 13.0 ± 0.3 and 8.8 ± 0.9 �Gy h−1 (n = 5;p < 0.05) at 0, 10, 25, 50, 75 and 100 �g L−1 U respectively.

ize (eggs per brood) in relation to brood number and uranium concentration.C) Individual dry mass of egg in relation to brood number and uranium con-entration. ANOVA: *p < 0.05, **p < 0.01, ***p < 0.001. Errors bars = S.E.M.

tabilized at 9.6 ± 0.4 (n = 10, p < 0.05) for brood 5. Exposureo uranium affected egg dry mass, with a significant decreasebserved for brood 3 above a concentration of 25 �g L−1 U7.6 ± 0.6 �g per egg, n = 10, p < 0.05) and for brood 5 above

concentration of 50 �g L−1 U (6.8 ± 0.3 �g per egg, n = 10,< 0.05). In contrast with egg dry mass, dry mass of <24-h oldeonates was not significantly affected by uranium exposure.

.2.3. Ingestion, respiration and scope for growthMass specific filtration rate showed no clear difference

etween the control and exposed daphnids after 7- and 14-ay exposure (broods 1 and 3). Mass specific ingestion ratesecreased significantly after 21 days for daphnids exposed atoncentrations ≥75 �g L−1 U (Fig. 4A). However, ingestionxpressed per daphnid was reduced significantly at concentra-ions of 25, 75 and 100 �g L−1 U on day 7 (first brood) and atvery concentration above 25 �g L−1 U on day 21 (brood 5, dataot shown). No effects were observed for brood 3. Mass specific

espiration rates (mass specific oxygen consumption rates) wereigher on day 7 (brood 1) above a concentration of 25 �g L−1 Uhan in the control (Fig. 4B). On day 14 (brood 3), mass specificxygen consumption rates were affected only at 100 �g L−1 U.

Fcsgb

ology 86 (2008) 370–378 375

ass specific oxygen consumption rates were not affected byranium exposure on day 21 (brood 5). As a result, mass specificcope for growth (Fig. 4C) decreased significantly for first broodt concentrations of 25 and 100 �g L−1 U and for fifth brood atoncentrations of 25, 75 and 100 �g L−1 U.

.3. Bioaccumulation and dose rates

Uranium concentration in daphnids increased with exposureoncentration in medium and exposure time (Fig. 5) exceptt the highest concentration. Bioaccumulation did not differignificantly between 75 and 100 �g L−1 at brood 1. A signif-cant decrease was even observed between 75 and 100 �g L−1,t broods 3 and 5, due to the highly altered physiology ofrganisms. Dose rates calculated from bioaccumulation mea-

ig. 4. Scope for growth in relation to brood number and uraniumoncentration. (A) Mass-specific ingestion rate (�g C mg−1 DW d−1); (B) Mass-pecific respiration rate (�g O2 mg−1 DW h−1); (C) Mass-specific scope forrowth (J mg−1 DW h−1). ANOVA: *p < 0.05, **p < 0.01, ***p < 0.001. Errorsars = S.E.M.

376 F.A. Zeman et al. / Aquatic Toxic

Fig. 5. Uranium bioaccumulation in somatic dry mass of daphnids (ng U mg−1

DW) in relation to brood number and uranium concentration. ANOVA: “a”isp

4

4

lcobticcUa(Miaaoepasle1tpcoemc

•

•

•

4

emctcatcr1i(2mCdarosiatroapert

4a

mei(rofs

ndicates statistically significant difference compared to control; “b” indicatestatistically significant difference compared to preceding concentration, with< 0.05. Errors bars = S.E.M.

. Discussion

.1. Influence of chemistry on acute toxicity of uranium

The bioavailability and toxicity of dissolved trace metals,ike uranium, was recognized to depend on their chemical spe-iation in solution. In this work, we modified the compositionf exposure medium in order to increase the concentration ofioavailable species of uranium while remaining within theolerance range of daphnid physiology. According to the freeon activity model (Campbell, 1995), uranium uptake and itsonsequent toxicity varies as a function of free ion UO2

2+

oncentration, rather than total uranium concentration. Thus,O2

2+ complexation with inorganic ligands (e.g. carbonatesnd hydroxyl) leads to a decrease in uranium bioavailabilityFortin et al., 2004). Moreover, other cations (e.g. Ca2+ and

g2+) can also contribute to reduce metal uptake and toxic-ty through competition for surface binding sites (Di Toro etl., 2005; Riethmuller et al., 2001). Indeed, water hardness andlkalinity were shown to reduce uranium toxicity for aquaticrganisms (Barata et al., 1998; Parkhurst et al., 1984; Postont al., 1984; Sheppard et al., 2005). In this study, reducingH and alkalinity increased UO2

2+ proportion and resulted indecrease in 48 h LC50 expressed as total uranium. At con-

tant competition with cations, 50%-lethality at 48 h should beogically associated with a constant UO2

2+concentration. How-ver, UO2

2+ concentration calculated at 48 h LC50 varied from.7 × 10−1 to 1.7 × 10−4 �g L−1 between pH 7 and 8, and downo 6.5 × 10−6 �g L−1 at pH 8.6 in water conditions reported inrevious studies (Table 1). Competition with Ca2+ and Mg2+

ould not solely explain this variation, as water hardness variednly from 70 to 250 mg L−1 CaCO3. This suggested that influ-nce of chemical speciation and pH on uranium toxicity to D.agna was more complex than anticipated. Several hypotheses

ould be provided, regarding the role of pH (Fig. 1):

A competition of UO 2+ with H+ as suggested by the lin-

2ear relationship between LC50 (expressed as −log[UO2

2+]concentration) and pH (Fig. 1), according to Di Toro et al.(2005).

rt2

ology 86 (2008) 370–378

A non-competitive inhibition of metal transport by protons,as recently reported by Fortin et al. (2007) in algae (with agreater uptake of UO2

2+ at pH 8 than at pH 7).The contribution of other uranium species to toxicity, includ-ing UO2OH+ or carbonated complexes (Fortin et al., 2007;Markich et al., 2000).

.2. Effects of uranium on daphnid energy budget

Scope for growth (SFG), defined as the difference betweennergy assimilated from food and energy metabolised for bodyaintenance (respiration and excretion), gives a valuable indi-

ation of organism health status (Baillieul et al., 2005). Inhis study, waterborne uranium was shown to induce signifi-ant reduction in mass-specific SFG at concentrations as lows 25 �g L−1 U (Fig. 4C). At the highest uranium concentra-ion, SFG represented only 10% of the value measured in theontrol. This strong decrease was mainly associated with theeduction in feeding activity, contributing 64% at brood 1 to00% at brood 5 to the reduction in available energy. Increasen energy demand associated with coping with uranium stressmeasured through respiration) accounted for a maximum of8% at brood 1. Those observations were in agreement in theetabolic cost theory (Calow and Sibly, 1990; Calow, 1991; Deoen and Janssen, 2003) which assumes that under stress con-itions, organisms set up defense and repair mechanisms thatre energy-consuming. As a consequence, any stress-inducededuction in available energy for organisms comes at the expensef growth and reproduction. Here, growth and reproductionhowed the same sensitivity to uranium exposure, both decreas-ng at the concentration of 50 �g L−1, as early as brood 1,nd of 25 �g L−1 U at brood 5. Maximum decline in alloca-ion of energy reached 48 and 63–68% for somatic growth andeproduction, respectively. For reproduction, both the numberf eggs produced over 21 days and dry mass per egg at broods 3nd 5 were affected by increasing uranium concentration, withotentially major consequences for population dynamics. Forxample, Ebert (1991) showed that the production of small eggsesulted in retarded somatic growth and delayed reproduction inhe offspring generation.

.3. Effects of uranium compared to radioactive substancesnd trace metals

Chemical toxicity of uranium is commonly considered ofuch greater concern in comparison to its radiotoxicity (Miller

t al., 2002; Sheppard et al., 2005). In this study, 50%-reductionn reproduction was observed at a sixfold lower concentration91 �g L−1) than previously reported by Poston et al. (1984), as aesult of increased bioavailability at pH 7. Parallel quantificationf uranium dose rates (from 2.1 to 13 �Gy h−1) provided a basisor comparing effects with those induced by other radioactiveubstances and trace metals.

Effects induced by Am-241 alpha irradiation over a higherange of dose rates (Alonzo et al., 2006) were far smallerhan those reported for uranium in this study. Thus with Am-41, ingestion and fecundity remained unaffected up to a dose

Toxic

rogduoeaam

irrw31tdg

hefrtmCnwagBTwds

pp2vYutu

A

tOWaERt

R

A

B

B

B

B

B

B

B

B

C

C

C

C

D

D

D

D

E

E

E

E

F.A. Zeman et al. / Aquatic

ate of 7.9 mGy h−1 while dry mass per egg and survivalf starved neonates decreased above 0.16 mGy h−1, somaticrowth reduced after 16 days at 0.9 mGy h−1 and respiratoryemand increased after 23 days at 7.9 mGy h−1. At 75 �g L−1

ranium, daphnids were exposed to the maximum dose ratef 13 �Gy h−1, a value where comparatively with Am-241,xpected effects of radiation stress should be very slight. Asconclusion, the observation of a strongly impaired feeding

ctivity and fecundity at 75 �g L−1 uranium necessarily resultedainly from chemical toxicity.Exposure to external gamma radiation, at dose rates rang-

ng from 0.4 to 31 mGy h−1, was shown to affect D. magnaeproduction (Gilbin et al., in press). The timing of broodelease changed in an opposite way when compared to whatas observed with uranium, with a 1 to 2-day advance at1 mGy h−1. At this dose rate, a reduction in brood size after5 days of exposure caused a 21% decrease in fecundity. In con-rast with results obtained with uranium, no change was eitheretected in egg dry mass, respiration, ingestion rate, or somaticrowth.

Effects of uranium are linked to its chemical toxicity as aeavy metal and similarity with other trace metals might bexamined. We estimated a 21-day EC10 of 59 �M (14 �g L−1 U)or reproduction. This concentration was comparable to valueseported for cadmium (33 �M) and copper (118 �M), suggestinghat uranium might be considered among some of the most toxic

etals. This was also interesting to note that toxicity of Cd andu closely resembled that reported for uranium, in terms of theature of effects produced: SFG under cadmium or copper stressas strongly reduced, mainly as a result of decreasing feeding

ctivity; consequences for organisms included a reduced somaticrowth and strongly impaired fecundity (Bodar et al., 1988;aird et al., 1990; Knops et al., 2001; Baillieul et al., 2005).his contrasted with the situation with Am-241 where ingestionas unaffected and daphnids maintained a constant fecundity,espite a reduced SFG associated with increased energy con-umption (Alonzo et al., 2006).

Toxicity mechanisms of Cd and Cu are known to involve theroduction of reactive oxygen species which disrupts normalhysiological processes (Stohs and Bagchi, 1995; Livingstone,001). Like some transition metals, uranium chemically acti-ates oxygen species via redox reactions (Miller et al., 2002;azzie et al., 2003). Further research is needed to test whetherranium, cadmium and copper share the same mechanism ofoxicity and to improve our understanding of consequences ofranium exposure for freshwater organisms.

cknowledgements

We specially thank Nadine Cauvin for her technical assis-ance with D. magna culturing, Virginie Camillieri and Danielrjollet for ICP-AES and ionic chromatography measurements.e are also grateful to Claire Della-Vedova for her helpful

dvice on the statistical analyses. This study is a part of theNVIRHOM research program supported by the Institute foradioprotection and Nuclear Safety. We are very grateful to

wo anonymous referees for the helpful comments on this work.

F

ology 86 (2008) 370–378 377

eferences

lonzo, F., Gilbin, R., Bourrachot, S., Floriani, M., Morello, M., Garnier-Laplace, J., 2006. Effects of chronic internal alpha irradiation on physiology,growth and reproductive success of Daphnia magna. Aquat. Toxicol. 80,228–236.

aillieul, M., Selens, M., Blust, R., 1996. Scope for growth and fitness ofDaphnia magna in salinity-stressed conditions. Funct. Ecol. 10, 227–233.

aillieul, M., Smolders, R., Blust, R., 2005. The effect of environmental stresson absolute and mass-specific scope for growth in Daphnia magna Strauss.Comp. Biochem. Physiol. C 140, 364–373.

aird, D.J., Barber, I., Calow, P., 1990. Clonal variation in general responses ofDaphnia magna Straus to toxic stress. I. Chronic life-history effects. Funct.Ecol. 4, 399–407.

arata, C., Baird, D.J., Markich, S.J., 1998. Influence of genetic and environ-mental factors on the tolerance of Daphnia magna Straus to essential andnon-essential metals. Aquat. Toxicol. 42, 115–137.

eaugelin-Seiller, K., Jasserand, F., Garnier-Laplace, J., Gariel, J.C., 2006. Mod-eling radiological dose in non-human species: principles, computerization,and application. Health Phys. 90, 485–493.

odar, C.W.M., Van der Sluis, I., Voogt, P.A., Zandee, D.I., 1988. Effects of cad-mium on consumption, assimilation and biochemical parameters of Daphniamagna: possible implication for reproduction. Comp. Biochem. Physiol. C90, 341–346.

onin, B., Blanc, P.L., 2001. L’uranium dans le milieu naturel, des originesjusqu’a la mine. In L’uranium de l’environnement a l’homme. EDP Sciences,Les Ulis, France, pp. 8–41.

ox, G.E.P., Cox, D.R., 1964. An analysis of transformations. J. R. Stat. Soc. B26, 296–311.

ampbell, P.G.C., 1995. Interactions between trace metals and aquatic organ-isms: a critique of the free-ion activity model. In: Tessier, A., Turner, D.R.(Eds.), Metal Speciation and Bioavailability in Aquatic Systems. Wiley, NewYork, pp. 45–102.

alow, P., 1991. Physiological costs of combating chemical toxicants: ecologicalimplications. Comp. Biochem. Physiol. C 100, 3–6.

alow, P., Sibly, R.M., 1990. A physiological basis of population processes:ecotoxicological implications. Funct. Ecol. 4, 283–288.

olle, C., Garnier-Laplace, J., Roussel-Debet, S., Adam, C., Baudin, J.-P.,2001. Comportement de l’uranium dans l’environnement. In L’Uraniumde l’environnement a‘l’homme. EDP Sciences, Les Ulis, France, pp. 187–211.

e Coen, W.M., Janssen, C.R., 2003. The missing biomarker link: relationshipbetween effects on the cellular energy allocation biomarker of toxicant-stressed Daphnia magna and corresponding population characteristics.Environ. Toxicol. Chem. 22, 1632–1641.

elacroix, D., Guerre, J.-P., Leblanc, P., 2004. Guide pratique Radionucleideset Radioprotection. Radioprotection, 4th ed. EDP Science, Les Ulis, France.

enison, F.H., Garnier-Laplace, J., 2005. The effects of database parameteruncertainty on uranium(VI) equilibrium calculations. Geochim. Cos-mochim. Acta 69, 2183–2191.

i Toro, D.M., McGrath, J.A., Hansen, D.J., Berry, W.J., Paquin, P.R., Mathew,R., Wu, K.B., Santore, R.C., 2005. Predicting sediment metal toxicity using asediment biotic ligand model: methodology and initial application. Environ.Toxicol. Chem. 24, 2410–2427.

bert, D., 1991. The effect of size at birth, maturation threshold andgenetic differences on the life-history of Daphnia magna. Oecologia 86,243–250.

lendt, B.P., 1990. Selenium deficiency in Crustacea: an ultrastructural approachto antennal damage in Daphnia magna Straus. Protoplasma, 25–33.

lliot, J.M., Davison, W., 1975. Energy equivalents of oxygen consumption inanimal energetics. Oecologia, 195–201.

nvironment-Canada and Health-Canada, 2000. Releases of radionuclidesfrom nuclear facilities (impact on non-human biota). Priority substances

list—Assement report. Canadian Environmental Protection Act, Environ-ment Canada Health Canada, 1999.

ortin, C., Denison, F.H., Garnier-Laplace, J., 2007. Metal-phytoplankton inter-actions: modeling the effect of competing ions (H+, Ca2+, and Mg2+) onuranium uptake. Environ. Toxicol. Chem. 26, 242–248.

3 Toxic

F

F

G

K

K

K

L

M

M

M

O

O

O

P

P

P

R

R

R

R

S

S

S

S

V

78 F.A. Zeman et al. / Aquatic

ortin, C., Dutel, L., Garnier-Laplace, J., 2004. Uranium complexation anduptake by a green alga in relation to chemical speciation: the importanceof the free uranyl ion. Environ. Toxicol. Chem. 23, 974–981.

rost, B.W., 1972. Effects of size and concentration of food particles on the feed-ing behaviour of a marine planktonic copepod Calanus pacificus. Limnol.Oceanogr., 805–815.

ilbin, R., Alonzo, F., Garnier-Laplace, J., in press. Effects of chronic externalgamma irradiation on growth and reproductive success of Daphnia magna.J. Environ. Radioact.

nops, M., Altenburger, R., Segner, H., 2001. Alterations of physiological ener-getics, growth and reproduction of Daphnia magna under toxicant stress.Aquat. Toxicol. 53, 79–90.

ooijman, S.A.L.M., 2000. Dynamic Energy and Mass Budgets in BiologicalSystems. Cambridge University Press, Cambridge.

uhne, W.W., Caldwell, C.A., Gould, W.R., Fresquez, P.R., Finger, S., 2002.Effects of depleted uranium on the health and survival of Ceriodaphnia dubiaand Hyalella azteca. Environ. Toxicol. Chem. 21, 2198–2203.

ivingstone, D.R., 2001. Contaminant-stimulated reactive oxygen species pro-duction and oxidative damage in aquatic organisms. Mar. Pollut. Bull. 42,656–666.

arkich, S.J., Brown, P.L., Jeffree, R.A., Lim, R.P., 2000. Valve movementresponses of Velesunio angasi (Bivalvia: Hyriidae) to manganese and ura-nium: an exception to the free ion activity model. Aquat. Toxicol. 51,155–175.

ayzaud, P., Razouls, S., 2002. Degradation of gut pigment during feedingby a subantarctic copepod: importance of feeding history and digestiveacclimation. Limnol. Oceanogr., 393–404.

iller, A.C., Stewart, M., Brooks, K., Shi, L., Page, N., 2002. Depleted uranium-catalyzed oxidative DNA damage: absence of significant alpha particledecay. J. Inorg. Biochem. 91, 246–252.

ECD, 1998. Daphnia magna reproduction test, guidelines for testing ofchemicals, Rep. No. 211. Organisation for Economic Cooperation andDevelopment, Paris, France.

ECD, 2003. Draft Guidance Document for on the Statistical Analysis of Eco-

toxicity Data. Organisation for Economic Cooperation and Development,Paris, France.

ECD-NEA, 1996. Survey of Thermodynamic and Kinetic Databases. Organ-isation for Economic Cooperation and Development and DevelopmentNuclear Energy Authority, Paris, France.

W

Y

ology 86 (2008) 370–378

arkhurst, B.R., Elder, R.G., Meyer, J.S., Sanchez, D.A., Pennak, R.A., Waller,W.T., 1984. An environmental hazard evaluation of uranium in a RockyMountain stream. Environ. Toxicol. Chem., 113–124.

ickett, J.B., Specht, W.L., Keyes, J.L., 1993. Acute and chronic toxicity ofuranium compounds to Ceriodaphnia dubia, Rep. No. WSRC-RP-92-995.Prepared for the Department of Energy under contract DE-AC09-89SR18035.

oston, T.M., Hanf Jr., R.W., Simmons, M.A., 1984. Toxicity of uranium toDaphnia magna. Water Air Soil Poll. 22, 289–298.

agnarsdottir, K.V., Charlet, L., 2000. Uranium behavior in naturalenvironments. In Environmental Mineralogy: Microbial Interactions,Anthropogenic Influences, Contaminated Lands and Waste Management,vol. Series 9. Mineralogical Society of Great Britain and Ireland, London,UK, pp. 333–377.

Development Core Team, 2006. R: A language and Environment for StatisticalComputing. R Development Core Team, Vienna, Austria.

iethmuller, N., Markich, S.J., van Dam, R.A., Parry, D., 2001. Effects of waterhardness and alkalinity on the toxicity of uranium to a tropical freshwaterhydra (Hydra viridissima). Biomarkers, 45–51.

itz, C., Streibig, J.C., 2005. Bioassay analysis using R. J. Stat. Soft. 12,1–22.

emaan, M., Holdway, D., van Dam, A.R.A., 2001. Comparative sensitivity ofthree populations of the cladoceran Moinodaphnia macleayi to acute andchronic uranium exposure. Environ. Toxicol. 16, 365–376.

heppard, S.C., Sheppard, M.I., Gallerand, M.O., Sanipelli, B., 2005. Deriva-tion of ecotoxicity thresholds for uranium. J. Environ. Radioact. 79, 55–83.

molders, R., Bervoets, L., Blust, R., 2002. Transplanted zebra mussels (Dreis-sena polymorpha) as active biomonitors in an effluent-dominated river.Environ. Toxicol. Chem., 1889–1896.

tohs, S.J., Bagchi, D., 1995. Oxidative mechanisms in the toxicity of metalions. Free Radical Biol. Med. 18, 321–336.

an der Lee, J., 1998. Thermodynamic and mathematical concepts of Chess.GIG-Ecole des Mines de Paris, Fontainebleau, France.

HO, 2001. Depleted uranium: sources, exposure, and health effects, Rep. No.WHO/SDE/PHE/01.1. Geneva, Switzerland.

azzie, M., Gamble, S.L., Civitello, E.R., Stearns, D.M., 2003. Uranyl acetatecauses DNA single strand breaks in vitro in the presence of ascorbate (vita-min C). Chem. Res. Toxicol. 16, 524–530.