Science of the Total Environment 443 (2013) 673–680

Contents lists available at SciVerse ScienceDirect

Science of the Total Environment

j ourna l homepage: www.e lsev ie r .com/ locate /sc i totenv

Metal bioaccumulation, genotoxicity and gene expression in the European woodmouse (Apodemus sylvaticus) inhabiting an abandoned uranium mining area

Joana Lourenço a,b,⁎, Ruth Pereira c,b, Fernando Gonçalves a,b, Sónia Mendo a,b

a Departamento de Biologia, Universidade de Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugalb CESAM, Centro de Estudos do Ambiente e do Mar, Universidade de Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugalc Departamento de Biologia, Faculdade de Ciências, Universidade do Porto, Rua do Campo Alegre, 4169-007 Porto, Portugal

H I G H L I G H T S

► Long term effects of chronic pollution in natural population of rodents.► Bioaccumulation of cadmium and uranium by organisms exposed to uranium wastes.► P53 upregulation in the liver and SNPs in the Rb gene detected in the kidney.► Significant DNA damages detected by the comet assay.► Concerns on the risks of human populations living nearby uranium mining areas.

⁎ Corresponding author at: Department of Biology &Campus Universitário de Santiago, 3810-193 Aveiro, Pofax: +351 234372587.

E-mail address: joanalourenco@ua.pt (J. Lourenço).

0048-9697/$ – see front matter © 2012 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.scitotenv.2012.10.105

a b s t r a c t

a r t i c l e i n f o

Article history:Received 19 June 2012Received in revised form 30 October 2012Accepted 31 October 2012Available online 7 December 2012

Keywords:DNA damageSNPsGene expressionMetalsRadionuclidesBioaccumulation

Genotoxic effects caused by the exposure to wastes containing metals and radionuclides were investigated inthe European wood mice (Apodemus sylvaticus). The animals were captured in the surroundings of an aban-doned uranium mining site. DNA damage was assessed by comet assay; gene expression and single nucleo-tide polymorphisms (SNPs) were assessed, respectively, by Real-Time PCR and melt curve analysis. Thebioaccumulation of metals in the liver, kidney and bones was also determined to help clarify cause–effectrelationships. Results confirmed the bioaccumulation of cadmium and uranium in organisms exposed touranium mining wastes. P53 gene was found to be significantly up-regulated in the liver of those organismsand SNPs in the Rb gene were also detected in the kidney. Our results showed that uranium mining wastescaused serious DNA damage resulting in genomic instability, disclosed by the significant increase in DNAstrand breaks and P53 gene expression disturbance. These effects can have severe consequences, since theymay contribute for the emergence of serious genetic diseases. The fact that mice are often used asbioindicator species for the evaluation of risks of environmental exposure to humans, raises concerns onthe risks for human populations living near uranium mining areas.

© 2012 Elsevier B.V. All rights reserved.

1. Introduction

The impact of uranium mining and milling on the environmenthas long been recognized (Lozano et al., 2000). The extraction ofuranium has an important radioactive and chemical impact, since itcauses the surface exposure of geologic material rich in uraniumand its radioactive descendants, as well as other metals and metal-loids (André et al., 2009; Carvalho et al., 2007; Lourenço et al.,2012; Lozano et al., 2000; Pereira et al., 2008). Consequently, evenafter the cessation of mining activities, the impacts on the environ-ment persist for several decades due to weathering conditions

CESAM, University of Aveiro,rtugal. Tel.: +351 234370967;

rights reserved.

(Lozano et al., 2000; Pereira et al., 2006). Chronic exposure to suchcontaminants may result in continued bioaccumulation and hasbeen linked to several harmful effects in living organisms, sincethey may have mutagenic, carcinogenic and teratogenic properties,among others (Amaral et al., 2007; Lourenço et al., 2012). Datacollected from vertebrates living in/exposed to contaminated areasare of utmost importance since they can bring pertinent informationto be used in human risk assessment (Pereira et al., 2006). Further-more, field studies provide crucial ecotoxicological data, given that,under laboratory conditions, it is difficult to mimic the exposure tocomplex mixtures of contaminants, synergistic and/or antagonisticeffects, as well as the complex interactions between abiotic and bioticfactors and their role in mixture toxicity (Léon et al., 2007; Pereiraet al., 2006; Sánchez-Chardi et al., 2007). Environmental exposure ofterrestrial vertebrates to non-essential metals in contaminated areasis known to be associated with a wide range of toxic effects at

674 J. Lourenço et al. / Science of the Total Environment 443 (2013) 673–680

different levels of biological organization (Paternain et al., 1989;Raymond-Whish et al., 2007; Sánchez-Chardi et al., 2007, 2009), be-cause these organisms bioaccumulate them in different tissues andorgans (Amaral et al., 2007; Marcheselli et al., 2010; Pereira et al.,2006; Sánchez-Chardi et al., 2007).

The environmental monitoring of genotoxic effects requires the se-lection of representative organisms as sentinels, as well as the develop-ment of suitable and sensitive assays (Festa et al., 2003). Among theseorganisms, small mammals, such as rodents, are preferentially selected,for they are considered to be sensitive to contaminants and appropriatebioindicator species for monitoring exposures and effects (Beernaertet al., 2007; Festa et al., 2003; Marcheselli et al., 2010). Small mammalsare used as sentinel species to monitor exposure risks because theyallow a better assessment of the effects on consumers, and thereforeconsent the prediction of possible risks to human health (Marcheselliet al., 2010). Although sentinel species data are not used as the sole fac-tor in assessing human health risks, that information can be useful for aweight-of-evidence approach in risk assessment decisions, by providingearly warning of situations requiring further study, or by suggesting po-tential causes and effects (Van der Schalie et al., 1999). Studies on smallmammals have demonstrated their ability to accumulate a wide spec-trum of pollutants widespread in their distribution areas (Ieradi et al.,1998; Sánchez-Chardi et al., 2007). Additionally, small mammals usual-ly fulfill the criteria to be selected as bioindicator species, namely: highabundance, relatively small feeding and breeding area and wide life ex-pectancy, enough to estimate possible long-term effects (Marcheselliet al., 2010). The ubiquitous wood mouse Apodemus sylvaticus isone of the species that meets these criteria (Marcheselli et al., 2010;Sánchez-Chardi et al., 2007). Although occasionally showing carnivo-rous behavior, this mouse is mainly a primary consumer and, thereby,is often used as a bioindicator of pollution (Marcheselli et al., 2010;Rogival et al., 2006; Sánchez-Chardi et al., 2007, 2009). This species isexposed to contaminants mainly via inhalation of soil particles and in-gestion of contaminated food and water (Marcheselli et al., 2010).Significant correlations between metal contamination (Festa et al.,2003), radioactivity and genetic damage in free-living rodents havebeen detected (Ieradi et al., 1998), which corroborates their sensitivityto these stress agents.

Comet assay is one of the many methodologies used to detect ge-netic damage. It is a simple, rapid and sensitive tool to assess DNAdamage and repair in individual cells. It has increasingly found appli-cation in several areas ranging from genetic toxicology to humanepidemiology (Dhawan et al., 2009). The determination of DNA dam-age in sentinel species can provide information about the genotoxicpotential of their habitat at an early stage, allowing the implementa-tion of strategies for the prevention and reduction of deleterioushealth effects in wildlife species and in humans (Dhawan et al.,2009). The exposure to potentially toxic chemical compounds also in-volves changes in gene expression profiles in receptor organisms(Brulle et al., 2010). Therefore, the analysis of changes in gene expres-sion is a powerful tool to diagnose the existence of a stress conditionand to analyze the response of organisms (Bernard et al., 2010; Brulleet al., 2008). Some molecular-genetic tools like quantitative reversetranscription PCR (RT-qPCR), widely developed in the last decade,allow the analysis of transcriptomes of stressed organisms, exposed inlaboratory or under field conditions (Brulle et al., 2010; Lee et al.,2005; Pirooznia et al., 2007). Gene expression analysis by qRT-PCRallows the accurate and sensitive measurement of gene expressionlevels (Willems et al., 2008). The environmental exposure to genotoxiccontaminants may also cause mutations (Pfeifer, 2010). Therefore, theevaluation of mutation occurrence is useful to assess the genomic insta-bility and also the predisposition for developing genetic diseases as aconsequence of exposure to genotoxic contaminants (Pfeifer, 2010).

This study aimed at evaluating the genotoxic effects occurring in theEuropean wood mouse inhabiting the deposition area of uraniummining wastes, an area rich in metals and radionuclides. Moreover,

changes in expression levels and the presence of SNPs in the tumor sup-pressor genes P53 and Rb in target organs were also investigated, sinceit is acknowledged that metals and radionuclides can have mutagenicand carcinogenic properties (Amaral et al., 2007; UNSCEAR, 2000). Rband P53 genes are crucial in the cellular response to a range of environ-mental and intracellular stresses (Meek, 2004). They act mainly in thecontrol of the G1/S checkpoint of the cell cycle, eliciting cell cycle arrestto allow DNA repair and/or apoptosis in response to a variety ofgenotoxic stresses (Genovese et al., 2006; Meek, 2004). In addition,data on the bioaccumulation ofmetals in target organswere also impor-tant to clarify cause–effect relationships.

2. Material and methods

2.1. Study site

The Cunha Baixa uranium mine (coordinates: 40.571393 N7.753322 W) was one of the largest uranium mines in Portugal(Carvalho et al., 2009). The mine is located southeast of Viseu, nearCunha Baixa village. The underground and open pit mining works,carried out between 1970 and 1991, produced about 1000 t of U3O8

and 1 million t of waste materials, that were disposed in a dump sur-rounding the mine area (Neves et al., 2005). For in situ leaching ofuranium in underground works, and in heap leaching of low gradeores at the surface sulfuric acid was used (Antunes et al., 2008a;Carvalho et al., 2009; Pereira et al., 2008). As a consequence, themine pit water is acidic (pHb3.5), and is able to move through per-meable heap leaching wastes. It follows the changes in the piezomet-ric level of the underlying aquifer, mobilizes elements and increasestheir dispersion (Neves et al., 2005). Since the end of the mining ac-tivity that the acidic water from the mine pit has been almost contin-uously pumped out to treatment plants. It has been treated withburned lime and barium chloride for pH neutralization and radionu-clide precipitation, respectively (Carvalho et al., 2009; Pereira et al.,2008). When the maximum capacity of the pond is reached, the treat-ed affluent is discharged in a stream passing nearby, and the sludgeformed by the accumulation of the chemical compounds at thebottom is removed and spread near the mine pit. The sludge hashigh levels of metals (Pereira et al., 2008) and radionuclides fromuranium series (Carvalho et al., 2007). Soils from the adjacent areasof the mine pit have suffered several impacts through the depositionof both tailings and sludge from the wastewater treatment plant(Antunes et al., 2008a; Pereira et al., 2008).

2.2. Test organisms and sampling procedure

Europeanwoodmice (A. sylvaticus)were collected at a pine-oak treeforest, 3 km from themining area (reference site), and in the surround-ings of the mine pit (contaminated site), which is also a pine-oak treeforest that includes tailings and sludge deposition sites (Fig. 1). Thereported average home range of A. sylvaticus is about 2250 m2

(Marcheselli et al., 2010; Mertens et al., 2001), which is smaller thanthe surface areas of the chosen study-sites. Thus, and also due to thefragmentation of the landscape, it was assumed that the wood micecaptured fed mainly inside the study-sites.

The sampling sites were selected based on the presence of minewastes and on data obtained from chemical and ecotoxicological stud-ies, previously performed by our team in the area (André et al., 2009;Antunes et al., 2008b; Lourenço et al., 2012; Pereira et al., 2008, 2009).They provided the information about the degree of contamination andtoxicological effects. The organisms were captured in 3-day capture pe-riods, using live traps baited with a mixture of canned tuna fish andbaby cereals. In total, 21 individuals were trapped, 11 from the contam-inated site and 10 from the reference site (Table 1).

The captured animals were taken alive to the laboratory and(according to the Directive 2010/63/EU of the European Parliament

Fig. 1. Map of the study area, showing the location of the Cunha Baixa village and the sampling sites. The approximate location of the mine pit and waste deposition areas are alsoindicated. The sampling points are marked with a black dot.

675J. Lourenço et al. / Science of the Total Environment 443 (2013) 673–680

and of the Council of 22 September 2010 on the protection of animalsused for scientific purposes, Annex IV), the animals were sacrificedusing the method of cervical dislocation. According to this directive,this method is appropriate for rodents under 150 g, which was thecase of the organisms used in this study, that had an average weightof 30 g. Thismethod is considered a humanemethod of killing small ro-dents, when performedby a trained person, for it causes extensive dam-age to the brainstem resulting in immediate unconsciousness anddeath. The person responsible for euthanizing the animals used in thisstudy is certified by national authorities (FI/285/2011) for animalexperimentation. After being weighed, the sex of each animal was de-termined and standard mammal body measurements were performed

Table 1Characteristics and standard mammal body measurements of the studied populations.

Reference site Contaminated site

Sample size 11 10Total weight (g) (Mean±SD) 26.3±5.6 24.3±5.2Total length (cm) (Mean±SD) 9±0.6 8.9±0.8Length of the rear paw (cm) (Mean±SD) 2.2±0.2 2.2±0.3Length of the ears (cm) (Mean±SD) 1.6±0.2 1.5±0.3Nº of females 4 3Nº of males 7 7

(Table 1). Liver, kidneys, and bones were dissected using a scalpel andweighted to the nearest 0.1 mg. Blood was extracted using a syringecontaining sodium EDTA and preserved at 4 °C with 10% of DMSOuntil analysis (about 4 h). A fraction of the liver and kidneys werefixed in RNA later (Qiagen, Germany) and stored at −80 °C until RNAextraction. The remaining fraction of these organs and the bones werefrozen at−20 °C until wet digestion andmetal content analysis. Organswere selected based on their role in the detoxification and storage ofmetals.

2.3. Comet assay

For the comet assay, 10 μL of whole blood samples was used directlyfrom the collection tube (11 and 10 samples from the reference andcontaminated sites, respectively). The alkaline version of the assaywas conducted under yellow light to prevent UV induced DNA damage,and performed with slight modifications of the protocol described byNogueira et al. (2006). Briefly, microscope slides were first coveredwith an agarose layer (1% normal melting point agarose), left to dryand covered by a second layer (0.5% low melting point agarose)containing the cells. Visual scoring of cellular DNA on each slide wasbased on the categorization of 100 cells randomly selected. Thecomet-like formations were visually graded into 5 classes, based onDNA damage, and scored as described by Garcia et al. (2004). Positive

676 J. Lourenço et al. / Science of the Total Environment 443 (2013) 673–680

controlswere always included and consisted of cells previously exposedto 200 μM of H2O2, for 1 h. H2O2 was used as a positive control for oxi-dative damage detection since most of the metals here analyzed areknown to cause oxidative stress. Although uranium can induce deleteri-ous effects via both chemical and radiological pathways, its chemicaltoxicity is the main environmental concern (Brugge and Buchner,2011) given that uranium has a very low level of radioactivity (Bruggeand Buchner, 2011). The chemical toxicity of uranium is similar to nick-el and chromium (Brugge and Buchner, 2011), which aremetals knownto cause significant DNA damages due to the induction of free radicalproduction leading to oxidative stress (Harris and Shi, 2003).

2.4. RNA extraction and RT-qPCR

RT-qPCR analyses were performed in kidney and liver samplesfrom 5 organisms (3 males and 2 females) randomly chosen fromthe total of animals captured in each study-site. Total RNA wasextracted using PureLink™ RNA Mini Kit (Invitrogen, USA), treatedon column with PureLink™ DNase (Invitrogen, USA) and quantifiedusing Qubit™ Fluorometer (Invitrogen, USA) and Quant-iT™ RNAAssay Kit (Invitrogen, USA). The extracted RNAs were also checkedfor quality by visually assessing the ratio of 28S:18S RNA on agarosegel stained with EtBr. First-strand cDNAwas synthesized using Super-Script® First-Strand Synthesis System for RT-PCR (Invitrogen, USA).First-strand cDNA was synthesized from 2 μg of total RNA in the pres-ence of random hexamer primers. The genes peptidylprolyl isomeraseA, beta-actin and 18S were used as reference genes for the kidneysamples. As for the liver samples, peptidylprolyl isomerase A (Ppia),beta-actin and glyceraldehyde 3-phosphate dehydrogenase (Gapdh)were used as reference genes. The cyclin G-associated kinase (Gak)and the signal recognition particle 72 (Srp72) genes were also testedto serve as reference genes, but were not suitable for this experiment.The stability of the reference genes used in this study was determinedusing the qbasePlus/qPCR data analysis software in which geNormsoftware is incorporated. qPCR reactions were performed in a finalvolume of 20 μL containing 10 μL of SsoFast™ EvaGreen® Supermix(Biorad, USA), 1 μL of diluted (1:10) cDNA and 200 nM primers. Anon-template control (NTC) was included in the qPCR analysis to de-termine the specificity of target cDNA amplification and also a non-RTcontrol (NRT), including RNA instead of cDNA to determine if theRNAs were contaminated with genomic DNA. The RT-qPCR wasperformed using the CFX96™ Real-Time System (Biorad, USA) andthe following cycling parameters: 98 °C for 30 s and 44 cycles of 5 sat 95 °C and 10 s at 57 °C. Melting curves were performed to identifythe presence of primer dimers and to analyze the specificity of the re-action. The amplification efficiency of each primer pair was calculatedusing a dilution series of cDNA. The primers used in RT-qPCR weredesigned using Mus musculus sequences deposited in the Genbankdatabase and are listed in Table 2. Two sets of primers for the

Table 2Sequences of the primers used in RT-qPCR.

Gene Sequence (5′–3′)

18S For: ATGGCCGTTCTTAGTTGGTGRev: CGCTGAGCCAGTCAGTGTAG

Ppia For: AGCATACAGGTCCTGGCATCRev: CACCTTCCCAAAGACCACAT

B-actin For: ACTGGGACGACATGGAGAAGRev: GGGGTGTTGAAGGTCTCAAA

Gapdh For: AACTTTGGCATTGTGGAAGGRev: ACACATTGGGGGTAGGAACA

P53 For: AACCGCCGACCTATCCTTACRev: CTTCTGTACGGCGGTCTCTC

Rb1 For: AGAGAGAACGCCACGAAAAARev: GATGGCTGATCACTTGCAGA

Rb2 For: TGCATGGCTTTCAGATTCACRev: ACAGGGCAAGGGAGGTAGAT

Rb gene are listed in Table 2, because one of them (Rb1) was not suit-able for the gene expression analysis in the liver samples. The primerset Rb1 was only used in liver samples for the detection of mutationsin the region between exons 11 and 12 of the Rb gene. The relative ex-pression levels of the target genes were calculated according to theefficiency correction method described by Pfaffl (2001).

The amplified fragments of the target genes P53 and Rb were alsoscreened for the presence of SNPs. Fragments were amplified as re-ferred earlier in a master mix containing the EvaGreen fluorescentdye (EvaGreen® Supermix (Biorad, USA)), and mutations weredetected after analysis of the EvaGreen melt curve behavior of eachamplification product (Temesvári et al., 2011) and further confirmedby DNA sequencing.

2.5. Chemical analysis

Metals were selected according to previous soil analysesperformed in both sampling sites (Pereira et al., 2008; Antuneset al., 2008b) to determine the bioavailability of these contaminants.The total concentration of Be, Al, Mn, Ni, Cu, Zn, Sr, Cd, Pb and Uwas determined in the liver, kidneys and bones of the organisms cap-tured in the mine and in the reference area (a total of 21 organisms),by inductively coupled plasma mass spectrometry (ICP–MS) (APHA,1995). The tissues were oven-dried at 105 °C until weight stabiliza-tion, and then the weights were recorded to the nearest 0.1 mg.Then, they were digested in Teflon vials containing 3 mL suprapurnitric acid (65%, Merck) in a 60 °C sand bath until there were nosolid fragments in the solution; 1.5 mL of suprapur hydrogen perox-ide (30%, Merck) was added to the vials and they were placed againin the sand bath until the solution became clear. Finally, ultra-purewater was added to a total volume of 5 mL. Reagent blanks wereobtained following the same procedure but without the tissues.Among all the analyzed metals, Be, Cd and Ni are recognized ashuman carcinogens (Beyersmann, 2002). Despite being weak muta-gens, these metals also have the ability to enhance the mutagenicityand carcinogenicity of directly acting genotoxic agents like, for exam-ple, ultraviolet or ionizing radiation (Beyersmann, 2002).

2.6. Statistical analysis

Regarding chemical analysis and comet assay results, the statisti-cal analysis was performed using one-way ANOVAs and Kruskal–Wallis one-way analysis of variance on ranks, whenever the datafailed to meet the normality and homoscedascity assumptions. Thedata obtained by real-time qPCR were analyzed using qbasePlus soft-ware, which performed Mann–Whitney tests on the results obtained.The level of significance defined for all the analyses was 0.05.

3. Results and discussion

The metal content in the analyzed organs revealed that Cd and Uwere significantly bioaccumulated bymice inhabiting the contaminatedsite (Table 3). Additionally, these animals also showed significantlyhigher levels of Cu in the bones (Table 3). Indeed, a study performedby Pereira et al. (2008), showed that the soils from the mine area (des-ignated in that study by A, B, C) where the mice were captured, hadhigher levels of these metals when compared to the reference site (des-ignated by J in that study). These results are in agreement with thelevels found in these organisms. However, the levels of Sr were higherin the organisms from the reference site (Table 3). In the same studypublished by Pereira et al. (2008) it was determined that the levels ofSr in the reference site were similar to those registered in the minearea. A higher bioavailable fraction of this metal in the reference areamay be an explanation for the bioaccumulation differences observedbetween the mice inhabiting the reference and contaminated sites.Another hypothetical explanation may be related to the fact that the

Table 3Metal concentration in the organs analyzed (μg·g−1 dry weight) (mean±standard deviation, n=11 for the reference site and n=10 for the contaminated site).

Sampling sites Organs Be Al Mn Ni Cu Zn Sr Cd Pb U

Organ metal concentration±SD (μg·g−1 dry weight)Control site Bone UDL 2.95±0.49 1.36±0.24 3.41±0.42 1.66±0.53 162±54.9 180±46.7 0.1±0.01 0.27±0.09 UDL

Kidney UDL 5.66±2.19 UDL 14.8±4.35 105.3±36.42 UDL 0.7±0.27 0.33±0.19 0.012±0.0005Liver 3.73±2.02 10.97±2.93 0.23±0.14 15.6±1.9 169.5±52.9 0.32±0.07 0.16±0.06 0.05±0.02 0.004±0.0007

Uranium mining site Bone UDL 6±4.95 1.35±0.73 3.46±0.58 2.03±0.36⁎ 170±70.25 85.9±38.14⁎ 0.14±0.08⁎ 0.29±0.17 0.17±0.16⁎

Kidney UDL 6.11±1.38 UDL 16.4±1.95 114.4±32.7 UDL 1.35±0.49⁎ 0.25±0.15 0.16±0.15⁎

Liver 4.22±2.52 9.19±3.83 0.25±0.11 18.9±6.33 163.3±68.57 0.15±0.06⁎ 0.32±0.17⁎ 0.04±0.02 0.03±0.04⁎

UDL: Under Detection Limits.⁎ Statistical significant differences between sampling sites, pb0.05.

677J. Lourenço et al. / Science of the Total Environment 443 (2013) 673–680

intake of uranium is associated with kidney function, increased frac-tional excretion of calcium and phosphate in urine (Kurttio et al.,2005). Due to their chemical similarity, strontium mimics calcium inthe formation of bones, (Dahl et al., 2001), as a consequence, the accu-mulation of uranium in the mice inhabiting the contaminated areasmay stimulate the excretion of the strontium accumulated in thebones, explaining the lower levels of this metal found in these animalswhen compared to those inhabiting the reference area.

As referred, the evaluation of DNA damage in exposed animals wasperformed using the comet assay. DNA is an important target of envi-ronmental stress in terrestrial organisms (Reinecke and Reinecke,2004). The loss of DNA integrity may determine the induction of muta-tions and other irreversible toxic effects (Kurelec, 1993). The resultsshowed that there was a significant loss of DNA integrity in the organ-isms living in the mining area (Fig. 2), probably caused by thebioaccumulation of uranium and cadmium and also by the exposureto uranium and its daughter radionuclides. The level of strand breakagein DNA has been proposed as a sensitive indicator of genotoxicity andan effective biomarker in environmental biomonitoring (Shugart andTheodorakis, 1998; Shugart, 2000). Metals like Cd and U, are knownto cause severe damage to organisms, such as tissue damage (Amaralet al., 2007; Lourenço et al., 2011; Pereira et al., 2006), apoptosis induc-tion (Amaral et al., 2007), DNA damage (Lourenço et al., 2012) andchromosomal aberrations involving gross alterations of the genetic

Fig. 2. Whole blood DNA integrity in Apodemus sylvaticus inhabiting the reference(light gray) and contaminated (dark gray) sites. Data are shown as average±standarddeviation. (*) stands for statistical significant differences (pb0.05) and A.U. stands forarbitrary units.

material (Au et al., 1995; Topashka-Ancheva et al., 2003; Zaire et al.,1997).Moreover, they are capable of inducing redox imbalance in livingorganisms, leading to the production and accumulation of reactive oxy-gen species (ROS) inside the cells, causing damage to the DNAmoleculelike strand breaks, DNA-protein cross-links, alkali labile sites and oxida-tive base modifications (Barillet et al., 2010). In addition, metals cancause direct DNA strand breaks, without any involvement of free radi-cals, and inhibit DNA repair (Barillet et al., 2010). Uranium extractionalso involves the generation of solid residues contaminated with radio-elements (U and its daughter radionuclides), which contributes to anextensive contamination of the soil compartment (Lourenço et al.,2012; Pereira et al., 2008). The radionuclides found in this area aremainly alpha emitters (Carvalho et al., 2009). Alpha particles (high en-ergy, relative large mass and momentum, low velocity) have relativelyhigh linear energy transfer (LET) values (Harrison and Day, 2008). Clus-tered DNA damage, along with the degree of complexity of the damage,has shown an increase with LET (Valentin, 2003). The biological conse-quence dependswhether the damage can be repaired andwithwhat fi-delity (Harrison and Day, 2008). In this work, accumulation ofradionuclides was not analyzed, preventing the authors of this studyfrom drawing any further conclusions regarding the exposure to alpharadiation, since these elements emit far more alpha radiation thandoes uranium itself in terms of specific activity. Both laboratory andfield studies have shown that sublethal effects at a low level of biologi-cal organization like, for example DNA damage, influence energy me-tabolism, fitness and reproductive success, which leads to populationlevel effects (Jha, 2008). Therefore, the wood mice population fromthe contaminated site may suffer from increased genomic instability,which has been suggested to play an important role in the fitness de-crease of the population, not only in laboratory studies, but also infield conditions (Jha, 2008). This will ultimately lead to adverse effectson long-termpopulation survival and to the deterioration of the ecosys-tem (Jha, 2008). Further studies are still required to determine theseeffects.

Gene expression changes are increasingly used to assess the effectsof exposure to environmental agents (Banda et al., 2008). Hence, the ex-pression of P53 and Rb genes was analyzed in the kidney and liver ofmice inhabiting a pine-oak tree forest and compared to that ofmice col-lected at an abandoned uraniummining area. No significant differenceswere found on the kidney samples between both sampling sites eitherfor the P53 (1.58 normalized fold expression; p=0.11) and theRb gene (1.04 normalized fold expression; p=0.84) (Fig. 3). Regardingthe liver, no significant differences were found for the Rb gene (0.96normalized fold expression; p=0.34) (Fig. 3); yet, significant differ-ences were found for the P53 gene (2.32 normalized fold expression;p=0.03) (Fig. 3).

The tumor suppressor gene P53 is a “cellular gatekeeper” that playsan important role in the control of cell proliferation, induction ofprogrammed cell death (apoptosis) and cell differentiation (Imazawaet al., 2003). The induction of P53 can occur in response to a range ofgenotoxic stresses leading to growth arrest or apoptosis (Meek, 2004)by transcriptional regulation of the genes involved in these processes(Gillet et al., 2000). Whereas in normal cells P53 is maintained at an

Fig. 3. Rb and P53 gene expression analysis in the liver and kidney of Apodemus sylvaticus inhabiting the reference (white) and contaminated (dark gray) sites. Data are shown asaverage±standard error. (*) stands for statistical significant differences (pb0.05).

Fig. 4. Single nucleotide polymorphisms in the Rb gene sequence amplified for gene expression analysis in the kidney samples. The image on top shows the different configurationsof the melt peaks, betweenMus musculus Rb gene sequence and the normal and polymorphic sequences of Apodemus sylvaticus. The image bellow shows a schematic representationof the Rb gene sequences from both species, and the location of the polymorphisms. Sequence alignment of both normal and polymorphic Rb proteins are shown in the bottom. Thisfigure also shows the location of the exons and primers used for the amplification.

678 J. Lourenço et al. / Science of the Total Environment 443 (2013) 673–680

679J. Lourenço et al. / Science of the Total Environment 443 (2013) 673–680

undetectable or very low level, in response to a genotoxic stress (suchasUV, ionizing radiation andmultiple chemical DNA-damaging agents),P53 expression is induced and exerts its functions (Gillet et al., 2000).Most of P53 activities are dependent on its ability to activate or represstranscription (Gillet et al., 2000). If DNA damage is extensive, theexpression of P53 is induced to prevent replication of highly damagedcells (Aardema and MacGregor, 2002). Thus, in the present study, itappears that the induction of P53 gene expression is most likely a con-sequence of the DNA damages induced by the contaminated environ-ment. These results suggest that the exposure to uranium miningresidues have the potential to induce serious genomic damages and in-stability, since the expression of a crucial gene, like P53, involved in theprotection of cells against severe DNA damages and cell transformationwas significantly induced. Moreover, single nucleotide polymorphisms(SNPs) were found in the fragment of the Rb gene amplified for thegene expression analysis in the kidney, which is a target organ for theaccumulation of metals like uranium and cadmium (Fig. 4). These poly-morphisms were found in 2 out of the 5 animals randomly chosen forgene expression analysis from the contaminated site. They are locatedin exon 11 (G to T and G to A substitutions), which results in a changeof glutamate to aspartate and a valine to methionine, respectively, inthe Rb protein (Fig. 4). These are nonsynonymous SNPs, as theaminoacid sequence of the amplified fragment was changed. Many ge-netic variations are single nucleotide polymorphisms (SNPs).

Non-synonymous SNPs are “neutral” if the resulting point-mutatedprotein is not functionally discernible from the wild type (Brombergand Rost, 2007), however, “non-neutral” non-synonymous SNPs areknown to cause numerous diseases (Bromberg and Rost, 2007). In thiscase, there are no evidences known to the authors of this study, wheth-er these SNPs are “neutral” or not, as there is no information on thepresence and level of such polymorphisms in natural populations of ro-dents. Further studies are required to clarify this issue.

The Rb gene is also a tumor suppressor gene and member of theretinoblastoma gene family (Dannenberg et al., 2004). This genecontrols the G1/S transition of the cell cycle by modulating theactivity of E2F transcription factors (Dannenberg et al., 2004). Thehypophosphorylated pRb binds to E2Fs and forms complexes, whichactively repress genes that are essential for cell cycle regulation,DNA replication, DNA repair, G2/M checkpoints and differentiation(Dannenberg et al., 2004). Polymorphisms in this gene may havethree outcomes: i) the polymorphisms will prove to be a result of agenetic drift, potentially improving the function of this gene and con-ferring the organism with a better protection against tumor forma-tion; ii) the polymorphisms were caused by exposure to uraniummining wastes and are “neutral”, causing no problems in the transla-tion process, and as a consequence no harm to the organisms; iii) thepolymorphisms were caused by exposure to uranium mining wastesand are not “neutral”, potentially causing the inactivation of the Rbprotein, which may result in genomic instability and cell immortality,after severe DNA damage (Manning et al., 2010). Cell immortality,aneuploidy and continuous capability for division are key factors incarcinogenesis.

4. Conclusions

This study shows that the exposure to wastes from abandoneduranium mines had serious genotoxic effects in organisms inhabitingthe contaminated area. These effects include DNA strand breaks andchanges in the expression of important genes like P53, which is crucialfor the organism defense against tumor formation, and is reported tobe induced in cells with extensive DNA damage. These events showthe potential danger of the exposure to these contaminants. Two mainreasons can be highlighted for the relevance of the present study: i) itwas conducted inmice, which are frequently employed asmodel organ-isms for the evaluation of toxic effects of contaminants in humans, andare therefore often used as bioindicator species for the assessment of

risks of human environmental exposure; ii) it consists of an in situevaluation of the exposure to contaminants. Moreover, the presenceof nonsynonymous SNPs was also reported in the fragment of the Rbgene amplified in this study. However, it is not clear if these polymor-phisms result from a genetic drift or if they are “neutral”, preventingthe authors of drawing any conclusion on their origin and potentialconsequences.

Since small mammals are often used as bioindicator species for theassessment of human risks, the results of this study raise concerns onpotential effects in human populations living near uranium mines,despite the obvious differences in the routes and in the level of expo-sure of wood mice and human populations.

Acknowledgments

Fundação para a Ciência e Tecnologia (Portuguese Ministry ofScience, Technology and Higher Education) has funded this studythrough a PhD Grant to Joana Lourenço (ref. SFRH/BD/37007/2007)and through the research Project ENGENUR (ref. PTDC/AAC-AMB/114057/2009). This research was also funded by FEDER throughCOMPETE and also partially funded by FSE, POPH funds (ProgramaCiência 2007) and the Portuguese Government (Programa 16 CiênciaInovação 2010).

References

Aardema MJ, MacGregor JT. Toxicology and genetic toxicology in the new era of“toxicogenomics”: impact of “-omics” technologies. Mutat Res-Fund Mol M2002;499:13–25.

Amaral A, Cabral C, Guedes C, Rodrigues A. Apoptosis, metallothionein, and bioavail-able metals in domestic mice (Mus musculus L.) from a human-inhabited volcanicarea. Ecotoxicology 2007;16:475–82.

André A, Antunes SC, Gonçalves F, Pereira R. Bait-lamina assay as a tool to assess theeffects of metal contamination in the feeding activity of soil invertebrates withina uranium mine area. Environ Pollut 2009;157:2368–77.

Antunes S, Castro B, Nunes B, Pereira R, Gonçalves F. In situ bioassay with Eisenia andreito assess soil toxicity in an abandoned uranium mine. Ecotoxicol Environ Saf2008a;71:620–31.

Antunes S, Castro B, Pereira R, Gonçalves F. Contribution for tier 1 of the ecological riskassessment of Cunha Baixa uraniummine (Central Portugal): II. Soil ecotoxicologicalscreening. Sci Total Environ 2008b;390:387–95.

APHA. Standardmethods for the examination of water andwastewater; 1995 (Washington:American Public Health Association).

Au WW, Lane RG, Legator MS, Whorton EB, Wilkinson GS, Gabehart GJ. Biomarkermonitoring of a population residing near uranium mining activities. EnvironHealth Perspect 1995;103:466.

Banda M, Bommineni A, Thomas RA, Luckinbill LS, Tucker JD. Evaluation and validationof housekeeping genes in response to ionizing radiation and chemical exposure fornormalizing RNA expression in real-time PCR. Mutat Res Genet Toxicol Environ2008;649:126–34.

Barillet S, Adam-Guillermin C, Palluel O, Porcher J-M, Devaux A. Uraniumbioaccumulation and biological disorders induced in zebrafish (Danio rerio) aftera depleted uranium waterborne exposure. Environ Pollut 2010;159:495–502.

Beernaert J, Scheirs J, Leirs H, Blust R, Verhagen R. Non-destructive pollution exposureassessment by means of wood mice hair. Environ Pollut 2007;145:443–51.

Bernard F, Brulle F, Douay F, Lemière S, Demuynck S, Vandenbulcke F. Metallic traceelement body burdens and gene expression analysis of biomarker candidates inEisenia fetida, using an “exposure/depuration” experimental scheme with fieldsoils. Ecotoxicol Environ Saf 2010;73:1034–45.

Beyersmann D. Effects of carcinogenic metals on gene expression. Toxicol Lett 2002;127:63–8.

Bromberg Y, Rost B. SNAP: predict effect of non-synonymous polymorphisms on func-tion. Nucleic Acids Res 2007;35:3823–35.

Brugge D, Buchner V. Health effects of uranium: new research findings. Rev EnvironHealth 2011;26:231–49.

Brulle F, Cocquerelle C, Mitta G, Castric V, Douay F, Leprêtre A, et al. Identification andexpression profile of gene transcripts differentially expressed during metallicexposure in Eisenia fetida coelomocytes. Dev Comp Immunol 2008;32:1441–53.

Brulle F, Morgan AJ, Cocquerelle C, Vandenbulcke F. Transcriptomic underpinning oftoxicant-mediated physiological function alterations in three terrestrial inverte-brate taxa: a review. Environ Pollut 2010;158:2793–808.

Carvalho F, Madruga M, Reis M, Alves J, Oliveira J, Gouveia J, et al. Radioactivity in theenvironment around past radium and uranium mining sites of Portugal. J EnvironRadioact 2007;96:39–46.

Carvalho F, Oliveira J, Faria I. Alpha emitting radionuclides in drainage from Quinta doBispo and Cunha Baixa uraniummines (Portugal) and associated radiotoxicologicalrisk. Bull Environ Contam Toxicol 2009;83:668–73.

680 J. Lourenço et al. / Science of the Total Environment 443 (2013) 673–680

Dahl SG, Allain P, Marie PJ, Mauras Y, Boivin G, Ammann P, et al. Incorporation anddistribution of strontium in bone. Bone 2001;28:446–53.

Dannenberg J-H, Schuijff L, Dekker M, van der Valk M. Riele Ht. Tissue-specific tumorsuppressor activity of retinoblastoma gene homologs p107 and p130. Genes Dev2004;18:2952–62.

Dhawan A, Bajpayee M, Parmar D. Comet assay: a reliable tool for the assessment ofDNA damage in different models. Cell Biol Toxicol 2009;25:5-32.

Festa F, Cristaldi M, Ieradi LA, Moreno S, Cozzi R. The Comet assay for the detection ofDNA damage in Mus spretus from Doñana National Park. Environ Res 2003;91:54–61.

Garcia O, Mandina T, Lamadrid AI, Diaz A, Remigio A, Gonzalez Y, et al. Sensitivity andvariability of visual scoring in the comet assay — results of an inter-laboratoryscoring exercise with the use of silver staining. Mutat Res Fundam Mol MechMutagen 2004;556:25–34.

Genovese C, Trani D, Caputi M, Claudio PP. Cell cycle control and beyond: emergingroles for the retinoblastoma gene family. Oncogene 2006;25:5201–9.

Gillet R, Grimber G, Bennoun M, Caron de Fromentel C, Briand P, Joulin V. The conse-quence of p53 overexpression for liver tumor development and the response oftransformed murine hepatocytes to genotoxic agents. Oncogene 2000;19:3498–507.

Harris GK, Shi X. Signaling by carcinogenic metals and metal-induced reactive oxygenspecies. Mutat Res Fundam Mol Mech 2003;533:183–200.

Harrison J, Day P. Radiation doses and risks from internal emitters. J Radiol Prot2008;28:137–59.

Ieradi LA, Moreno S, Bolívar JP, Cappai A, Di Benedetto A, Cristaldi M. Free-livingrodents as bioindicators of genetic risk in natural protected areas. Environ Pollut1998;102:265–8.

Imazawa T, Nishikawa A, Toyoda K, Furukawa F, Mitsui M, Hirose M. Sequential alter-ation of apoptosis, p53 expression, and cell proliferation in the rat pancreas treatedwith 4-hydroxyaminoquinoline 1-oxide. Toxicol Pathol 2003;31:625–31.

Jha AN. Ecotoxicological applications and significance of the comet assay. Mutagenesis2008;23:207–21.

Kurelec B. The genotoxic disease syndrome. Mar Environ Res 1993;35:341–8.Kurttio P, Komulainen H, Leino A, Salonen L, Auvinen A, Saha H. Bone as a possible

target of chemical toxicity of natural uranium in drinking water. Environ HealthPerspect 2005;113:68.

Lee MS, Cho SJ, Tak ES, Lee JA, Cho HJ, Park BJ, et al. Transcriptome analysis in themidgut of the earthworm (Eisenia andrei) using expressed sequence tags. BiochemBiophys Res Commun 2005;328:1196–204.

Léon G, Pérez LE, Linares JC, Hartmann A, Quintana M. Genotoxic effects in wild rodents(Rattus rattus and Mus musculus) in an open coal mining area. Mutat Res GenetToxicol Environ 2007;630:42–9.

Lourenço J, Silva A, Carvalho F, Oliveira J, Malta M, Mendo S, et al. Histopathologicalchanges in the earthworm Eisenia andrei associated with the exposure to metalsand radionuclides. Chemosphere 2011;85:1630–4.

Lourenço J, Pereira R, Silva A, Carvalho F, Oliveira J, Malta M, et al. Evaluation of thesensitivity of genotoxicity and cytotoxicity endpoints in earthworms exposed insitu to uranium mining wastes. Ecotoxicol Environ Saf 2012;75:46–54.

Lozano JC, Vera Tomé F, Gómez Escobar V, RodrI B. Radiological characterization of auranium mine with no mining activity. Appl Radiat Isot 2000;53:337–43.

Manning AL, Longworth MS, Dyson NJ. Loss of pRB causes centromere dysfunction andchromosomal instability. Genes Dev 2010;24:1364–76.

Marcheselli M, Sala L, Mauri M. Bioaccumulation of PGEs and other traffic-relatedmetals in populations of the small mammal Apodemus sylvaticus. Chemosphere2010;80:1247–54.

Meek DW. The p53 response to DNA damage. DNA Repair 2004;3:1049–56.Mertens J, Luyssaert S, Verbeeren S, Vervaeke P, Lust N. Cd and Zn concentrations in small

mammals and willow leaves on disposal facilities for dredged material. EnvironPollut 2001;115:17–22.

Neves MO, Matias MJ, Abreu MM, Magalhães MCF, Basto MJ. Abandoned mine site char-acterization for remediation: the case of the Cunha Baixa uranium mine (Viseu,

Portugal). Environmental contamination from uranium production facilities andtheir remediationIAEA Proceeding Series, ISBN: 92-0-104305-8; 2005. p. 159–69.

Nogueira PR, Lourenço J, Mendo S, Rotchell JM. Mutation analysis of ras gene in theliver of European eel (Anguilla anguilla L.) exposed to benzo[a]pyrene. Mar PollutBull 2006;52:1611–6.

Paternain JL, Domingo JL, Ortega A, Llobet JM. The effects of uranium on reproduction,gestation, and postnatal survival in mice. Ecotoxicol Environ Saf 1989;17:291–6.

Pereira R, Pereira ML, Ribeiro R, Gonçalves F. Tissues and hair residues and histopathologyin wild rats (Rattus rattus L.) and Algerian mice (Mus spretus Lataste) from an aban-doned mine area (Southeast Portugal). Environ Pollut 2006;139:561–75.

Pereira R, Antunes SC, Marques SM, Gonçalves F. Contribution for tier 1 of the ecologicalrisk assessment of Cunha Baixa uranium mine (Central Portugal): I soil chemicalcharacterization. Sci Total Environ 2008;390:377–86.

Pereira R, Marques CR, Ferreira MJS, Neves MFJV, Caetano AL, Antunes SC, et al. Phyto-toxicity and genotoxicity of soils from an abandoned uranium mine area. Appl SoilEcol 2009;42:209–20.

Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR.Nucleic Acids Res 2001;29:e45.

Pfeifer G. Environmental exposures and mutational patterns of cancer genomes.Genome Med 2010;2:54.

Pirooznia M, Gong P, Guan X, Inouye L, Yang K, Perkins E, et al. Cloning, analysis andfunctional annotation of expressed sequence tags from the earthworm Eiseniafetida. BMC Bioinf 2007;8:S7.

Raymond-Whish S, Mayer LP, O'Neal T, Martinez A, Sellers MA, Christian PJ, et al.Drinking water with uranium below the US EPA water standard causes estrogenreceptor-dependent responses in female mice. Environ Health Perspect2007;115:1711.

Reinecke SA, Reinecke AJ. The comet assay as biomarker of heavy metal genotoxicity inearthworms. Arch Environ Contam Toxicol 2004;46:208–15.

Rogival D, Scheirs J, De CoenW, Verhagen R, Blust R. Metal blood levels and hematolog-ical characteristics in wood mice (Apodemus sylvaticus L.) along a metal pollutiongradient. Environ Toxicol Chem 2006;25:149–57.

Sánchez-Chardi A, Peñarroja-Matutano C, Ribeiro CAO, Nadal J. Bioaccumulation ofmetals and effects of a landfill in small mammals. Part II. The wood mouse,Apodemus sylvaticus. Chemosphere 2007;70:101–9.

Sánchez-Chardi A, Peñarroja-Matutano C, Borrás M, Nadal J. Bioaccumulation of metalsand effects of a landfill in small mammals Part III: structural alterations. EnvironRes 2009;109:960–7.

Shugart LR. DNA damage as a biomarker of exposure. Ecotoxicology 2000;9:329–40.Shugart L, Theodorakis C. New trends in biological monitoring: application of

biomarkers to genetic ecotoxicology. Biotherapy 1998;11:119–27.Temesvári M, Paulik J, Kóbori L, Monostory K. High-resolution melting curve analysis to

establish CYP2C19*2 single nucleotide polymorphism: comparison with hydrolysisSNP analysis. Mol Cell Probes 2011;25:130–3.

Topashka-Ancheva M, Metcheva R, Teodorova S. Bioaccumulation and damaging actionof polymetal industrial dust on laboratory mice Mus musculus alba II. Genetic, cell,and metabolic disturbances. Environ Res 2003;92:152–60.

UNSCEAR. Sources and effects of ionizing radiation. Report to the general assembly,with scientific annexes. United Nations Scientific Committee on the Effects ofAtomic RadiationNew York: United Nations; 2000.

Valentin J. Relative biological effectiveness (RBE), quality factor (Q), and radiationweighting factor (wR): ICRP Publication 92. Ann ICRP 2003;33:1-121.

Van der Schalie WH, Gardner Jr HS, Bantle JA, De Rosa CT, Finch RA, Reif JS, et al.Animals as sentinels of human health hazards of environmental chemicals. EnvironHealth Perspect 1999;107:309.

Willems E, Leyns L, Vandesompele J. Standardization of real-time PCR gene expressiondata from independent biological replicates. Anal Biochem 2008;379:127–9.

Zaire R, Notter M, Riedel W, Thiel E. Unexpected rates of chromosomal instabilities andalterations of hormone levels in Namibian uranium miners. Radiat Res 1997:579–84.