A systematic study of the disposition and metabolismof mercury species in mice after exposure to low levelsof thimerosal (ethylmercury)$

Maria Fernanda Hornos Carneiro a, Juliana Maria Oliveira Souza a, Denise Grotto a,b,Bruno Lemos Batista a,c, Vanessa Cristina de Oliveira Souza a, Fernando Barbosa Jr.a,n

a Laboratório de Toxicologia e Essencialidade de Metais, Faculdade de Ciências Farmacêuticas de Ribeirão Preto – USP, Avenida do Café, s/n, Monte Alegre,CEP 14040-903 Ribeirão Preto, SP, Brazilb Programa de Pós-Graduação em Ciências Farmacêuticas, Universidade de Sorocaba, Rodovia Raposo Tavares km 92.5, CEP 18023-000 Sorocaba, SP, Brazilc Centro de Ciências Naturais e Humanas, Universidade Federal do ABC, Bloco B, Avenida dos Estados 5001, CEP 0910-170 Santo André, SP, Brazil

a r t i c l e i n f o

Article history:Received 7 June 2014Received in revised form15 July 2014Accepted 17 July 2014

Keywords:ThimerosalEthylmercuryHalf-lifeDistributionDealkylation

a b s t r a c t

Thimerosal (TM) is an ethylmercury (etHg)-containing preservative used in some vaccines despite verylimited knowledge on the kinetics and direct interaction/effects in mammals' tissues after exposure.Thus, this study aimed to evaluate the kinetics of Hg species in mice in a time course analysis afterintramuscular injection of TM, by estimating Hg half-lives in blood and tissues. Mice were exposed toone single intramuscular dose of 20 mg of Hg as TM. Blood, brain, heart, kidney and liver were collected at0.5 hour (h), 1 h, 8 h, 16 h, 144 h, 720 h and 1980 h after TM exposure (n¼4). Hg species in animal tissueswere identified and quantified by speciation analysis via liquid chromatography hyphenated withinductively coupled mass spectrometry (LC–ICP-MS). It was found that the transport of etHg frommuscle to tissues and its conversion to inorganic Hg (inoHg) occur rapidly. Moreover, the conversionextent is modulated in part by the partitioning between EtHg in plasma and in whole blood, since etHg israpidly converted in red cells but not in a plasma compartment. Furthermore, the dealkylationmechanism in red cells appears to be mediated by the Fenton reaction (hydroxyl radical formation).Interestingly, after 0.5 h of TM exposure, the highest levels of both etHg and inoHg were found inkidneys (accounting for more than 70% of the total Hg in the animal body), whereas the braincontributed least to the Hg body burden (accounts for o1.0% of total body Hg). Thirty days after TMexposure, most Hg had been excreted while the liver presented the majority of the remaining Hg.Estimated half-lives (in days) were 8.8 for blood, 10.7 for brain, 7.8 for heart, 7.7 for liver and 45.2 forkidney. Taken together, our findings demonstrated that TM (etHg) kinetics more closely approximatesHg2þ than methylmercury (meHg) while the kidney must be considered a potential target for etHgtoxicity.

& 2014 Elsevier Inc. All rights reserved.

1. Introduction

Thimerosal (TM), which contains ethylmercury (etHg), hasbeen widely used as a preservative in a number of drug products,including vaccines, to help prevent life-threatening contamination

with microbes (Tan and Parkin, 2000). However, the potentialneurotoxic effects of organomercurial compounds, even at lowexposures (Lebel et al., 1998; Berman et al., 2008; Delong, 2011;Bose et al., 2012; Petroni et al., 2012; Ida-Eto et al., 2013), haveprovoked concerns about the use of thimerosal in vaccines andother products (Clements et al., 2000; Ball et al., 2001).

The toxic properties of Hg compounds are directly related tothe chemical form of the element. In general, exposure to organicforms of Hg is associated with nervous system damage, whileinorganic forms are closely connected to renal damage (Clarksonand Magos, 2006). However, the toxicokinetics and potential toxicproperties of TM (etHg) are mostly unknown (WHO, 2012).

Due to the lack of information about the behavior of TM in themammalian body, the initial risk assessments for etHg were basedon studies of oral methylmercury (meHg) toxicity. However, recent

Contents lists available at ScienceDirect

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

Environmental Research

http://dx.doi.org/10.1016/j.envres.2014.07.0090013-9351/& 2014 Elsevier Inc. All rights reserved.

☆Funding sources supporting the work: Fundação de Amparo à Pesquisa do Estadode São Paulo (FAPESP) and Conselho Nacional de Desenvolvimento Científico eTecnológico (CNPq).

n Corresponding author. Fax: þ55 16 3602 4725.E-mail addresses: mafehoca@fcfrp.usp.br (M.F.H. Carneiro),

souza.jmo@gmail.com (J.M. Oliveira Souza), denise.grotto@prof.uniso.br (D. Grotto),bruno.lemos@ufabc.edu.br (B.L. Batista),vcosouza@fcfrp.usp.br (V.C. de Oliveira Souza),fbarbosa@fcfrp.usp.br (F. Barbosa Jr.).

Environmental Research 134 (2014) 218–227

data indicate that the kinetics of tissue disposition and metabo-lism differ substantially between these two forms of organic Hg(Clarkson, 2002; Magos, 2003; Burbacher et al., 2005), indicatingthat meHg is not a suitable reference for risk assessment fromexposure to TM-derived Hg. Therefore, the knowledge of thetoxicokinetics of TM is mandatory to afford a meaningful assess-ment of the developmental effects of TM-containing vaccines.

Thus, the present study aimed to investigate the body burdendisposition of TM in mice after a single low-dose exposure. Twoforms of Hg found after the exposure (etHg and the dealkylatedform (inoHg)) were measured in blood and animal tissues by usinghyphenation of High Performance Liquid Chromatography toInductively Coupled Plasma Mass Spectrometry (HPLC–ICP-MS)in a time course analysis to estimate Hg half-lives. It was alsodemonstrated that the mechanism of etHg dealkylation appears tobe mediated by the Fenton reaction (hydroxyl radical formation).

2. Material and methods

2.1. Reagents

All reagents used were of analytical grade and the solutionswere prepared using high purity water with a resistivity of18.2 MΩ cm (Milli-Q Plus, Millipore, Bedford, MA, USA). A solutionof 37% hydrochloric acid (Merck, Darmstadt, Germany) was doublydistilled in a quartz sub-boiling apparatus (Kürner Analysentech-nik, Rosenheim, Germany). A clean laboratory and laminar-flowhood capable of producing class 100 were used for preparingsolutions and samples for the HPLC–ICP-MS technique. All solu-tions used were stored in high-density polyethylene bottles.Materials were cleaned by soaking in 10% v/v HNO3 for 24 h,rinsing five times with ultrapure water and dried in a class 100laminar flow hood before use. All operations were performed on aclean bench. A 10 mg/l standard solution of inoHg was obtainedfrom Perkin-Elmer (PerkinElmer, Norwalk, CT). A 1000 mg/l stan-dard solution of meHg chloride (CH3HgCl) and 1000 mg/l standardsolution of etHg chloride (CH3CH2HgCl) in water were obtainedfrom Alfa Aesar (MA, USA). Analytical calibration standards of Hgspecies were prepared daily over the range of 0.0–20.0 mg/l for theHPLC–ICP-MS by suitable serial dilutions of the stock solution.

Additional reagents used for the speciation studies includedmethanol (99.9% v/v, HPLC-grade, Merck, Germany), mercap-toethanol (Sigma-Aldrich, USA) and L-cysteine (Fluka, Japan).Ammonium acetate (99.99%) was obtained from Aldrich ChemicalCompany (Milwaukee, USA) and formic acid, dimethyl sulfoxide(DMSO) and hydrogen peroxide (H2O2) were purchased fromSigma-Aldrich (USA).

2.2. Instrumentation and method

All measurements were carried out with a HPLC (Perkin-Elmermodel L-200, a six-port injector Rheodyne 9725 and a reverse-phase column C8 Brownlee 3 mm, 33 mm�4.6 mm) hyphenatedto an ICP-MS (Elan DRCII PerkinElmer, Norwalk, CT, USA). Argonwith a purity of 99.999% (White Martins, São Paulo, Brazil) wasused throughout. A Vibracell VC 100 ultrasonic processor witha titanium probe controlled by USS-100 (Sonics & Materials,Danbury, CT, USA) was employed for Hg extraction from samples.The ultrasonic probe was set at an amplitude of 80%, power of50 W and a frequency of 20 kHz. Samples volume injection was100 ml. All separations were performed at room temperature underisocratic conditions. The isocratic mobile phase was 3% v/vmethanolþ97% v/v (0.5% v/v 2-mercaptoethanolþ0.05% v/vformic acid). The flow rate was 1.2 ml/min. Data were evaluatedusing the software Chromeras (supplied with the instrument) and

quantified by external calibration using a peak area. The completedescription of experimental conditions for both ICP-MS and HPLCwere provided by Souza et al. (2013). Data were validated by theanalysis of Standard Reference Materials (SRM) 966 Toxic Metals inBovine Blood from the National Institute of Standards and Tech-nology (NIST, Gaithersburg, MD) and Certified Reference Materials(CRMs) DOLT-3, DORM-3 and TORT-2 from the National ResearchCouncil Canada (NRCC). Found values were in adequate agreementwith target values according to a t test at the 95% confidence level.

2.3. Experimental design

2.3.1. In vivo studyAnimals were handled and treated according to the guidelines

of the Committee on the Care and Use of Experimental AnimalResources of the University of São Paulo, Brazil (Opinion number12.1.1158.53.1, issued on the 14th of December of 2012). Male Swissmice weighing approximately 25 g (sixth week of life) wereobtained from the central animal facility. Animals were subjectedto 12 h light/12 h dark cycles in an air-conditioned room(22–25 1C) with free access to food and water.

The animals (n¼28) were exposed to 20 mg of Hg in the form ofTM (sodium ethylmercury thiosalicylate, Sigma Chemical, St Louis,MO) through IM injection, which corresponded to approximately0.8 mg Hg/kg. The dose was adapted to the mouse body weightand corresponded to a dose 20 times higher than that a child of3.5 months of age receives in terms of Hg, only from vaccines(Clements et al., 2000). Animals were euthanized by intraperito-neal injection of sodium pentobarbital (150 mg/kg) to collectblood and organs (brain, heart, kidney and liver) after 0.5 hour(h), 1 h, 8 h, 16 h, 144 h (or 6 days (d)), 720 h (or 30 d) and 1980 h(or 80 d) after TM exposure (4 animals per time of study). Theblood and organs were placed in metal-free Eppendorf tubes andkept at �80 1C until analysis. Plasma samples were not included inthe present study due to the limitation to collect the necessaryvolumes from mice for Hg speciation.

The total mass of each Hg species accumulated in each samplewas calculated, dividing the concentration found (mg/ml or mg/g)by the total volume of blood in the animal body (mice of 25 g,approximately 1.6 ml) or the mean mass obtained of the respectiveorgan (in grams)—previously weighed in 3 control animals atapproximately 25 g. This mass balance was performed due to thefact that Hg concentrations (as mg/ml or mg/g) do not reflectdirectly the absolute amount of Hg accumulated in tissues, i.e.,tissues with smaller sizes and lower Hg concentration may containmore Hg than tissues with higher Hg concentrations.

Moreover, along the time course, the total amount of Hgobtained at each moment—i.e., the sum of Hg levels measured inall monitored tissues—was considered to be 100% (since most ofthe Hg is concentrated in these compartments). This assumptionwas made in order to better illustrate the contribution of eachorgan to Hg accumulation over time. Then, the percentages ofinoHg and etHg were calculated in each organ at one time pointassuming the whole amount of the exposure dose (20 mg) tobe 100%.

2.3.2. In vitro studyAn in vitro experiment was conducted to evaluate the stability

of TM and etHg in whole blood, plasma and erythrocytes. This partof the study was conducted in three different experiments to:(i) evaluate the stability of TM and etHg in whole blood, plasmaand erythrocytes samples; (ii) and (iii) to elucidate the mechanismbehind the dealkylation of etHg in blood. These experimentswere performed on samples collected from two healthy humanvolunteers.

M.F.H. Carneiro et al. / Environmental Research 134 (2014) 218–227 219

The procedure to obtain the samples was identical in allexperiments. Human blood samples (30 ml) were collected fromthe cubital median vein in metal-free tubes containing heparin.After homogenization, 15 ml of blood was centrifuged at 3000 rpmfor 10 min to obtain plasma and erythrocytes. After plasmacollection and buffy coat removal, erythrocytes were washed threetimes in cold isotonic phosphate buffered saline (PBS) solutionaccording to the procedure described by Harisa et al. (2012). Allsolutions used had been freshly prepared.

In a preliminary in vitro experiment, aliquots of whole blood,plasma and erythrocytes were kept in a water bath at 37 1C. Then,to these aliquots, TM or etHg solution was added resulting in afinal Hg concentration of 3 mg/l. 24 h after the TM/etHg incuba-tion, samples were analyzed for Hg species determination (n¼4).Since there was no statistical difference in the levels of Hg speciesafter incubation of samples with TM or etHg, in the followingin vitro studies, we decided to incubate the samples withonly etHg.

Subsequent in vitro experiments were carried out by the sameprocedure followed in the preliminary experiment. Briefly,aliquots of whole blood, plasma and erythrocytes were maintainedin a water bath at 37 1C. Then, to these aliquots, solutions of etHgand iron (Fe) chloride (FeCl3) were added resulting in finalconcentrations of 3 mg/l of Hg and 400 mg/dl of Fe. This Feconcentration is about twice the typical concentration found inhuman plasma. 24 h after the incubation with etHg or etHg/Fe,samples were analyzed for Hg species determination (n¼4).

In the last in vitro experiment, the analysis was focused only onplasma samples. Human plasma aliquots were kept in a water bathat 37 1C after collection and centrifugation. Then, to each aliquot,solutions of etHg, H2O2 and DMSO were added to obtain plasmasamples with the following concentrations:

(a) etHg (3 mg/l); (b) etHg (3 mg/l) plus H2O2 (10 mM); (c) etHg(3 mg/l) plus H2O2 (10 mM) plus DMSO (100 mM). Then, 24 h later,samples were analyzed for Hg species determination (n¼4).

2.4. Statistical analysis

Descriptive analysis was used to elucidate some findings of thepresent study. For the in vivo study, a t test was employed tocompare the mean Hg levels (inoHg and etHg)7standard devia-tion (SD) obtained in blood and organs at each time point in thestudy. Hg half-lives were estimated using the pharmacokineticmodeling software SAAM II (SAAM Institute, Seattle, WA, USA). Inthe kidneys, a one-compartment model failed to provide asatisfactory fit of the mean Hg concentrations–time data. Afterlog-scale transformation, the half-lives for the mono/biexponentialdecline of Hg concentration were estimated by standard formulas(Bonate, 2006; Gibaldi and Perrier, 1982; Wolfsegger, 2006). Theestimation of initial and terminal half-lives was made using thebiexponential model: y¼a1 nexp(�b1 næ)þa2 nexp(�b2 næ) with aparameterization to ensure b14b240 fitted by the least squarescriteria with optimization of the package function based on themethod of Nelder–Mead. Curve peeling (Foss, 1969) was employedto obtain starting values for nonlinear model fitting. When noadequate starting values were determined by curve peeling, asingle exponential model was fitted with starting values obtainedfrom an Ordinary Least Squares regression on log-transformedvalues with a parameterization to ensure a slope40.

In the in vitro study, One-way ANOVA followed by Tukey's test,when needed, was used to compare the mean concentrations ofHg species obtained in whole blood, plasma and human erythro-cytes. The influence of Fe/H2O2/H2O2þDMSO upon etHg con-version was tested by means of the same combination ofstatistical tests.

The statistical significance was set at 5%. Statistical proceduresand figures preparation were performed using GraphPad Prismversion 5 (GraphPad Software, Inc. CA, USA).

3. Results and discussion

3.1. Hg distribution in tissues of mice exposed to thimerosal

Table 1 shows the Hg (etHg and InoHg) amounts in mass unitsaccumulated in the organs and blood of mice at different timepoints after TM exposure. In each compartment, Hg amounts wereexpressed in mg, obtained by dividing the concentration found(mg/ml or mg/g) by the total volume of blood in the animal body(mice of 25 g, about 1.6 ml) or the mean mass of the respectiveorgan (in grams)—previously weighed in 3 control animals with abody weight of 25 g. This mass balance was performed due to thefact that Hg concentrations (as mg/ml or mg/g) do not reflectdirectly the absolute amount of Hg accumulated in tissues, i.e.,tissues with smaller sizes and lower Hg concentration may containmore Hg than tissues with higher Hg concentrations.

The total amount of Hg measured 0.5 h after TM exposure (summeasured in organs and blood) was 17.370.8 mg (Table 1). Theremaining quantity of Hg (considering the dose administered of20 mg) had been distributed to other organs or excreted. However,it is reasonable to consider the sum of Hg measured in the animalblood, brain, heart, liver and kidney as approximating 100% of thetotal Hg in the body, since the contribution of other compartmentsis much less relevant. Taking this in account, the contribution ofeach organ and blood to the total amount is also represented as apercentage (Table 1).

It was apparent that there is no excretion of Hg from 0.5 h to16 h after the TM exposure, since the sum of Hg levels obtainedin the monitored organs changed slightly (ranging from17.3170.8 mg (0.5 h) to 17.1770.36 mg (16 h)). Moreover, duringthis same period there was no significant change in the contribu-tion of blood and organs in terms of percentage of the Hg bodyburden—considering each time point individually (Table 1). Thekidney was the organ with the greatest contribution to Hg bodyburden (around 70–75%) followed by liver (20–24%). A major causeof accumulation of Hg in the renal tissue is likely to be related toits physiological function of filtration. The large renal blood flow ofabout 25% of the cardiac output is another important featurecontributing to uptake of toxic chemicals, since the kidney is full ofessential transport systems (Berndt et al., 1985).

After 0.5 h of TM exposure, the amount of Hg in the bloodcompartment corresponded to approximately 7.6% of the injecteddose. This is consistent with the results obtained for Hg levels inblood from infants after receiving 12.5 mg of Hg as TM (from 4.2%to 5.1% of the injected dose) (Stajich et al., 2000; Aschner andCeccatelli, 2010). On the other hand, the brain accounted for lessthan 1% of the total Hg in animal body and was the organ with lessHg accumulated after TM exposure and the tissue that has thehighest loss of the accumulated Hg in percentage (from 0.5 h to6 d¼94.1%, see Table 1). The brain-to-blood Hg ratio was ratherlower than 1 (ranging from 0.1 to 0.4—Table 1). This rate is in closeagreement with other studies (Harry et al., 2004; Orct et al., 2006;Zareba et al., 2007; Rodrigues et al., 2010; Blanuša et al., 2012).Moreover, from 0.5 h to 16 h, Hg was found in brain tissue as bothinoHg and etHg (approximately 50/50), indicating that despite thegreater molecular size compared with meHg, etHg can cross theblood–brain-barrier. In fact, the uptake of meHg and etHg com-plexes with cysteine into glioma cells was shown to be mediated,at least in part, through the L-type neutral amino acid carriertransport system (Zimmermann et al., 2013), which corroboratesthis result. However, at 6 d after exposure, no etHg was detected in

M.F.H. Carneiro et al. / Environmental Research 134 (2014) 218–227220

this tissue (inoHg: 100%). Two main scenarios may explain thepresence of inoHg in the brain: (i) etHg crosses the blood–brainbarrier and, once inside the brain, is converted to inoHg, asobserved for meHg by Vahter et al. (1995) or; (ii) due to directuptake of inoHg from blood, since studies have reported inoHg inbrain after exposure to inoHg (Arvidson, 1992; Pamphlett andWaley, 1996; Pamphlett and Jew, 2013). However, both mechan-isms can be occurring simultaneously. Despite the low Hg levelsfound in the brain of mice after TM exposure, and the predomi-nance of inoHg, toxic risk cannot be excluded, since tissuesrespond differently to Hg. Yasutake et al. (2010) found in micean increased rate of movement in the open field test associatedwith brain Hg levels of 0.4 mg/g, 3 weeks after inoHg exposure byan intraventricular injection. This concentration is close to theinoHg levels we found in the present study after TM exposure(around 0.2 mg/g at 0.5 h and 0.1 mg/g at 1 h—Fig. 1). Therefore, asstated by Blair et al. (1975), any Hg level in brain, either organic orinorganic, should be viewed as potentially hazardous.

It is also interesting to observe that from 16 h to 6 d after TMexposure there was a considerable drop in the total Hg measuredas the sum of blood and tissue levels (from 17.2 mg to 4.83 mg). Atthe same time, the kidney Hg level dropped from 12.1 mg to 3.5 mg.

Thus, the contribution of the kidney to the total reduction in theHg levels was 69% (8.5 mg), meaning that a significant part of Hgwas excreted by urine. On the other hand, in the same period, liverHg levels account for less than 25% of the total Hg in the rodentbody, meaning that the liver-biliary system is not the main routeof Hg excretion. We did not measure Hg levels in urine or feces inthe present study; however, it is reasonable to deduce that asignificant part of the Hg excreted is eliminated by urine, sincemost of the Hg found in kidney was found in the inorganic form.This finding is contrary to the observations of Pichichero et al.(2002), who reported that the Hg levels determined in stool andurine samples collected from children given vaccines containingTM indicated a substantial excretion of Hg taking place via thefecal route. However, in their study Pichichero et al. (2002)collected spot urine samples (at 1 time interval after vaccination)and reported the Hg levels in concentration and not in absoluteamount of Hg excreted (i.e. without volume correction).

At 80 d after TM exposure, Hg was still detected in the kidneybut not in other compartments. It is also important to point outthat elevated kidney Hg levels were also observed in an etHgpoisoning episode in Iraq (Hilmy et al., 1976). Suzuki et al. (1973)reported levels of total and inoHg in the tissues of a 13-year-old

Table 1Hg levels (μg) determined in mice organs after thimerosal exposure and the percentage of it considering the sum of Hg measured in the animal blood, brain, heart, liver andkidney as 100% in each time point.

Time elapsed Organ Total Hg (lg)7SD inoHg (%) etHg (%)% of Hg considering the

sum per time pointHg [organ]/[blood]

0.5 h

Heart 0.1170.02 100 0 0.65 0.1Liver 3.2670.31 53 47 18.8 2.5Kidney 12.4473.34 60 40 71.9 9.4Brain 0.1770.04 52 48 0.97 0.1Blood 1.3370.32 46 54 7.66 1.0Total 17.3170.8

1 h

Heart 0.4370.05 94 6 2.35 0.4Liver 3.6670.44 48 52 20 3.1Kidney 12.9571.91 65 35 70.8 10.9Brain 0.0770.01 56 44 0.39 0.1Blood 1.1970.10 71 29 6.48 1.0Total 18.3072.44

8 h

Heart 0.2670.02 95 5 1.46 0.7Liver 3.4670.33 42 58 19.33 9.4Kidney 13.7671.48 67 33 76.9 37.4Brain 0.0470.001 53 47 0.2 0.1Blood 0.3770.02 75 25 2.06 1.0Total 17.8970.37

16 h

Heart 0.2470.02 87 13 1.42 0.5Liver 4.2770.76 43 57 24.9 8.4Kidney 12.1070.82 80 20 70.4 23.9Brain 0.0670.0001 55 45 0.32 0.1Blood 0.5170.19 83 27 2.95 1.0Total 17.1770.36

144 h (6 d)

Heart 0.0470.01 77 23 0.88 0.5Liver 1.1970.26 67 33 24.7 14.0Kidney 3.571.29 92 8 72.5 41.2Brain 0.0170.0001 100 0 0.21 0.1Blood 0.0970.03 100 0 1.76 1.0Total 4.8370.26

720 h (30 d)

Heart 0.0170.001 100 0 1.90 0.2Liver 0.1970.03 83 17 54 4.7Kidney 0.1070.02 94 6 27.9 2.4Brain 0.0270.001 100 0 4.63 0.4Blood 0.0470.02 100 0 11.6 1.0Total 0.3670.01

1980 h (80 d)

Heart ND – – – –

Liver ND – – – –

Kidney 0.02 100 0 100 –

Brain ND – – – –

Blood ND – – – –

Total 0.02

ND: not detected.

M.F.H. Carneiro et al. / Environmental Research 134 (2014) 218–227 221

boy who died 5 d after receiving an infusion of plasma containinghigh content of TM (284–540 mg Hg). Considerable Hg levels werefound in the kidneys.

It has been assumed that the major effect from chronicexposure to inoHg is nephropathy induced mainly by apoptosisand necrosis (Sharma et al., 2007; Gewin et al., 2012). AlthoughTM is supposed to be decomposed rapidly into inoHg after TMinjection, the effects and mechanisms behind both etHg and inoHglevels should be deeply explored in this organ. Considering inoHgtoxicity upon kidneys, the literature shows that even at low levelsof exposure to inoHg (approximately four times higher than theHg concentration we administered in the present study) ratsexhibited an impaired glomerular filtration rate (Zalups, 1997).Moreover, the currently available literature reports renal failure,amongst other organ injuries, in a patient who ingested 83 mg/kgof TM (Pfab et al., 1996). However, little information on toxiceffects in kidneys after administration of TM-containing vaccinesis available. Moreover, to date, very little is known about themechanisms and molecular interactions underlying toxicity ofetHg/thimerosal in the kidney (Carneiro et al., 2013, 2014).

Due to an immature or impaired renal system, newborns andrenal patients, respectively, may be at high risk of toxicity after TMexposure through vaccines. Therefore, the kidney appears to be animportant post-TM target organ, despite the little attention thathas been given in the literature to the possible harmful renaleffects of Hg after TM exposure.

Table 1 also shows the fraction of the total Hg found as etHg orinoHg in each compartment analyzed. In this table, a considerablechange is evident in the proportion between inoHg and etHg inthe blood compartment from 0.5 to 1 h after TM exposure,indicating a rapid dealkylation of etHg in this fluid or transferenceof etHg from blood to tissues. However, no considerable changes inthe amount of etHg was observed in the tissues in the same timecourse, which indicates that etHg is rapidly converted to inoHg inblood compartment when it reaches the bloodstream. Moreover,during the time course of the exposure, the inorganic fractionincreases in all compartments, which confirms that the conversionof etHg to inoHg after TM exposure is a dynamic process.

Table 2 presents a comparison between Hg levels found inblood and tissues after TM exposure under different experimental

Fig. 1. Mean levels of inorganic mercury (inoHg) (circles) and ethyl Hg (etHg) (triangules)7standard deviation obtained in mice blood, heart, brain, kidney and liver afterintramuscular injection of 20 mg of Hg in the form of thimerosal (TM) (n¼4).

M.F.H. Carneiro et al. / Environmental Research 134 (2014) 218–227222

conditions and relatively similar exposure doses. As can be noted,in general Hg levels in blood and tissues of the present study waswithin the range of concentrations reported in other studies ondifferent animal species, with the exception of Harry et al. (2004).

3.2. Biological half-lives of etHg in blood and tissues of mice

As far as we know, this study is the first to estimate the half-lifeof TM in the blood, brain, kidney, liver and heart of mice after alow-dose exposure. Fig. 1 shows the mean levels of inoHg andetHg7SD obtained in the blood, heart, brain, kidney and liver ofmice at 0.5 h, 1 h, 8 h, 16 h, 6 d, 30 d and 80 d after the IMinjection of 20 mg of Hg as TM. These data were employed toestimate the biological half-lives in blood and tissues discussedthroughout the text.

Table 3 displays the estimated half-lives derived from one ortwo-compartment analysis of log-transformed Hg levels obtainedfrom TM exposed mice (see Section 2 for details). The time courseof Hg levels in blood, brain, heart and liver was better described bya one compartment model, whereas kidney levels are bestdescribed by a two compartment model with a rapid initial phasefollowed by a slower terminal phase of clearance.

The half-life of Hg in blood was estimated in 8.8 d, a valuevery close to the previous reports on those in blood obtainedfrom humans and other animal species exposed to TM (Table 4)but much shorter than those observed after meHg exposure(Skerfving, 1974; Yaginuma-Sakurai et al., 2012).

Burbacher et al. (2005) reported a terminal half-life of 8.62 d inthe blood of infant monkeys after TM injection, whereas Barregårdet al. (2011) and Pichichero et al. (2002, 2009) reported values of5.6 d, 7 d and 6.3 d, for adults and newborn infants. Therefore,although the biological interaction of Hg may differ betweenhumans and other animals (USEPA, 1997), our results on Hg half-life in blood of mice after TM exposure strongly indicate simila-rities between the two categories. This fact consistently stressesthe contribution of the present work to the establishment of atoxicological profile for TM. Moreover, since the clearance of Hgfrom blood after TM exposure is much faster than after meHgexposure (Burbacher et al., 2005), the latter is not a suitablereference for risk assessment of TM exposure. Moreover, given the

Table

2Com

parison

sof

Hgleve

lsin

bloo

dan

dtissues

afterTM

exposure

inex

perim

entalstudies.

Spec

iespro

file

Experim

entaldesigndose

ofHgas

TM(appro

ximate)

Route

of

exposu

reElap

sedtimeafter

exposu

rea

Hg

determined

Hgleve

ls(appro

ximate)

inmg/lorlg/g

Referen

ce

Blood

Brain

Hea

rtLive

rKidney

Adu

ltCD-1

mice(m

ale)

70mgHg/kg

(1.75mgHg)

IM24

hTo

tal

22.5

2.6

––

97Harry

etal.(20

04)

Neo

natal

ICRmice(both

gender)

1.4mgHg/kg

(35mg)

IM24

hinoH

g/orga

nic

fraction

0.8

0.04

/0.2

–0.9/2.3

0.7/2.5

Zareba

etal.(20

07)

Adu

ltSw

issmice(m

ale)

0.8mgHg/kg

(20mgHg)

IM8h

inoH

g/etHg

0.2/0.06

0.04

/0.04

2.25

/0.12

1.24

/4.08

21.15/10

.12

Presen

tstudy

16h

0.3/0.06

0.07

/0.05

2.1/0.32

1.04

/2.83

21.33/5.6

6d

0.08

/ND

0.03

/ND

0.33

/0.09

0.72

/0.35

10.7/0.9

Suck

lingW

istarrats

(male)

0.81

mmol

Hg/kg

(17mg

Hg),

3times

SC72

hTo

tal

0.85

0.10

8–

1.37

2Orctet

al.(20

06)

Suck

lingW

istarrats

(male)

0.81

mmol

Hg/kg

(17mg

Hg),

3times

SC24

hTo

tal

1.2

0.08

–1.5

1.7

Blanuša

etal.

(201

2)6d

0.5

0.08

1.2

2.1

Adu

ltW

istarrats

(male)

0.5mgHg/kg

(100mg

Hg)

Oral

12h/120

h:bloo

d(Total

Hg)

5d:

tissues

(inoH

g/etHg)

0.4/0.05

0.16

/0.035

0.84

/ND

3.0/0.5

9.5/0.38

Rod

rigu

eset

al.

(201

0)Adu

ltsq

uirrelmon

keys

(bothge

nder)

207mgHg—

low

dose

IN6mon

thsof

exposure

inoH

g/orga

nic

fraction

0.01

2/0.004

0.01

2/0.02

0–

0.07

3/0.007

0.41

/0.05

Blair

etal.(19

75)

Infantmon

keys

M.

fascicularis

(both

gender)

35mgHgdivided

in4ad

ministrations(birth,7,14

and21

)

IM7d

Total

0.006

––

––

Burbacher

etal.

(2005

)

aMostco

mmon

lyem

ploye

dtimepoints

amon

gthereferencescited(som

epreviou

sstudiespresentedother

timepoints).ND,n

otdetected.IM,intram

uscular.SC

,subc

utaneo

us.

IN,intran

asal.

Table 3Estimated half-lives for blood and tissues derived from one and two-compartmentanalysis of the log-transformed Hg levels obtained from thimerosal exposed mice.

Organ/tissue T1/2 (days)

Blood 8.8Heart 7.8Liver 7.7Brain 10.7

T1/2 α (days) T1/2 β (days)Kidney 2.7 45.2

Table 4Estimated blood half-lives after TM administration in different animal species.

Animal species Blood T1/2 (days) Reference

Mice 8.8 Present workMonkeys Initial: 2.1 Burbacher et al. (2005)

Terminal: 8.6

HumansAdults 5.6 Barregård et al. (2011)Newborns 7 Pichichero et al. (2002)Newborns 6.3 Pichichero et al. (2009)

M.F.H. Carneiro et al. / Environmental Research 134 (2014) 218–227 223

low Hg levels in blood and shorter half-life, a risk for toxicityseems to be lower than that after meHg exposure.

For the heart and liver, half-lives were estimated at 7.8 d and7.7 d (Table 3), versus 10.7 d for the brain, which is much shorterthan the 26 d half-life reported for brain after meHg exposure(Magos and Butler, 1976). On the other hand, the initial kidneyhalf-life was about 2.7 d and the terminal, 45.2 d (Table 3), anotherindicative that kidney is the organ where Hg persists for a longerduration after TM exposure. As a comparison, in rabbits exposed tomeHg, the total Hg half-life in liver and kidneys (0.125 mmol/kg,twice a week for 9 weeks, by intravenous injection) was estimatedat 28 d (Petersson et al., 1991). This period is about 4 times longerthan the Hg half-life we observed in the liver but lower than theHg half-life in the kidneys of TM-exposed mice. Accordingly, thekidney Hg half-life is closer to the 58 d (ranging from 35 to 90 d)reported after inhalation of Hg vapor (Clarkson and Magos, 2006).

3.3. Retention of initial Hg dose by each tissue along time course

Fig. 2 shows a schematic representation of Hg distribution afterTM exposure where the size of the organs represents a roughapproximation of the percentage of the initial dose (20 mg of Hg asTM) retained in each organ/tissue along a time course. To draw thefigure we used the amount of total Hg in mg in each tissue at eachtime point to calculate the percentage regarding the initial dose of20 mg Hg as TM as 100%.

Blood Hg levels did not exceed 7% of the initial dose of 20 mg ofHg throughout the time course. The maximum and minimumvalues were observed at 0.5 h and 30 d, respectively (6.63% and0.21%). In the heart, Hg levels peaked at 2.15% of the initial Hg dose1 h after the exposure then decreased during the follow-up until

reaching the minimum percentage observed as 0.03% of theinitial dose.

In the brain, the Hg percentage ranged from 0.82 at 0.5 h to0.08% at 30 d, whereas in the liver it reached 21.34% at 16 h,dropping to 0.96% at 30 d. Indeed, the percentage of initial doseretained in the kidney ranged from 0.11% at 80 d to about 69% at8 h after TM exposure, which corresponds to approximately 22 ngand 14 mg of Hg, respectively.

3.4. Evaluation of possible mechanisms associated withetHg conversion into inoHg in blood samples

With increasing size, the bond between Hg and the organicradical becomes less stable, which results in easier degradation ofetHg into inoHg. Accordingly, the bond lengths between Hg andcarbon in meHg and etHg have been estimated by Kaupp andMalkina (1998) as 2.115 and 2.150 Å, respectively. Besides this, inpreliminary experiments our group observed that etHg is highlystable at room temperature in plasma samples, but converts toinoHg in red blood cells (unpublished data). Thus, an in vitroexperiment was conducted to evaluate the stability of TM andetHg in whole blood, plasma and erythrocytes and also toelucidate the mechanism behind the dealkylation of etHg in blood.These experiments were carried out on human blood samples.

Fig. 3 shows Hg species concentrations (mean7SD) measuredafter adding etHg or TM solution (3 mg/l) in human samples ofwhole blood, plasma or red blood cells and incubated for 24 h.There was no difference between the levels of Hg species formedafter the addition of TM or etHg considering whole blood,erythrocytes or plasma individually, which indicates that the anionthiosalicylate presents undetectable interference in the dealkyla-tion of etHg present in TM (Fig. 3A). On the other hand, after the

Fig. 2. Schematic representation of mercury (Hg) distribution after thimerosal (TM) exposure. The size of the organs represents a rough approximation of the percentage ofthe initial dose retained in each of the organ/tissue along time.

M.F.H. Carneiro et al. / Environmental Research 134 (2014) 218–227224

incubation time, the etHg levels in whole blood were lower thanthose found both in erythrocytes and plasma for both compounds.Moreover, in whole blood, etHg is converted more intensivelywhen compared to red blood cells while no dealkylation wasobserved in plasma samples (Fig. 3A). This finding shows that thepartitioning of circulating etHg between red cells determines atleast in part the extension of etHg conversion to inoHg andmight contribute to augmenting the half-life of etHg in the animalbody.

There is some evidence that the dealkylation of organomercur-ials is triggered by reactive oxygen species and microbial organ-isms; however, the mechanisms behind the conversion are stillunknown (Suda et al., 1991, 1992; Qvarnström et al., 2003).Therefore, subsequent experiments were carried out to ascertainthe mechanisms associated with etHg conversion into inoHgin blood.

It is known that Fe is a cofactor of a Fenton Haber–Weissreaction and that plasma has a much lower Fe concentration incomparison to whole blood and red cells. Fig. 3B shows that theincubation of FeCl3 together with etHg resulted in a significantlyhigher inoHg concentration in plasma samples when compared toplasma without Fe addition. Consequently, greater conversion hasoccurred in the plasma possibly by the action of OHd radicalformed in the presence of a greater quantity of Fe. Furthermore,after Fe incubation, an increase in etHg conversion to inoHg wasobserved in erythrocytes (Fig. 3B) (Hg species levels were statis-tically the same in comparison to whole blood). Furthermore,erythrocytes and whole blood are known to present very similarFe concentrations (Helmer and Emerson, 1934; Fairbanks, 1994).Based on this information, one would expect that whole blood anderythrocytes would convert etHg to inoHg at similar rates. How-ever, in red blood cells, a lower conversion of etHg was observed incomparison to whole blood (Fig. 3A and B). This suggests involve-ment of other mechanisms in the dealkylation of etHg beyond theoxidizing action of OHd radical as previously reported by Suda andTakahashi (1992) and Suda et al. (1991). Potentially, other cells,molecules and enzymes in whole blood might be involved in thisprocess. For instance, polymorphonuclear leukocytes from humanblood and intraperitoneal cavity of rabbits, rats and guinea pigs, inaddition to guinea pig macrophages and eosinophils and humanmonocytes were already reported to participate in the conversion

of organic Hg into inoHg (Suda and Hirayama, 1992; Suda et al.,1992).

In order to confirm that the conversion of etHg into inoHgoccurs through OHd radical formation, a further experiment wasdeveloped with the addition of H2O2 and DMSO. The incubation ofplasma with H2O2 enhanced the conversion of etHg to inoHg asshown in Fig. 3C. The H2O2 is formed in the second step of aFenton reaction and serves as a substrate for OHd formation(Halliwell and Chirico, 1993). Later, H2O2 was added together withDMSO, which acts as a potential scavenger of OHd, and under thiscondition no statistical difference in terms of inoHg levels wasobserved when compared to the plasma samples without H2O2 orDMSO addition (Fig. 3C). The results strongly indicate that one ofthe mechanisms behind the conversion of etHg to inoHg in livingorganisms is mediated by OHd radicals. Moreover, few otherplausible explanations for the non-conversion of etHg in plasmainclude the absence of cells, hemoglobin and lesser content ofproteins and other compounds, which renders plasma a lessoxidative environment. This condition may also contribute to thehigher stability of etHg in plasma. However, further studies arenecessary to confirm this hypothesis.

Based on these findings and considering that etHg in plasma is“more freely available to interact with tissues” than those in redblood cells, we hypothesized that the etHg found in tissues istransported by plasma and not by red blood cells. Moreover, oncein plasma, etHg can be either sequestered by red blood cells andconverted into inoHg or be retained in tissues. etHg might be alsoconverted by oxidative reactions/specific enzymes within cells—something not yet investigated—or be returned to plasma/redblood cells. Hence, the partition coefficient of etHg in red bloodcells/plasma regulates the transport of this Hg species within theorganism, which is influenced primarily by the conversion rate ofetHg in red blood cells. This conversion can also have an impactupon the toxic effects of Hg after the administration of TM-containing vaccines.

4. Conclusion

The present study reinforces the evidence that the transport ofetHg from muscle to tissues and its conversion into inoHg are

Fig. 3. Mean7SD of mercury (Hg) species concentrations measured after 24 h of adding ethyl Hg (etHg) or thimerosal (TM) solution (3 mg/l) in whole blood, plasma or redblood cells obtained along the first (A), second (B) or third (C) experiments of the in vitro study. One way ANOVA followed by Tukey's test. nConcentrations of etHg and inoHgstatistically lower and higher, respectively, than those obtained in plasma and erythrocytes; #Concentrations of etHg and inoHg statistically lower and higher, respectively,than those obtained in plasma; nnConcentrations of etHg and inoHg statistically higher and lower, respectively, than those obtained in erythrocytes without Fe, and in bothblood groups; ##Concentrations of etHg and inoHg statistically lower and higher, respectively, than those obtained in plasma without Fe, Po0.05. nnnConcentrations of inoHgand etHg statistically higher and lower, respectively, than those obtained in plasmaþetHg and plasmaþetHgþH2O2þDMSO samples, Po0.05.

M.F.H. Carneiro et al. / Environmental Research 134 (2014) 218–227 225

rapid. Moreover, the dealkylation process occurs in red blood cells,but not in plasma, with hydroxyl radical as a potential effector.Thus, the partition coefficient of etHg in red blood cells andplasma potentially defines the amount of etHg that reaches tissuesafter TM exposure.

Although some tissues accumulated significant amounts of Hgafter TM exposure, the clearance time of etHg from the animalbody is shorter than that of meHg. Moreover, since etHg is quicklymetabolized to inoHg, the kinetics of the latter might determinethe elimination of Hg after TM exposure. Despite the smallercontribution of the brain to the body burden of Hg, the potentialneurotoxicity of TM must be continuously monitored, since any Hglevel in the brain, either organic or inorganic, should be viewed aspotentially hazardous. Additionally, the longer Hg persistence andhigher accumulation of both Hg species in kidneys make thetissue a potential candidate for toxic effects after TM exposure.Therefore, further studies must be carried out to evaluate thenephrotoxicity of TM.

Conflict of interest statement

The authors declare that they have no conflicts of interest.

Acknowledgments

The authors are grateful to the Fundação de Amparo à Pesquisado Estado de São Paulo (FAPESP-2011/08467-0) and the ConselhoNacional de Desenvolvimento Científico e Tecnológico (CNPq-147713/2010-2) for financial support and fellowships. We alsothank Samuel Simião de Souza, Kátia de Marco and Kim-Anh LeCao for technical support.

References

Arvidson, B., 1992. Accumulation of inorganic mercury in lower motoneurons ofmice. Neurotoxicology 13 (1), 277–280.

Aschner, M., Ceccatelli, S., 2010. Are neuropathological conditions relevant toethylmercury exposure? Neurotox. Res. 18 (1), 59–68.

Ball, L.K., Ball, R., Pratt, R.D., 2001. An assessment of thimerosal use in childhoodvaccines. Pediatrics 107 (5), 1147–1154.

Barregård, L., Rekic, D., Horvat, M., Elmberg, L., Lundh, T., Zachrisson, O., 2011.Toxicokinetics of mercury after long-term repeated exposure to thimerosal-containing vaccine. Toxicol. Sci. 120 (2), 499–506.

Berman, R.F., Pessah, I.N., Mouton, P.R., Mav, D., Harry, J., 2008. Low-level neonatalthimerosal exposure: further evaluation of altered neurotoxic potential in SJLmice. Toxicol. Sci. 101 (2), 294–309.

Berndt, W.O., Baggett, J.M., Blacker, A., Houser, M., 1985. Renal glutathione andmercury uptake by kidney. Fundam. Appl. Toxicol. 5 (5), 832–839.

Blair, A., Clark, B., Clarke, A.J., Wood, P., 1975. Tissue concentrations of mercury afterchronic dosing of squirrel monkeys with thiomersal. Toxicology 3 (2), 171–176.

Blanuša, M., Orct, T., Vihnanek Lazarus, M., Sekovanic, A., Piasek, M., 2012. Mercurydisposition in suckling rats: comparative assessment following parenteralexposure to thiomersal and mercuric chloride. J. Biomed. Biotechnol. 2012,256965.

Bonate, P.L., 2006. Pharmacokinetic—Pharmacodynamic Modeling and Simulation.Springer, New York.

Bose, R., Onishchenko, N., Edoff, K., Janson, L.A.M., Ceccatelli, S., 2012. Inheritedeffects of low-dose exposure to methylmercury in neural stem cells. Toxicol. Sci.130 (2), 383–390.

Burbacher, T.M., Shen, D.D., Liberato, N., Grant, K.S., Cernichiari, E., Clarkson, T.,2005. Comparison of blood and brain mercury levels in infant monkeysexposed to methylmercury or vaccines containing thimerosal. Environ. HealthPerspect. 113 (8), 1015–1021.

Carneiro, M.F.H., Morais, C., Barbosa Jr., F., Gobe, G., 2013. Thimerosal in childhoodvaccines contributes to accumulating mercury toxicity in the kidney. Toxicol.Environ. Chem. 95, 1424–1447.

Carneiro, M.F.H., Morais, C., Small, D.M., Vesey, D.A., Barbosa Jr., F., Gobe, G.C., 2014.Thimerosal induces apoptotic and fibrotic changes to kidney epithelial cellsin vitro. Environ. Toxicol. http://dx.doi.org/10.1002/tox.22012, in press.

Clarkson, T.W., 2002. The three modern faces of mercury. Environ. Health Perspect.110, 11–23.

Clarkson, T.W., Magos, L., 2006. The toxicology of mercury and its chemicalcompounds. Crit. Rev. Toxicol. 36, 609–662.

Clements, C.J., Ball, L.K., Ball, R., Pratt, D., 2000. Thiomersal in vaccines. Lancet 355(9211), 1279–1280.

Delong, G., 2011. A positive association found between autism prevalence andchildhood vaccination uptake across the U.S. population. J. Toxicol. Environ.Health A 74 (14), 903–916.

Fairbanks, V.F., 1994. Iron in medicine and nutritionModern Nutrition in Health andDisease8th ed. Lea & Febiger, Philadelphia, pp. 191–202.

Foss, S.D., 1969. A method for obtaining initial estimates of the parameters inexponential curve fitting. Biometrics 25, 580–584.

Gewin, L., Vadivelu, S., Neelisetty, S., Srichai, M.B., Paueksakon, P., Pozzi, A., Harris,R.C., Zent, R., 2012. Deleting the TGF-beta receptor attenuates acute proximaltubule injury. J. Am. Soc. Nephrol. 23 (12), 2001–2011.

Gibaldi, M., Perrier, D., 1982. Pharmacokinetics. Marcel Dekker, New York.Halliwell, B., Chirico, S., 1993. Lipid peroxidation: its mechanism, measurement,

and significance. Am. J. Clin. Nutr. 57 (5 Suppl), S715–S725.Harisa, G.I., Alanazi, F.K., El-Bassat, R.A., Malik, A., Abdallah, G.M., 2012. Protective

effect of pravastatin against mercury induced vascular cells damage:erythrocytes as surrogate markers. Environ. Toxicol. Pharmacol. 34 (2)428–435.

Harry, G.J., Harris, M.W., Burka, L.T., 2004. Mercury concentrations in brain andkidney following ethylmercury, methylmercury and thimerosal administrationto neonatal mice. Toxicol. Lett. 154 (3), 183–189.

Helmer, O.M., Emerson, C.P., 1934. The iron content of the whole blood of normalindividuals. J. Biol. Chem. 104 (1), 157–161.

Hilmy, M.I., Rahim, S.A., Abbas, A.H., 1976. Normal and lethal mercury levels inhuman beings. Toxicology 6, 155–159.

Ida-Eto, M., Oyabu, A., Ohkawara, T., Tashiro, Y., Narita, N., Narita, M., 2013. Prenatalexposure to organomercury, thimerosal, persistently impairs the serotonergicand dopaminergic systems in the rat brain: implications for association withdevelopmental disorders. Brain Dev. 35 (3), 261–264.

Kaupp, M., Malkina, O.L., 1998. Density-functional analysis of C-13 and H-1chemical-shifts and bonding in mercurimethanes and organomercuryhydrades—the role of scalar relativistic, spin-orbit, and substituent effects. J.Chem. Phys. 108, 3648–3659.

Lebel, J., Mergler, D., Branches, F., Lucotte, M., Amorim, M., Larribe, F., Dolbec, J.,1998. Neurotoxic effects of low-level methylmercury contamination in theAmazonian Basin. Environ. Res. 79 (1), 20–32.

Magos, L., 2003. Neurotoxic character of thimerosal and the allometric extrapola-tion of adult clearance half-time to infants. J. Appl. Toxicol. 23, 263–269.

Magos, L., Butler, W.H., 1976. The kinetics of methylmercury administered repeat-edly to rats. Arch. Toxicol. 35 (1), 25–39.

Orct, T., Blanusa, M., Lazarus, M., Varnai, V.M., Kostial, K., 2006. Comparison oforganic and inorganic mercury distribution in suckling rat. J. Appl. Toxicol. 26(6), 536–539.

Pamphlett, R., Jew, S.K., 2013. Uptake of inorganic mercury by human locus ceruleusand corticomotor neurons: implications for amyotrophic lateral sclerosis. ActaNeuropathol. Commun. 1, 13.

Pamphlett, R., Waley, P., 1996. Uptake of inorganic mercury by the human brain.Acta Neuropathol. 92, 525–527.

Petersson, K., Dock, L., Soderling, K., Vahter, M., 1991. Distribution of mercury inrabbits subchronically exposed to low levels of radiolabeled methyl mercury.Pharmacol. Toxicol. 68 (6), 464–468.

Petroni, D., Tsai, J., Agrawal, K., Mondal, D., George, W., 2012. Low-dose methyl-mercury-induced oxidative stress, cytotoxicity, and tau-hyperphosphorylationin human neuroblastoma (SH-SY5Y) cells. Environ. Toxicol. 27 (9),549–555.

Pfab, R., Muckter, H., Roider, G., Zilker, T., 1996. Clinical course of severe poisoningwith thiomersal. J. Toxicol. Clin. Toxicol. 34 (4), 453–460.

Pichichero, M.E., Cernichiari, E., Lopreiato, J., Treanor, J., 2002. Mercury concentra-tions and metabolism in infants receiving vaccines containing thiomersal:a descriptive study. Lancet 360, 1737–1741.

Pichichero, M.E., Gentile, A., Giglio, N., Alonso, M.M., Fernandez Mentaberri, M.V.,Zareba, G., Clarkson, T., Gotelli, C., Gotelli, M., Yan, L., Treanor, J., 2009. Mercurylevels in premature and low birth weight newborn infants after receipt ofthimerosal-containing vaccines. J. Pediatr. 155 (4), 495–499.

Qvarnström, J., Lambertsson, L., Havarinasab, S., Hultman, P., Frech, W., 2003.Determination of methylmercury, ethylmercury, and inorganic mercury inmouse tissues, following administration of thimerosal, by species-specificisotope dilution GC-inductively coupled plasma-MS. Anal. Chem. 75 (16),4120–4124.

Rodrigues, J.L., Serpeloni, J.M., Batista, B.L., Souza, S.S., Barbosa Jr., F., 2010.Identification and distribution of mercury species in rat tissues followingadministration of thimerosal or methylmercury. Arch. Toxicol. 84 (11),891–896.

Sharma, M.K., Sharma, A., Kumar, A., Kumar, M., 2007. Evaluation of protectiveefficacy of Spirulina fusiformis against mercury induced nephrotoxicity inSwiss albino mice. Food Chem. Toxicol. 45 (6), 879–887.

Skerfving, S., 1974. Methylmercury exposure, mercury levels in blood and hair, andhealth status in Swedes consuming contaminated fish. Toxicology 2 (1), 3–23.

Souza, S.S., Campiglia, A.D., Barbosa Jr., F., 2013. A simple method for methylmer-cury, inorganic mercury and ethylmercury determination in plasma samples byhigh performance liquid chromatography-cold-vapor-inductively coupledplasma mass spectrometry. Anal. Chim. Acta 761, 11–17.

Stajich, G.V., Lopez, G.P., Harry, S.W., Sexson, W.R., 2000. Iatrogenic exposure tomercury after hepatitis B vaccination in preterm infants. J. Pediatr. 136,679–681.

M.F.H. Carneiro et al. / Environmental Research 134 (2014) 218–227226

Suda, I., Hirayama, K., 1992. Degradation of methyl and ethyl mercury intoinorganic mercury by hydroxyl radical produced from rat-liver microssomes.Arch. Toxicol. 66 (6), 398–402.

Suda, I., Takahashi, H., 1992. Degradation of methyl and ethyl mercury intoinorganic mercury by other reactive oxygen species besides hydroxyl radical.Arch. Toxicol. 66 (1), 34–39.

Suda, I., Totoki, S., Takahashi, H., 1991. Degradation of methyl and ethyl mercuryinto inorganic mercury by oxygen free radical-producing systems: involvementof hydroxyl radical. Arch. Toxicol. 65 (2), 129–134.

Suda, I., Totoki, S., Uchida, T., Takahashi, H., 1992. Degradation of methyl and ethylmercury into inorganic mercury by various phagocytic cells. Arch. Toxicol. 66(1), 40–44.

Suzuki, T., Takemoto, T.-I., Kashiwazaki, H., Miyama, T., 1973. Metabolic fate ofethylmercury salts in man and animal. In: Miller, MW, Clarkson, TW (Eds.),Mercury, Mercurials and Mercaptans. Charles C. Thomas, Springfield, IL,pp. 209–232.

Tan, M., Parkin, J.E., 2000. Route of decomposition of thiomersal (thimerosal). Int. J.Pharm. 208 (1–2), 23–34.

USEPA (United States Environmental Protection Agency), 1997. Mercury StudyReport to Congress (Available from). Office of Air Quality Planning andStandards and Office of Research and Development, Washington D.C., USA(accessed October, 2013).

Vahter, M., Mottet, N.K., Friberg, L.T., Lind, S.B., Charleston, J.S., Burbacher, T.M.,1995. Demethylation of methyl mercury in different brain sites of Macaca

fascicularis monkeys during long-term subclinical methyl mercury exposure.Toxicol. Appl. Pharmacol. 134 (2), 273–284.

Wolfsegger, M.J., 2006. The R package PK for basic pharmacokinetics. Biometrie undMedizin 5, 61–68.

WHO (World Health Organization), 2012. Children's exposure to mercury com-pounds. Available from ⟨http://www.who.int/ceh/publications/children_exposure/en/⟩ (accessed March, 2014).

Yaginuma-Sakurai, K., Murata, K., Iwai-Shimada, M., Nakai, K., Kurokawa, N.,Tatsuta, N., Satoh, H., 2012. Hair-to-blood ratio and biological half-life ofmercury: experimental study of methylmercury exposure through fish con-sumption in humans. J. Toxicol. Sci. 37 (1), 123–130.

Yasutake, A., Marumoto, M., Yoshida, M., 2010. Neurotoxic action of inorganicmercury injected in the intraventricular space of mouse cerebrum. J. Toxicol.Sci. 35, 767–771.

Zalups, R.K., 1997. Reductions in renal mass and the nephropathy induced bymercury. Toxicol. Appl. Pharmacol. 143, 366–379.

Zareba, G., Cernichiari, E., Hojo, R., Nitt, S.M., Weiss, B., Mumtaz, M.M., Jones, D.E.,Clarkson, T.W., 2007. Thimerosal distribution and metabolism in neonatal mice:comparison with methyl mercury. J. Appl. Toxicol. 27 (5), 511–518.

Zimmermann, L.T., Santos, D.B., Naime, A.A., Leal, R.B., Dórea, J.G., Barbosa Jr., F.,Aschner, M., Rocha, J.B., Farina, M., 2013. Comparative study on methyl- andethylmercury-induced toxicity in C6 glioma cells and the potential role of LAT-1in mediating mercurial-thiol complexes uptake. Neurotoxicology 38C, 1–8.

M.F.H. Carneiro et al. / Environmental Research 134 (2014) 218–227 227