Journal of Inorganic Biochemistry 105 (2011) 1464–1468

Contents lists available at SciVerse ScienceDirect

Journal of Inorganic Biochemistry

j ourna l homepage: www.e lsev ie r.com/ locate / j inorgb io

Aluminium effects on thyroid gland function: Iodide uptake, hormone biosynthesisand secretion

Daniel Orihuela ⁎Laboratorio de Investigaciones Fisiológicas Experimentales, Facultad de Bioquímica y Ciencias Biológicas, Universidad Nacional del Litoral, Santa Fe, Argentina

⁎ Corresponding author at: Cátedra de Fisiología HumCiencias Biológicas, Universidad Nacional del Litoral, PisoEl Pozo, (3000) Santa Fe, Argentina. Fax: +54 342 4595

E-mail address: orihuela@fbcb.unl.edu.ar.

0162-0134/$ – see front matter © 2011 Elsevier Inc. Aldoi:10.1016/j.jinorgbio.2011.08.004

a b s t r a c t

a r t i c l e i n f o

Article history:Received 20 April 2011Received in revised form 5 August 2011Accepted 5 August 2011Available online 12 August 2011

Keywords:AluminiumRadiolabeled iodide 125I−

Sodium–iodide symporterThyroxineThyrotropinRat

The effects of aluminium (Al) on thyroid function were evaluated in adult Wistar rats intraperitoneally (i.p)injected with 7 mg Al (as lactate)/kg body weight (b.w) per day during a six week period. The time-coursekinetics of Na125I (3 μCi per 100 g b.w, i.p) was analysed bymeasuring gamma-radioactivity of thyroid, serum,serum protein precipitate and bile, at times ranging from 2 to 96 h post-dosing. In Al-treated group the 125I−

thyroid uptake at 24 h (15,840±570 vs. 18,030±630 dpm/mg, Pb0.05) as well as the rate of 125I− releasefrom the gland, calculated as the slope of the plot between 24 and 96 h (84±8 vs. 129±11 dpm/mg/h,Pb0.05) were significantly reduced as compared to control. The biliary 125I− excretion was not modified at allstudied times. The Al content and lipid peroxidation (69.1±8.5 vs. 53.2±7.0 nmol MDA/g wet weight,Pb0.05) of thyroid tissue were increased in Al-treated rats. The serum concentrations of total thyroxine (T4,3.78±0.14 vs. 4.68±0.12 μg/dL, Pb0.05) and total triiodothyronine (T3, 47±4 vs. 66±5 ng/dL, Pb0.05)were decreased by effect of Al, but free-T4 (1.05±0.05 vs. 1.04±0.04 ng/dL, NS) and thyrotropin (TSH, 2.7±0.4 vs. 2.6±0.5 ng/ml, NS) remain unchanged. In spite of the Al could indirectly affect thyroid iodide uptakeand hormones secretion by a mechanism involving the induction of an oxidative stress state, however, thesechanges could be managed by the hypothalamus–pituitary–thyroid endocrine axis. We can conclude that inadult rats the Al would not act as a thyroid disruptor.

ana, Facultad de Bioquímica y4, Ciudad Universitaria, Paraje522.

l rights reserved.

© 2011 Elsevier Inc. All rights reserved.

1. Introduction

Aluminium (Al) is a ubiquitous element widely spread in theenvironment. The expanding Al use in industrial and householdapplications, has rendered humans highly exposed to its toxicity. Alcan accumulate in vulnerable regions of the brain, and has been stronglylinked to a number of neurodegenerative diseases (dialysis encepha-lopathy, amyotrophic lateral sclerosis, Parkinson's and Alzheimer'sdiseases) [1].

A role for Al as disruptor of certain endocrine systems has recentlyemerged. For instance, it has been well established that Al can beclassified as a metalloestrogen [2,3].

It has been reported that in a group of 227 subjects withoccupational exposure to aluminium, thyroid stimulating hormone(thyrotropin, TSH) values were reduced after 1 year of work and werestill reduced 6 months later [4]. This finding suggests that Al may alterthe pituitary endocrine regulation of thyroid gland.

The iodide-containing thyroid hormones T3 and its precursor T4are crucial for normal development, growth, and regulation ofnumerous metabolic pathways. The synthesis of thyroid hormones

(TH) requires uptake of iodide across the basolateral membrane intothe thyrocytes, transport across the cell, and efflux through the apicalmembrane into the follicular lumen. Uptake of iodide is mediated bythe sodium–iodide symporter (NIS), which cotransports two sodiumions along with one iodide ion, with the sodium gradient serving asthe driving force. The efflux of iodide across the apical membrane ismediated, at least in part, by pendrin. Once iodide reaches the cell–colloid interface, it is oxidised and rapidly organified by incorporationinto selected tyrosyl residues of thyroglobulin. This reaction, referredto as organification, is catalysed by thyroid peroxidase (TPO) in thepresence of hydrogen peroxide and results in the formation of mono-and diiodotyrosines (MIT and DIT). TPO also catalyses the coupling oftwo iodotyrosines to form either T3 or T4. To release thyroidhormones, thyroglobulin is engulfed by pinocytosis, digested inlysosomes, and then secreted into the bloodstream at the basolateralmembrane [5,6]. In serum, most of T3 and T4 (N99%) are bound tospecific carrier proteins, namely, thyroxine binding globulin (TBG),transthyretin (TTR) and albumin [7,8].

By tracing the literature about animal models used to investigateeffects of Al on thyroidmetabolism and function, scarce information isfound. An old study carried out with aluminium-fed and iodine-fedanimals (pigs and rats) concluded that aluminium occurs at only avery low level in the thyroid and that there is no correlation betweenaluminium and iodine in normal metabolism [9]. Aluminiumadministered intraperitoneally enhanced the permeability of the

1465D. Orihuela / Journal of Inorganic Biochemistry 105 (2011) 1464–1468

blood–brain barrier to labelled thyroxine, a substance which crossesthe blood–brain barrier by carrier-mediated transport, but not tolabelled TSH or iodide [10]. After a single i.v aluminium infusion torabbits, Al concentration did not significantly increase above controlin the thyroid gland 4 h after the dose [11]. Short-term oral high dosealuminium treatment reduced thyroxine serum levels in euthyroidrats [12].

Thus, the goal of this work was to analyse the possible effects ofaluminium on thyroid function involving radiolabeled iodide uptake,hormone biosynthesis and release, as well as the circulating levels ofmain thyroid-related hormones in a rodent model of mid-termsystemic administration.

2. Materials and methods

2.1. Chemicals

Na[125I] (specific activity 17 mCi/mg) was obtained from NENPerkinElmer (Boston, USA). All chemicals were of analytical gradefrom Sigma Co. (St. Louis, USA).

Al lactate to be injected was dissolved in 0.9% saline and weeklyprepared.

2.2. Animals and treatment

Five-month old male rats of a Wistar strain inbred in ourlaboratory were used (mean weight 475±22 g). Animals werehoused at controlled room temperature (22±3 °C) on a 12 hlight/12 h dark cycle. Standard commercial rodent diet was suppliedto all animals and tap water was provided ad libitum. Theexperimental procedures were approved by Ethical Committee ofthe Faculty and conducted in accordance with international rules onthe use of animals in toxicological studies [13].

The rats were randomly split in two groups: 1) Aluminium-treated: they received intraperitoneally (i.p) during a period of sixweeks 7 mg Al (as lactate)/kg body weight per day (0.5 ml/injection).2) Control: they were given daily i.p injection of 0.5 ml of normalsaline during the same period. This Al dose and administration routehas been repeatedly used to achieve systemic Al overload in rats [14].The period of treatment with Al was chosen on the basis of a previouspilot study, causing in our rats a significant decrease of total T4 serumlevel.

2.3. Radiolabeled iodide test

At the end of treatment period, a tracer dose of 3 μCi Na125I per100 g body weight was injected i.p into each rat, dissolved in 1 ml ofnormal saline. Time-course kinetics was analysed by measuringgamma-radioactivity of thyroid, serum, serum protein precipitateand bile, at different times: 2, 4, 6, 12, 24, 48, 72 and 96 h postradiolabeled iodide dosing [15]. At least 2 rats of each experimentalgroup were included at each time. Animals were anaesthetised withsodium pentobarbital (50 mg/kg b.w, i.p) and samples were obtainedas follows:

Thyroid: glands were immediately removed, cleaned, weighed andthen transferred into Pyrex counting tubes with 1 ml of 5 N NaOH.Upon completion of digestion, usually within 2 h at room tempera-ture, the radioactivity in the thyroid gland preparations wasdetermined [16]. Results were expressed as disintegrations/min(dpm) per mg of tissue.

Serum and protein precipitate: blood was drawn from the heartthrough a syringe, after clotting the serum was obtained by centrifu-gation.Analiquot of 1 ml serumwasmixedwith anequal amountof 30%trichloroacetic acid. The precipitate obtained by centrifugation wassuccessively washed three times with 5% trichloroacetic acid [16]. Theradioactivities of 1 ml serum and trichloroacetic acid precipitate

(protein-bound 125I, PBI125) were counted. Results were expressed asdpm/ml.

Bile: in some animals, before performing the heart puncture, asample of bile was collected during a 15 min period by means of acatheter (size PC-10) introduced into the bile duct, and radioactivitywas counted. Results were expressed as dpm/ml/min.

This sample was included to evaluate the catabolic pathway ofthyroid hormones which occurs mainly by hepatic glucuronidationand sulfation, products that suffer biliary excretion [8].

2.4. Hormones serum levels quantitation

In another group of rats subjected to the same protocol asindicated above (item 2.2), when the experimental period ended, ablood sample was withdrawn by heart puncture under pentobarbitalanaesthesia and sera were frozen at −20 °C until hormonalmeasurements were performed. Total thyroxine and triiodothyronine,as well as the free-thyroxine concentrations were determined byradioimmunoassay (Institute of Isotopes Ltd., Budapest, Hungary). Ratthyrotropin (rTSH) in serum was also determined by radioimmuno-assay (Immunodiagnostic Systems Ltd., Liege, Belgium).

2.5. Thyroid tissue lipid peroxidation

Since it is known that Al can act as a pro-oxidant agent in severalorgans [17], an index of oxidative stress of thyroid tissuewasmeasured.Lipid peroxidation levels were determined as thiobarbituric acidreactive substances (TBARS) in the homogenates of thyroid glands(10% w/w, 0.15 M KCl), employing 1,1,3,3-tetramethoxypropane asexternal standard, according to Ohkawa et al. [18]. Results wereexpressed as nmol malondialdehyde (MDA) produced/g wet weight.

2.6. Aluminium measurements

In samples of sera and thyroid tissues the Al content were analysedby electrothermal atomic absorption spectrometry. Both lobes of ratthyroid were dissected, weighted and digested in 0.3 ml of pure nitricacid [19]. The Al concentrations were determined by atomicabsorption spectroscopy employing a Perkin Elmer model 5000apparatus equipped with graphite furnace model HGA-500 (USA).

2.7. Radioactivity measurements

Iodine-125 is a radionuclide with a half-life of 59.4 days, emittinggamma-rays with a maximum energy of 35 keV. Radioactivities of thesamples were measured in a single-well scintillation spectrometer(AlfaNuclear gammacounter, Argentina). Disintegrations per minutewere calculated using a Na[125I] standard solution of known absoluteactivity. Counting time for each sample was set to maintain statisticalerror below 1%. Corrections were made for background and isotopicdecay.

2.8. Statistical analysis

The results in the text are presented as mean±S.E.M. Statisticalsignificance of the differences between two groups was assessed bygrouped Student's t-test. Linear regression analysis was performed bythe least squares method. P values of 0.05 or less were consideredsignificant. Data were processed with a registered version of theInStat/Prism 3.03 software (GraphPad Software Inc., San Diego, CA,USA).

Table 1Serum levels of thyroid-related hormones.

Experimental group Total T4μg/dL

Free T4ng/dL

Total T3ng/dL

TSHng/ml

Control 4.68±0.12 1.04±0.04 66±5 2.6±0.5Al-treated 3.78±0.14⁎ 1.05±0.05 47±4⁎ 2.7±0.4

Data are mean±S.E.M. (n=10).Al = aluminium; T3 = triiodothyronine; T4 = thyroxine ; TSH = thyroid stimulatinghormone.⁎ Pb0.05 vs. control.

1466 D. Orihuela / Journal of Inorganic Biochemistry 105 (2011) 1464–1468

3. Results

3.1. Radiolabeled iodide test

As shown in Fig. 1 the peak of radioactivity of 125I− in thyroid gland,observedat 24 h,was significantly lower inAl-treatedgroupas comparedto control (15,840±570 vs. 18,030±630 dpm/mg, Pb0.05). The rate ofrelease of 125I− from the gland, associated with the hormone secretionprocess, was calculated by fitting a regression line among values of theplot from 24 to 96 h. A significant decrease of the slope was obtained inAl-treated group as compared to control (84±8 vs. 129±11 dpm/mg/h,Pb0.05). ThePBI125 values at 24 and48 hwere significantly reduced afterAl treatment with regard to control (Fig. 1C). In Al-treated rats, neitherserum nor bile radioactivities were modified, at all studied times, whencompared to respective controls.

3.2. Hormones serum levels

The seric concentrations of thyroid-related hormones are displayedin Table 1. Total T3 and T4 were decreased, but free-T4 and TSHconcentrations remained unaltered, as a consequence of Al treatment.The ratio total T4/T3, calculated with the concentrations expressed inthe same units, was not modified by Al as compared to control (80.7±9.2 vs. 80.6±5.9, n=10, P=0.999, NS).

3.3. Aluminium measurements

As can be seen in Table 2, the Al serum concentration, as well as theAl content of thyroid tissue in Al-treated group were higher thancontrols.

3.4. Thyroid tissue lipid peroxidation

Fig. 2 depicts the TBARS value of thyroid tissue of rats. The Altreatment produced an increase of this parameter as compared tocontrol (69.1±8.5 vs. 53.2±7.0 nmol MDA/g wet weight, Pb0.05).

Fig. 1. Effect of aluminium on time-course of radiolabeled iodide (125I−). For each time data abound. Panel D: biliary excretion.*Pb0.05 vs. respective time of Control.

4. Discussion

The purpose of the present work was to elucidate whether Al can beconsidered as a thyroid disruptor in adult rats. Referred to animalmodels, the basic areas of interference of chemicals with thyroidmetabolism are: inhibition of iodide uptake at the cellular membrane ofthe thyrocyte via blockage of theNIS transporter; synthesis inhibition viathyroperoxidase; binding of transport protein TTR in the bloodstream;altered hepatic phase II catabolism by glucuronosyltransferase andsulfotransferase metabolism of T3 and T4; alteration of deiodinase-regulated T4 metabolism; and alterations of transport across cellularmembranes and alteration of TH receptors in target cells [8,20]. Inparticular for Al, some of the major mechanisms were analysed here.

The main finding was that free-T4 and TSH circulating levels, themost sensitive markers of thyroid function in mammals [21], were notaffected by Al, regardless of the observed effects on iodide uptake.According to our results the accumulation of radiolabeled iodide bythyroid was diminished in Al-treated rats (Fig. 1A). Despite the Alcannot act as a direct competitor of iodine [6], even so, could indirectlyalter the function of the sodium-iodide symporter (NIS). Thistransport system moves iodine ions from bloodstream into theepithelial cell of thyroid follicles, and works synchronically withNa/K-ATPase which expels out the two Na ions carried along withiodine, in order to maintain the electrochemical balance of the cell. It

remean±S.E.M. (n=3). Panel A: thyroid gland. Panel B: serum. Panel C: serum protein-

Table 2Levels of aluminium in serum and thyroid tissue of rats.

Experimental group Serum(μg/L)

Thyroid gland(ng/mg)

Control 0.07±0.02 BDLAl-treated 320±18⁎ (1.15±0.06)×10−3⁎

Data are mean±S.E.M. (n=4).BDL = below detection limit.⁎ Pb0.05 vs. control.

1467D. Orihuela / Journal of Inorganic Biochemistry 105 (2011) 1464–1468

has been demonstrated that Al inhibits Na/K-ATPase in brain cortexsynaptosomes and kidney homogenates of rats exposed to Al.Inhibition of Na/K-ATPase activity occurred, preceding possiblealterations of expression of catalytic subunits, cellular energydepletion, and disturbances in cellular membrane integrity. Thedecrease in total Na/K-ATPase activity was ensured by partialinhibition of isozymes containing α1-, α 2-, and α 3-subunits[22,23]. Since the energy required to produce the Na gradient,which serves as the driving force to transport iodine ions by NIS, isprovided by Na/K-ATPase [5], a failure of this enzyme could lead todiminished iodine uptake by thyrocytes. Besides, the increase ofthyroid TBARS level in Al-treated group (Fig. 2) would be indicating acertain degree of oxidative stress of thyroid tissue, which maycontribute to impair NIS activity because of the cell membranedisarrangement implicated [17,24]. Although a small amount of Alwas found in thyroid tissue of Al-treated group, it is not possible toknow its internal distribution: inside or outside the follicles (theparafollicular tissue). Even the latter, Al could still exert its action onthe basal membrane of thyrocytes.

The lower availability of iodine within thyrocytes may explain byitself the falling of total T3 and T4 serum levels, though an effect onhormone secretion through basal membrane into bloodstream, as issuggested by the decreased slope of 125I− releases from thyroid glandin Al-treated rats, cannot be ruled out. Nevertheless, the mechanismsof TH secretion at the basolateral membrane and the involved channel(s) have not been characterised [5].

However, the overall impact of this reduction in total THcirculating levels produced by Al on the endocrine regulatory systemof thyroid, seems to be unimportant, since TSH level remainedunaffected.

Circulating levels of THs are maintained within a relatively narrowrange in large part by a negative feedback relationship betweencirculating levels of THs and those of TSH [7,25]. THs appear to exertthis effect by acting on one subtype of the TH receptor (TRβ2), whichis expressed in the pituitary gland and in the hypothalamicparaventricular nucleus, and appears to be the predominant mediatorof the negative feedback action of TH on TSH. Thus, environmentalchemicals that interact with TRs will affect this negative feedbacksystem if, and only if, they interact with the TRβ2 isoform [26]. Thereis no evidence that Al is able to perform such an interaction.

Fig. 2. Level of lipid peroxidation, expressed as TBARS, of thyroid tissue of rats treatedwith Al.*Pb0.05 vs. Control. Data are mean±S.E.M. (n=4).

An important issue to take into account to explain this lackof response of the hypothalamic–pituitary–thyroid (HPT) axis inAl-treated rats is that although circulating levels of THs are generallymaintained within a narrow range, the range is far narrower for anindividual than for the population, and individual genetics is a majorcontributor defining the set-point around which the HPT axis isregulated [26]. Thereby, the observed variability in measures of totalTHs between Al-treated and control groups would be relativized.

The scenario deployed by our outcomes is a reduction of total THswith no variation of free-T4 and TSH serum levels in Al-treated group.To find a possible explanation we need to analyse the factors thatregulate the serum half-life of THs.

The long serum half-life of T4, when compared to other hormones, islargely the result of tightnon-covalent binding to three principal bindingproteins: TBG, TTR and albumin [7,8]. The bound T4 is in rapidequilibriumwith unbound or free T4 that is available for cellular uptake.Because T4 is more avidly bound to these proteins, it has a much longerhalf-life than T3. Chemicals that can displace TH from these bindingproteins may cause a very rapid decline (in a timeframe of minutes) inserum hormone levels (both bound and free), in combinationwith theirability to increase the biliary clearance by inducing phase I–III enzymes[27]. Theprincipal pathwayof THclearance fromserumisby conjugationto glucuronic acid or sulphate [8]. The constitutive androstane receptorand the pregnane X-receptor play central roles in the xenobiotic-induced clearance of T3 and T4 and the subsequent ability ofenvironmental chemicals to reduce serum concentrations of thesehormones [26]. In the case of Al, it seems unlikely an effect ofdisplacement of THs from their binding proteins, since the reduction oftotal THs was not so remarkable and, in addition, free-T4 was notchanged; related to this, the rate of biliary excretion of 125I−, used as anindex of hepatic catabolism of THs, was not modified by Al. Then, the Alwould not substantially alter the half-life of THs.

Moreover, there are physiological factors, not yet well-known, thatit would help to understand in part these results. According to anidealised model of the HPT axis, if a xenobiotic chemical causes adecrease in circulating levels of T4, then serum TSH should increase[7,26]. However, there are a number of chemicals that can cause adecrease in serum total and free T4 without causing a concomitantincrease in serum TSH. One of the best-known examples is that ofpolychlorinated biphenyls [27,20]. An explanation for this behaviourcomes from a critical issue of the functioning of HPT axis calledcompensation: this concept posits that as THs levels decline, a numberof systemic and tissue-level responses are activated which amelioratethe consequences of low TH [26].

Another important topic must be addressed. T4 is mainly aprohormone that becomes activated upon its conversion to T3,considered the bioactive form of TH [7,25]. Nuclear bound T3 intarget cells is partly derived from plasma and partly from localgeneration as a result of the deiodination of T4. This task is performedby three different isoforms of deiodinase enzymes (DIO; type I, II andIII), which are selenocysteine-containing proteins, responsible for theproduction and recycling of T3 and T4 inside specific types of cells[28]. The ratio between T4 and T3 total serum concentrations can beused as an indirect marker of the DIO activity [21]. Even though theconcentrations of total THs diminished in serum, the T4/T3 ratio didnot vary, thus, we can reasonably assume that DIO activity was notaltered by Al treatment.

In summary, despite the Al could indirectly affect thyroid iodideuptake and hormones secretion by a mechanism involving theinduction of an oxidative stress state that would alter the functionof basal membrane of thyrocytes, however, the magnitude of thesechanges would not be so relevant, and they could be managed by thehypothalamus–pituitary–thyroid endocrine axis as is pointed out bythe absence of variation of TSH and the biologically active free fractionof T4, the standard measures of thyroid function. We can concludethat in adult rats the Al would not act as a thyroid disruptor.

1468 D. Orihuela / Journal of Inorganic Biochemistry 105 (2011) 1464–1468

5. Abbreviations

NIS Sodium-iodide symporterT3 TriiodothyronineT4 ThyroxineTSH Thyroid stimulating hormone, ThyrotropinTBARS Thiobarbituric acid reactive substancesTH Thyroid hormonesTPO Thyroid peroxidaseMIT MonoiodotyrosinesDIT DiiodotyrosinesTBG Thyroxine binding globulinTTR TransthyretinHPT Hypothalamic–pituitary–thyroid axis

Conflict of interest statement

No conflict of interest has to be declared by authors in relationwith this work.

Acknowledgements

This research was supported by grant C.A.I+D 2009-12/B354-PROG 069, Universidad Nacional del Litoral, Argentina.

References

[1] C.M. Nday, B.D. Drever, T. Salifoglou, B. Platt, J. Inorg. Biochem. 104 (2010)919–927.

[2] P.D. Darbre, J. Inorg. Biochem. 99 (2005) 1912–1919.[3] P.D. Darbre, J. Appl. Toxicol. 26 (2006) 191–197.

[4] L. Alessio, P. Apostoli, A. Ferioli, I. Di Sipio, I. Mussi, C. Rigosa, A. Albertini, Med. Lav.80 (1989) 290–300.

[5] A. Bizhanova, P. Kopp, Endocrinology 150 (2009) 1084–1090.[6] O. Dohan, A. De La Vieja, V. Paroder, C. Riedel, M. Artani, M. Reed, C.S. Ginter, N.

Carrasco, Endocr. Rev. 24 (2003) 48–77.[7] A.W. Norman, Thyroid hormones, in: A.W. Norman, G. Litwack (Eds.), Hormones,

Academic Press, New York, 1997, pp. 180–181.[8] L. Patrick, Altern. Med. Rev. 14 (2009) 326–346.[9] K. Mackenzie, Biochem. J. 26 (1932) 833–845.

[10] W.A. Banks, A.J. Kastin, Neuropharmacology 24 (1985) 407–412.[11] R.A. Yokel, P.J. McNamara, Toxicol. Appl. Pharmacol. 99 (1989) 133–138.[12] D. Orihuela, J. Inorg. Biochem. 103 (2009) 1542–1547.[13] Guiding Principles in the Use of Animals in Toxicology revised, Society of

Toxicology, Virginia, USA, March 1999.[14] J.L. Esparza, M. Gomez, M.R. Nogues, J.L. Paternain, J. Mallol, J.L. Domingo, J. Pineal

Res. 39 (2005) 129–136.[15] E.A. Merrill, R.A. Clewell, J.M. Gearhart, P.J. Robinson, T.R. Sterner, K.O. Yu, D.R.

Mattie, J.W. Fisher, Toxicol. Sci. 73 (2003) 256–269.[16] J.M. Hsu, A.W. Root, G.E. Duckett, J.C. Smith, A.A. Yunice, G. Kepford, J. Nutr. 114

(1984) 1510–1517.[17] C. Exley, Free Radic. Biol. Med. 36 (2004) 380–387.[18] H. Ohkawa, N. Ohishi, K. Yagi, Anal. Biochem. 95 (1979) 351–358.[19] D. Bohrer, M.B. Dessuy, R. Kaizer, P.C. do Nascimento, M.R. Schetinger, V.M.

Morsch, L.M. de Carvalho, S.C. Garcia, Anal. Biochem. 377 (2008) 120–127.[20] M. Boas, U. Feldt-Rasmussen, N.E. Skakkebæk, K.M. Main, Eur. J. Endocrinol. 154

(2006) 599–611.[21] M. DeVito, L. Biegel, A. Brouwer, S. Brown, F. Brucker-Davis, A. Oliver Cheek, R.

Christensen, T. Colborn, P. Cooke, J. Crissman, K. Crofton, D. Doerge, E. Gray, P.Hauser, P. Hurley, M. Kohn, J. Lazar, S. McMaster, M. McClain, E. McConnell, C.Meier, R. Miller, J. Tietge, R. Tyl, Environ. Health Perspect. 107 (1999) 407–415.

[22] V.S. Silva, A.I. Duarte, A.C. Rego, C.R. Oliveira, P.P. Goncalves, Toxicol. Sci. 88 (2005)485–494.

[23] V.S. Silva, P.P. Goncalves, J. Inorg. Biochem. 97 (2003) 143–150.[24] M. Suwalsky, B. Norris, F. Villena, F. Cuevas, P. Sotomayor, P. Zatta, Food Chem.

Toxicol. 42 (2004) 925–933.[25] Y. Shi, J.W.A. Ritchie, P.M. Taylor, Pharmacol. Ther. 94 (2002) 235–251.[26] T.R. Zoeller, Hormones 9 (2010) 28–40.[27] E.N. Pearce, L.E. Braverman, Best Pract. Res. Clin. Endocrinol. Metab. 23 (2009)

801–813.[28] J. Kohrle, Best Pract. Res. Clin. Endocrinol. Metab. 21 (2007) 173–191.