- Email: [email protected]
Effects of Methylmercury on the Microvasculature of the Developing Brain
NeuroToxicology 25 (2004) 849–857
Effects of Methylmercury on the Microvasculature of the Developing Brain M. Bertossi1,*, F. Girolamo2, M. Errede2, D. Virgintino2, G. Elia3, L. Ambrosi4, L. Roncali2 1
Dipartimento di Scienze Biomediche, Facolta` di Medicina e Chirurgia, Universita` di Foggia, via L. Pinto, 71100 Foggia, Italy 2 Department of Human Anatomy and Histology, Bari University, Medical School, Italy 3 S. Maugeri Foundation, IRCSS, Bari, Italy 4 Department of Medical and Occupational Sciences, Foggia University, Italy Received 12 May 2003; accepted 22 December 2003 Available online 5 March 2004
Abstract The study, undertaken with the aim of further investigating the effects of methylmercury (MeHg) exposure on the developing brain, was performed in the cerebellum of chick embryos, chronically treated with a MeHgCl solution dropped onto the chorioallantoic membrane, and in control embryo cerebella. Quantitative evaluations, performed by cold vapour atomic absorption spectrophotometry, demonstrated a high mercury content in the chorioallantoic membrane, encephalon, liver and kidney of the treated embryos. The morphological observations showed severe neuronal damage consisting of degenerative changes of the granules and Purkinje neurons. The effects on astrocytes were even more severe, since they were extremely rare both in the neuropil and around the vessel wall. Compared with the controls, the cerebellar vessels of MeHg-treated embryos showed immature morphology, poor differentiation of endothelial barrier devices, and high permeability to the exogenous protein horseradish peroxidase. These findings support the hypothesis that MeHgrelated neuronal sufferance may be secondary to astrocytic damage and suggest that the developmental neurotoxicity of this compound could also be related to astrocyte loss-dependent impairment of blood–brain barrier (BBB) differentiation. # 2004 Elsevier Inc. All rights reserved.
Keywords: Methylmercury; Development; Neurotoxicity; Cerebellum; Astrocytes; Blood–brain barrier
INTRODUCTION The evidence of the neurotoxicity of methylmercury (MeHg) dates back to the 50s–60s, when it was ascertained that the neurological diseases affecting thousands of people in Minamata and Niigata (Japan) were caused by consumption of fish contaminated by MeHg (Harada, 1978; Irukayama et al., 1962; Takeuchi et al., 1962). MeHg exposure remains a major public health concern because of natural and anthropogenic release of inorganic mercury into the aquatic environment (Crinnion, 2000; Pirrone, 2001), where it is biotrans*
Corresponding author. Tel.: þ39-0805478310; fax: þ39-0805478310. E-mail addresses: [email protected], [email protected] (M. Bertossi).
formed by algae and bacteria into MeHg. This can pass along the food chain and, eventually, to man (Jensen and Jernelov, 1969; Wood et al., 1968). Neuropathological changes, such as damage to specific cerebral and cerebellar areas and sensory fibres of the spinal nerves, have been described in both man and laboratory animals exposed to MeHg (Eto, 1997; Garman et al., 1975; Hunter and Russell, 1954; Nagashima, 1997). However, the severity of the damage differs according to exposure time and dose, as well as animal species and age (Berlin, 1986; Burbacher et al., 1990; Kakita et al., 2000a,b; Wakabayashi et al., 1995; WHO, 1993; Willes, 1977). The developing brain is extremely sensitive to MeHg poisoning; it is well known, in fact, that prolonged preand/or perinatal exposure to MeHg, even at moderate doses, results in widespread neuronal damage, severe
0161-813X/$ – see front matter # 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.neuro.2004.01.005
850
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857
motor and sensory disturbances, neuropsychological dysfunctions, and mental retardation (Andersen et al., 2000; Crinnion, 2000; Choi et al., 1978; Grandjean et al., 1997; Harada, 1978; Kjellstrom et al., 1986, 1989; Mahaffey, 2000; Myers and Davidson, 2000; Sakamoto et al., 1998, 2002; Steuerwald et al., 2000; Watanabe and Sato, 1996). The demonstration that MeHg contained in food is almost all absorbed by the body and crosses the blood–brain barrier (BBB) and the placenta explains its neurotoxic and teratogenic effects (Kakita et al., 2000b; Mansour et al., 1974). Despite the strong evidence of MeHg neurotoxicity, the mechanism whereby the damage to the central nervous system (CNS) is wrought remains to be elucidated. For instance, there is some debate about the primary cellular target of MeHg (either neurons or astrocytes), although it has been demonstrated that astrocytes preferentially sequester MeHg and display significant alterations when exposed to it, in vivo or in vitro (Allen et al., 2002; Aschner, 2000, 2002; Charleston et al., 1996; Dare´ et al., 2001). Furthermore, no data have been published regarding the effects of developmental exposure to MeHg on the growth of the neural vessels and differentiation of BBB, which constitute essential events for the proper morphofunctional development of the CNS. To gain a better understanding of the damage following developmental exposure of CNS to MeHg, we undertook the present research with the aim of further investigating the changes occurring in the neurons and glial cells and ascertaining whether MeHg causes impairment of the developing neural vessels. Particular attention was paid to BBB differentiation, which takes place progressively during prenatal life and is thought to be controlled by astroglial cells (Akiyama et al., 2000; Bertossi et al., 1993, 1997, 1999, 2002, 2003; Hayashi et al., 1997; Hurwitz et al., 1993; Janzer and Raff, 1987; Kuchler-Bopp et al., 1999; Roncali et al., 1986; Stewart and Wiley, 1981; Virgintino et al., 1998, 2000; Wakai and Hirokawa, 1978). The study was carried out in the cerebellum, which is a well-known target of MeHg poisoning (Hunter and Russell, 1954; Nagashima, 1997).
MATERIALS AND METHODS For this study we utilized an experimental model previously tested in an investigation on the effects of lead acetate on the development of the chick embryo kidney (Errede et al., 2001). In brief, fertilized eggs of White Leghorn chicken were incubated in conditions
of constant humidity and temperature (37 8C). At the 3rd day of incubation, 2–3 cc of albumen were drained away to detach the surface of the chorioallantoic membrane (CAM) from the egg shell; a small window was then made in the shell and closed by a removable glass. A daily concentration of 5 mg/g egg weight of methylmercury (MeHg) chloride was chosen on the basis of literature data in the rat (Sakamoto et al., 1998) and subsequently reduced to 2.5 mg/g per day. In fact, a series of preliminary experiments with progressively lower MeHg concentrations demonstrated that doses higher than 2.5 mg/g per day caused poor survival of the embryos (less than 50% after 1–3 days of treatment), whereas the dose of 2.5 mg/g per day allowed similar survival of the treated embryos to that of the controls (70%). Starting from the 12th day of incubation, seventy embryos were treated daily with 2.5 mg/g egg weight of MeHgCl diluted in saline solution and dropped onto the CAM surface. Ten control embryos were treated daily with saline solution alone. To perform quantitative evaluations of the mercury content in different organs, the CAM, encephalon, liver, and kidney were isolated from the treated and control embryos which survived until hatching time (20–21 days of incubation), sacrificed by decapitation. Soon after sacrifice, three to four cerebellar lamellae were isolated from the encephalon of the embryos and submitted to the routine procedures for light and electron microscopy (see further). To evaluate the permeability of the cerebellar vessels to exogenous protein, three treated and three control embryos received an intracardial injection of horseradish peroxidase (HRP) (see further). Quantitative Evaluation of Mercury Content The glass wares utilized for the following procedures were cleaned with 0.5% nitric acid in an ultrasound bath for 2 h, washed in double distilled water and dried in an oven. The encephala, livers, kidneys and CAMs were separately dried in open glass vials; each sample was kept in an oven for 6 h at 60 2 8C, 10 h at 80 2 8C, 24 h at 110 2 8C, and weighed after cooling at room temperature. The dry samples were placed in screw capped glass vials and digested in 5 ml of an acid mixture of 60% sulphuric acid and 40% nitric acid (Instra-Analized, J.T. Baker, Italy) in a thermostatic bath at 45 0:5 8C for 24 h to obtain the complete dissolution of the tissues. Thereafter, the samples were cooled and reconstituted to 50 ml with double distilled water. The quantitative analysis of the mercury content was performed by cold vapour atomic absorption
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857
spectrophotometry with a Perkin–Elmer 5100 ZL spectrophotometer. The quantification was performed by testing five aliquots of samples (1–2 ml) and adding 0, 20, 30, 40 and 60 ng of Hg/sample, respectively. This method of quantification was chosen instead of quantification by an external standard to reduce the matrix effects on the Hg signal. The instrumental parameters were set as indicated according to the operation manual of the instrument used for the mercury analysis. Mercury standard solution and MeHgCl were purchased from Sigma–Aldrich (Milan, Italy). Light and Electron Microscopy Cerebellar lamellae were fixed in Bouin’s solution and embedded in paraffin; the histological sections were stained with hematoxylin-eosin or with toluidine blue. Small fragments of cerebellar lamellae were fixed in 3% phosphate-buffered glutaraldehyde, embedded in Epon 812 and cut in ultrathin sections (60 nm) with an ultramicrotome. The sections were double-stained with uranyl acetate and lead citrate. HRP Assay Three treated and three control embryos received an intracardial injection of HRP (0.3 mg Sigma type II HRP/g body weight in 0.1–0.3 ml saline solution). The embryos were sacrificed 10 min after the injection and the cerebella were fixed in a sodium-cacodylate buffered mixture of 2% glutaraldehyde and 2% paraformaldehyde. Thick sections (50 mm), cut with a vibratome, were incubated in a 0.05% solution of 30 ,30 diaminobenzidine in 0.05 M Tris–HCl buffer containing 0.01% H2O2. After incubation, the thick sections were dehydrated and mounted on slides; the peroxidase activity was apparent at the light microscope as a brown reaction product (Roncali et al., 1986). Light microscopy sections were photographed with a Vanox-T (Olympus) microscope equipped with a high resolution CCD/KAI-2000 video camera (SPOT Insight Color, Diagnostic Instruments, MI, USA). The ultrathin sections were photographed under the electron microscope EM109 (Carl Zeiss). Morphometric Analysis Image acquisition and measurement were performed by the image analyser VIDAS 2.5 (Kontron Elektronik GmbH, Eching, Germany) connected either via a highresolution (1600 1200 active pixels) CCD/KAI-2000 video camera to an Axioscop 20 (Carl Zeiss) light
851
microscope, or connected to a CCD/RGB video camera (TK-1070E; JVC, Japan) for electron microphotograph analysis. A semiautomatic/interactive morphometric analysis was applied to measure the number of Purkinje neurons affected by degenerative changes. Randomly chosen fields (50 for treated and 30 for control embryos) were acquired at 40 magnification from toluidine blue stained sections of cerebellar lamellae. On the digitised images, both the total number and the number of damaged Purkinje neurons were counted by a ‘mark points’ interactive measurement function and expressed as percent mean value standard deviation. Randomly chosen vascular fields (200 for treated and 160 for control embryos) were digitised from electron microphotographs of the cerebellar lamellae acquired at 18,000 magnification. The total vessel perimeter and the length of the vessel perimeter contacted by astrocytes were calculated by a ‘trace contour’ interactive function and expressed as mean value standard deviation of the percentage of the vessel perimeter lined by astrocytes. The statistical significance of the obtained values was assessed by the t-test.
RESULTS Quantitative Evaluation of Mercury Content No traces of mercury were present in the encephalon, liver, and kidney of the control embryos, while a mercury content of 0:8 0:12 mg/g dry weight was detected in the CAM. The organs of MeHg-poisoned embryos contained a large amount of mercury; the highest value was detected in the CAM (261:9 8:0 mg/g dry weight) and there were virtually similar amounts in the encephalon (75:6 9:8 mg/g dry weight), liver (76:4 9:2 mg/g dry weight), and kidney (79:2 7:8 mg/g dry weight). Light and Electron Microscopy Compared with the controls (Fig. 1a and b), the cerebellar lamellae of MeHg-exposed embryos (Fig. 1c and d) appeared hypotrophic and characterized by widespread cellular damage. In particular, the outer granules were arranged in a discontinuous layer, where a number of cells displayed pyknotic nuclei (Fig. 1d). A statistically significant percentage of Purkinje neurons (59:19 9:20% versus 3:78 0:78 in the controls, P ¼ 0:001) was affected by degenerative changes
852
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857
Fig. 1. Lamellae and details of the cerebellar cortex of control (a and b) and MeHg-treated (c and d) embryos. Note, in the treated embryo, the reduced size of the cerebellar lamella (c compared with a), as well as the disarrangement and irregular thickness of the external granular layer (EGL) (d compared with b). A number of outer (arrows) and inner (double arrow) granules, as well as Purkinje neurons (arrowheads), display nuclear pyknosis (d compared with b). Purkinje neurons featuring chromatolysis and an unstained cytoplasm are also present (d, double arrowheads). In the microvessels of a control embryo cerebellum, the permeability marker HRP has an intravascular localization (e). In a MeHg-treated embryo, HRP is widely diffused throughout the cerebellar neuropil (f). Bars: (a and c) 100 mm; (b and d) 50 mm; (e and f) 25 mm.
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857
853
Fig. 2. Cytoplasmic and nuclear alterations affecting two Purkinje neurons of a MeHg-treated embryo (a). Cerebellar microvessel of a control embryo displaying a regular course of the inner and outer surfaces of the vessel wall and extensive tight junctions (b, double arrowheads) joining the endothelial cells (E). The vessel wall is bordered by a continuous basal lamina (b, arrowheads) and astrocytic end feet (A) linked by gap junctions (b, double arrow). The microvessels of MeHg-treated embryos (c and d) show luminal expansions of the endothelial cells (E, arrows) and pericyte processes (d, P) extending toward the neuropil. The endothelial cells contain numerous vesicles and vacuoles, and are linked by short junctions (c, double arrowhead). The basal lamina enwraps only part of the vessel (c, arrowheads) or is entirely absent (d). The vessel shown in (c) is completely devoid of perivascular glia. Only two small astrocytic end feet (A) are appreciable in (d). Bars: (a) 2 mm and (b–d) 1 mm.
854
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857
consisting of cellular shrinkage, pyknosis, chromatolysis, or a scarce amount of Nissl’s substance (Figs. 1d and 2a). Neurons containing pyknotic nuclei were also recognizable in the inner granular layer (Fig. 1d). Unlike the cerebellum of control embryos, where the permeability marker HRP was retained within the microvessels (Fig. 1e), the cerebellum of all treated embryos displayed massive extravasation of the marker throughout the neuropil (Fig. 1f). By the electron microscope, the cerebellar microvessels of control embryos (Fig. 2b) showed a regular course of the inner and outer profiles of the endothelium-pericyte layer. The endothelial cells were sealed by extensive tight junctions and contained a low number of pinocytotic vesicles. The outer surface of the vessel wall was bordered by a continuous, relatively thick, basal lamina and by an uninterrupted layer of astrocytic end feet, joined by gap junctions and containing clusters of glycogen granules. In the treated embryos, on the contrary, the cerebellar microvessels (Fig. 2c and d) displayed irregular profiles, mainly at the luminal front, a high number of endothelial vesicles and vacuoles, and poorly developed interendothelial junctions; the nucleus and cytoplasmic organelles of endothelial cells and pericytes appeared structurally normal. In a number of microvessels, the basal lamina was discontinuous (Fig. 2c) or entirely lacking (Fig. 2d). Astrocyte bodies and processes were extremely rare, both in the neuropil and near the microvessels (Fig. 2c and d). The quantitative evaluation of the perivascular glia extension showed that astrocytic end feet covered 6:24 1:36% of the endotheliumpericyte layer in the treated embryos versus 96 3:4% in the control embryos (statistical significance P ¼ 0:001). No cellular debris was seen around the vessel walls.
DISCUSSION AND CONCLUSIONS The results of the quantitative evaluations, indicating high mercury concentrations in the brain, liver and kidney, demonstrate the efficacy of our experimental protocol, since these organs are known to be the main sites of mercury storage (Crinnion, 2000; Nagashima, 1997; Opitz et al., 1996; Sakamoto et al., 1998; WHO, 1993). The morphological observations demonstrate severe cellular sufferance in the cerebellum of treated embryos and indicate that the neurons most severely affected by MeHg are the cerebellar granules and Purkinje neurons. MeHg-induced damage to cerebellar
granules has been documented in several in vitro and in vivo studies, one of which suggests that granule cell death depends on MeHg-impairment of cell migration toward the inner granular layer. Instead, the Purkinje neurons of adult humans as well as of neonatal and adult rats appear to be relatively spared, or simply disarranged, under conditions of chronic MeHg exposure (Eto, 1997; Kunimoto, 1994; Kunimoto and Suzuki, 1997; Nagashima, 1997; Sakamoto et al., 1998, 2002; Wakabayashi et al., 1995). The discrepancy between these results and the extensive damage affecting the Purkinje neurons in the chick embryo corroborates the finding that sensitivity to MeHg varies according to both animal species and age (Berlin, 1986; Burbacher et al., 1990; Kakita et al., 2000a,b; Wakabayashi et al., 1995; Willes, 1977; WHO, 1993) and further demonstrates the higher risk of exposure to this compound during development. In this context, it must be remembered that the MeHg treatment in the chick embryo coincided temporally with crucial events of the cerebellar morphohistogenesis, such as differentiation of granule neurons, and their migration toward the inner granular layer, differentiation of Purkinje and astroglial cells, as well as arrangement of astrocytic end-feet around the growing microvessels (Bertossi et al., 1986, 1997; Mugnaini, 1969; Mugnaini and Forstrønen, 1967; Virgintino et al., 1993). Many in vitro studies have proposed different mechanisms through which MeHg exerts cytotoxic effects, such as oxidative stress, apoptosis, necrosis, altered transport of glutamate, and inhibition of mitosis via microtubule disruption (Allen et al., 2002; Amorim et al., 2000; Aschner, 2002; Aschner et al., 2000; Castoldi et al., 2001; Dare´ et al., 2001; Miura et al., 1999). Considering the widespread cellular damage observed in the cerebellum of treated embryos, it is reasonable to assume that MeHg may exert its action on the neuronal and glial populations through one or more of these mechanisms. In particular, the reduction of astrocytes could be related to impairment of proliferation of their precursor cells rather than to astroglia degeneration, since no cellular debris were observed in the MeHg-exposed cerebellum. The strong sensitivity to MeHg displayed by astroglial cells is in agreement with the results of investigations performed in vitro (Allen et al., 2002; Aschner, 2002; Aschner et al., 2000) and suggests that in vivo, and during development, astroglial cells are even more sensitive than neurons to MeHg exposure. It is well known that astrocytes play active roles in the development and function of the CNS. Thanks to the presence of gap junctions, they work as a
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857
functional syncytium that maintains the physiological composition of the extracellular fluid and, among other actions, preserves the neurons from glutamatemediated excitotoxicity (LoPachin and Aschner, 1993; Aschner and Kimelberg, 1996; Barres and Barde, 2000; Ransom et al., 2003). Therefore, it seems conceivable that MeHg-related neuronal sufferance may be secondary to, or at least enhanced by, astrocytic damage (Allen et al., 2002; Aschner, 2002; Brookes, 1992). The ultrastructural features and permeability to the marker HRP shown at hatching by the cerebellar microvessels of MeHg-poisoned embryos are similar to those observed in normally developed embryos at earlier stages (Bertossi et al., 1997, 2003). These findings indicate that MeHg exposure does not cause structural damage to endothelial cells and pericytes, but does lead to significantly delayed maturation of the vessels and imperfect acquirement of their barrier properties. The relationship between absence of perivascular astrocyte end feet and immature morphofunctional features of the cerebellar microvessels likely represents a further indication of the pivotal role played by astroglial cells in the proper differentiation of BBBprovided vessels. Taken as a whole, our observations strongly suggest that astrocyte loss-dependent impairment of the BBB is consistently involved in the developmental neurotoxicity of MeHg.
ACKNOWLEDGEMENTS The work was supported in part by grants from University of Foggia to M. Bertossi (60%, 2001, 2002). We thank Ms M.V.C. Pragnell, BA, for linguistic help and Ms M. Ambrosi and S. Cantatore for their skilful technical assistance.
REFERENCES Akiyama H, Kondoh T, Kokunai T, Nagashima T, Saito M, Tamaki N. Blood-brain barrier formation of grafted human umbilical vein endothelial cells in athymic mouse brain. Brain Res 2000;858:172–6. Allen JW, Shanker G, Tan KH, Aschner M. The consequences of methylmercury exposure on interactive functions between astrocytes and neurons. Neurotoxicology 2002;23:755–9. Amorim MLM, Mergler D, Bahia MO, Dubeau H, Miranda D, Lebel J, Burbano RR, Lucotte M. Cytogenetic damage related to low levels of methylmercury contamination in the Brazilian Amazon. An Acad Bras Ci 2000;72:497–507.
855
Andersen HR, Nielsen JB, Grandjean Ph. Toxicologic evidence of developmental neurotoxicity of environmental chemicals. Toxicology 2000;144:121–7. Aschner M. Neurotoxic mechanism of fish-borne methylmercury. Environ Toxicol Pharmacol 2002;12:101–4. Aschner M, Kimelberg AK, The role of glia in neurotoxicity. Boca Raton: CRC Press; 1996. Aschner M, Yao CP, Allen JW, Tan KH. Methylmercury alters glutamate transport in astrocytes. Neurochem Int 2000;37: 199–206. Barres BA, Barde Y-A. Neuronal and glial cell biology. Curr Opin Neurobiol 2000;10:642–8. Berlin M. Mercury. In: Friberg L, Nordberg GF, Vouk V, editors. Handbook on the toxicology of metals, vol. II. Amsterdam: Elsevier; 1986, p. 387–445. Bertossi M, Roncali L, Mancini L, Ribatti D, Nico B. Process of differentiation of cerebellar Purkinje neurons in the chick embryo. Anat Embryol 1986;175:25–34. Bertossi M, Roncali L, Nico B, Ribatti D, Mancini L, Virgintino D, Fabiani G, Guidazzoli A. Perivascular astrocytes and endothelium in the development of the blood-brain barrier in the optic tectum of the chick embryo. Anat Embryol 1993; 188:21–9. Bertossi M, Virgintino D, Roncali L, Riva A. Embryo cerebellum microvessels under scanning electron microscopy. In: Motta PM, editor. Recent advances in microscopy of cells, tissues and organs. Antonio Delfino Editore: Rome; 1997. p. 215–8 Bertossi M, Virgintino D, Errede M, Roncali L. Immunohistochemical and ultrastructural characterization of cortical plate microvasculature in the human fetus telencephalon. Microvasc Res 1999;58:49–61. Bertossi M, Girolamo F, Errede M, Benagiano V, Virgintino D, Roncali L. Developmental changes of HT7 expression in the microvessels of the chick embryo brain. Anat Embryol 2002; 205:229–33. Bertossi M, Girolamo F, Errede M, Virgintino D, Roncali L. Effects of the 6-aminonicotinamide gliotoxin on the blood-brain barrier differentiation in the chick embryo cerebellum. Anat Embryol 2003;207:209–19. Brookes N. In vitro evidence for the role of glutamate in the CNS toxicity of mercury. Toxicology 1992;76:245–56. Burbacher TM, Rodier PM, Weiss B. Methylmercury developmental neurotoxicity: a comparison of effects in humans and animals. Neurotoxicol Teratol 1990;12:191–202. Castoldi AF, Coccini T, Ceccatelli S, Manzo L. Neurotoxicity and molecular effects of methylmercury. Brain Res Bull 2001;55: 197–203. Charleston JS, Body RL, Bolender RP, Vahter ME, Burbas TM. Changes in the number of astrocytes and microglia in the thalamus of the monkey Macaca fascicularis following longterm subclinical methylmercury exposure. Neurotoxicology 1996;17:127–38. Choi BH, Lapham LW, Amin-Zaki L, Saleem T. Abnormal neuronal migration, deranged cerebral cortex organization, and diffuse white matter astrogliosis of human fetal brain: a major effect of methylmercury poisoning in utero. J Neuropathol Exp Neurol 1978;37:719–33. Crinnion WJ. Environmental medicine. Part 3. Long-term effects of chronic low-dose mercury exposure. Altern Med Rev 2000; 5:209–23.
856
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857
Dare´ E, Li W, Zhivotovsky B, Yuan X, Ceccatelli S. Methylmercury and H2O2 provoke lysosomal damage in human astrocytoma D384 cells followed by apoptosis. Free Radic Biol Med 2001;30:1347–56. Errede M, Elia G, Bertossi M, Mulas ML, Riva A, Virgintino D, Benagiano V, Girolamo F, Roncali L, Ambrosi L. Developmental effects of lead acetate on the chick embryo metanephros. J Submicrosc Cytol Pathol 2001;33:261–73. Eto K. Pathology of Minamata disease. Toxicol Pathol 1997;25: 614–23. Garman RH, Weiss B, Evans HL. Alkylmercurial encephalopathy in the monkey (Saimiri sciureu and Macaca arctoides): a histopathologic and autoradiographic study. Acta Neuropathol (Berl) 1975;32:61–4. Grandjean P, Weihe P, White RF, Debes F, Araki S, Yohoyama K, Murata K, Sorensen N, Dahl R, Jorgensen PJ. Cognitive deficit in 7-year-old-children with prenatal exposure to methylmercury. Neurotoxicol Teratol 1997;19:417–28. Harada M. Congenital Minamata disease: intrauterine methylmercury poisoning. Teratology 1978;18:285–8. Hayashi Y, Nomura M, Yamagishi SL, Harada SJ, Yamashita J, Yamamoto H. Induction of various blood-brain barrier properties in non-neural endothelial cells by close apposition to co-cultured astrocytes. Glia 1997;19:13–26. Hunter D, Russell DS. Focal cerebral and cerebellar atrophy in a human subject due to organic mercury compounds. J Neurol Neurosurg Psychiatr 1954;17:235–41. Hurwitz AA, Berman JW, Rashbaum WK, Lyman WD. Human fetal astrocytes induce the expression of blood-brain barrier specific proteins in autologous endothelial cells. Brain Res 1993;625:238–43. Irukayama K, Kai F, Fujiki M, Kondo T. Study on the origin of causative agent of Minamata disease. III. Industrial waste containing mercury compounds from Minamata factory. Kumamoto Med J 1962;15:57–68. Janzer RC, Raff MC. Astrocytes induce blood-brain barrier properties in endothelial cells. Nature 1987;325:253–7. Jensen S, Jernelov A. Biological methylation of mercury in aquatic organisms. Nature 1969;223:753–4. Kakita A, Wakabayashi K, Su M, Yoneoka Y, Sakamoto M, Ikuta F, Takahashi H. Intrauterine methylmercury intoxication. Consequence of the inherent brain lesions and cognitive dysfunction in maturity. Brain Res 2000a;877:322–30. Kakita A, Wakabayashi K, Su M, Sakamoto M, Ikuta F, Takahashi H. Distinct pattern of neuronal degeneration in the fetal rat brain induced by consecutive transplacental administration of methylmercury. Brain Res 2000;859:233–9. Kjellstrom T, Kennedy P, Wallis S, Mantell C. Physical and mental development of children with prenatal exposure to mercury from fish. Stage I: preliminary tests at age 4. Solna (Sweden): National Swedish Environmental Protection Board Report; 1986. p. 3080. Kjellstrom T, Kennedy P, Wallis S, Stewart A, Friberg L. Physical and mental development of children with prenatal exposure to mercury from fish: Stage II: interviews and psychological tests at age 6. Solna (Sweden): National Swedish Environmental Protection Board Report; 1989. p. 3642. Kuchler-Bopp S, Delaunoy JP, Artault JC, Zaepfel M, Dietrich JB. Astrocytes induce several blood-brain barrier properties in non-neural endothelial cells. Neuroreport 1999;10:1347– 53.
Kunimoto M. Methylmercury induces apoptosis of rat cerebellar neurons in primary culture. Biochem Biophys Res Commun 1994;204:310–7. Kunimoto M, Suzuki T. Migration of granule neurons in cerebellar organotypic cultures is impaired by methylmercury. Neurosci Lett 1997;226:183–6. LoPachin RM, Aschner M. Glial-neuronal interactions: the relevance to neurotoxic mechanisms. Toxicol Appl Pharmacol 1993;118:141–58. Mahaffey KR. Recent advances in recognition of low-level mercury poisoning. Curr Opin Neurol 2000;13:699–707. Mansour MM, Dyer NC, Hoffman LH, Davies J, Brill AB. Placental transfer of mercuric nitrate and methylmercury in the rat. Am J Obstet Gynecol 1974;119:557–62. Miura K, Koide N, Himeno S, Nagakawa I, Imura N. The involvement of microtubular disruption in methylmercury induced apoptosis in neuronal and nonneuronal cell lines. Toxicol Appl Pharmacol 1999;160:279–88. Mugnaini E, Ultrastructural studies on the cerebellar histogenesis. II. Maturation of nerve cell populations and establishment of synaptic connections in the cerebellar cortex of the chick. In: Llina`s RR, editor. Neurobiology of cerebellar evolution and development. Chicago: AMA Educational and Research Foundation; 1969. p. 749–82. Mugnaini E, Forstrønen PF. Ultrastructural studies on the cerebellar histogenesis. I. Differentiation of granule cells and development of glomeruli in the chick embryo. Z Zellforsch 1967;77:115–43. Myers GJ, Davidson PhW. Does methylmercury have a role in causing developmental disabilities in children? Environ Health Perspect 2000;108:413–20. Nagashima K. A review of experimental methylmercury toxicity in rats: Neuropathology and evidence for apoptosis. Toxicol Pathol 1997;25:624–31. Opitz H, Schweinsberg F, Grossmann T, Wendt-Gallitelli MF, Meyermann R. Demonstration of mercury in the human brain and other organs 17 years after metallic mercury exposure. Clin Neuropathol 1996;15:139–44. Pirrone N. Mercury research in Europe: towards the preparation of the EU air quality directive. Atmos Environ 2001;35: 2979–86. Ransom B, Behar T, Nedergaard M. New roles for astrocytes (stars at last). Trends Neurosciences 2003;26:620–2. Roncali L, Nico B, Ribatti D, Bertossi M, Mancini L. Microscopical and ultrastructural investigations on the development of the blood-brain barrier in the chick embryo optic tectum. Acta Neuropathol (Berl) 1986;70:193–201. Sakamoto M, Wakabayashi K, Kakita A, Takahashi H, Adachi T, Nakano A. Widespread neuronal degeneration in rats following oral administration of methylmercury during the postnatal developing phase: a model of fetal-type Minamata disease. Brain Res 1998;784:351–4. Sakamoto M, Kakita A, Wakabayashi K, Takahashi H, Nakano A, Akagi H. Evaluation of changes in methylmercury accumulation in the developing rat brain and its effects: a study with consecutive and moderate dose exposure throughout gestation and lactation periods. Brain Res 2002;949:51–9. Steuerwald U, Weihe P, Jorgensen PJ, Bjerve K, Brock J, Heinzow B, Budtz-Jorgensen E, Grandjean P. Maternal seafood diet, methylmercury exposure, and neonatal neurologic function. J Pediatr 2000;136:599–605.
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857 Stewart PA, Wiley MJ. Developing nervous tissue induces formation of blood-brain barrier characteristics in invading endothelial cells. Dev Biol 1981;84:183–92. Takeuchi T, Morikawa N, Matsumoto H, Shiraishi A. A pathological study of Minamata disease. Acta Neuropathol 1962;2:40–57. Virgintino D, Maiorano E, Bertossi M, Pollice L, Ambrosi G, Roncali L. Vimentin- and GFAP-immunoreactivity in developing and mature neural microvessels. Study in the chicken tectum and cerebellum. Eur J Histochem 1993;37:353–62. Virgintino D, Maiorano E, Errede M, Vimercati A, Greco P, Selvaggi L, Roncali L, Bertossi M. Astroglia-microvessel relationship in the developing human telencephalon. Int J Dev Biol 1998;42:1165–8. Virgintino D, Robertson D, Benagiano V, Errede M, Bertossi M, Ambrosi G, Roncali L. Immunogold cytochemistry of the blood-brain barrier glucose transporter GLUT1 and endogenous albumin in the developing human brain. Dev Brain Res 2000;123:95–101.
857
Wakabayashi K, Kakita A, Sakamoto M, Su M, Iwanaga K, Ikuta F. Variability of brain lesions in rat administered methylmercury at various postnatal developmental phases. Brain Res 1995;705:267–72. Wakai S, Hirokawa N. Development of the blood-brain barrier to horseradish peroxidase in the chick embryo. Cell Tiss Res 1978;195:195–203. Watanabe C, Sato H. Evolution of our understanding of methylmercury as a health threat. Environ Health Perspect 1996;104:367–79. Willes RF. Tissue distribution as a factor in species susceptibility to toxicity and hazard assessment. Example: methylmercury. J Environ Pathol Toxicol 1977;1:135–46. WHO. Evaluation of carcinogenic risks to humans, vol. 58; 1993. p. 2339–45 Wood JM, Kennedy FS, Rosen CE. Synthesis of methylmercury compounds by extracts of methanogenic bacterium. Nature 1968;220:173–4.
Effects of Methylmercury on the Microvasculature of the Developing Brain M. Bertossi1,*, F. Girolamo2, M. Errede2, D. Virgintino2, G. Elia3, L. Ambrosi4, L. Roncali2 1
Dipartimento di Scienze Biomediche, Facolta` di Medicina e Chirurgia, Universita` di Foggia, via L. Pinto, 71100 Foggia, Italy 2 Department of Human Anatomy and Histology, Bari University, Medical School, Italy 3 S. Maugeri Foundation, IRCSS, Bari, Italy 4 Department of Medical and Occupational Sciences, Foggia University, Italy Received 12 May 2003; accepted 22 December 2003 Available online 5 March 2004
Abstract The study, undertaken with the aim of further investigating the effects of methylmercury (MeHg) exposure on the developing brain, was performed in the cerebellum of chick embryos, chronically treated with a MeHgCl solution dropped onto the chorioallantoic membrane, and in control embryo cerebella. Quantitative evaluations, performed by cold vapour atomic absorption spectrophotometry, demonstrated a high mercury content in the chorioallantoic membrane, encephalon, liver and kidney of the treated embryos. The morphological observations showed severe neuronal damage consisting of degenerative changes of the granules and Purkinje neurons. The effects on astrocytes were even more severe, since they were extremely rare both in the neuropil and around the vessel wall. Compared with the controls, the cerebellar vessels of MeHg-treated embryos showed immature morphology, poor differentiation of endothelial barrier devices, and high permeability to the exogenous protein horseradish peroxidase. These findings support the hypothesis that MeHgrelated neuronal sufferance may be secondary to astrocytic damage and suggest that the developmental neurotoxicity of this compound could also be related to astrocyte loss-dependent impairment of blood–brain barrier (BBB) differentiation. # 2004 Elsevier Inc. All rights reserved.
Keywords: Methylmercury; Development; Neurotoxicity; Cerebellum; Astrocytes; Blood–brain barrier
INTRODUCTION The evidence of the neurotoxicity of methylmercury (MeHg) dates back to the 50s–60s, when it was ascertained that the neurological diseases affecting thousands of people in Minamata and Niigata (Japan) were caused by consumption of fish contaminated by MeHg (Harada, 1978; Irukayama et al., 1962; Takeuchi et al., 1962). MeHg exposure remains a major public health concern because of natural and anthropogenic release of inorganic mercury into the aquatic environment (Crinnion, 2000; Pirrone, 2001), where it is biotrans*
Corresponding author. Tel.: þ39-0805478310; fax: þ39-0805478310. E-mail addresses: [email protected], [email protected] (M. Bertossi).
formed by algae and bacteria into MeHg. This can pass along the food chain and, eventually, to man (Jensen and Jernelov, 1969; Wood et al., 1968). Neuropathological changes, such as damage to specific cerebral and cerebellar areas and sensory fibres of the spinal nerves, have been described in both man and laboratory animals exposed to MeHg (Eto, 1997; Garman et al., 1975; Hunter and Russell, 1954; Nagashima, 1997). However, the severity of the damage differs according to exposure time and dose, as well as animal species and age (Berlin, 1986; Burbacher et al., 1990; Kakita et al., 2000a,b; Wakabayashi et al., 1995; WHO, 1993; Willes, 1977). The developing brain is extremely sensitive to MeHg poisoning; it is well known, in fact, that prolonged preand/or perinatal exposure to MeHg, even at moderate doses, results in widespread neuronal damage, severe
0161-813X/$ – see front matter # 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.neuro.2004.01.005
850
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857
motor and sensory disturbances, neuropsychological dysfunctions, and mental retardation (Andersen et al., 2000; Crinnion, 2000; Choi et al., 1978; Grandjean et al., 1997; Harada, 1978; Kjellstrom et al., 1986, 1989; Mahaffey, 2000; Myers and Davidson, 2000; Sakamoto et al., 1998, 2002; Steuerwald et al., 2000; Watanabe and Sato, 1996). The demonstration that MeHg contained in food is almost all absorbed by the body and crosses the blood–brain barrier (BBB) and the placenta explains its neurotoxic and teratogenic effects (Kakita et al., 2000b; Mansour et al., 1974). Despite the strong evidence of MeHg neurotoxicity, the mechanism whereby the damage to the central nervous system (CNS) is wrought remains to be elucidated. For instance, there is some debate about the primary cellular target of MeHg (either neurons or astrocytes), although it has been demonstrated that astrocytes preferentially sequester MeHg and display significant alterations when exposed to it, in vivo or in vitro (Allen et al., 2002; Aschner, 2000, 2002; Charleston et al., 1996; Dare´ et al., 2001). Furthermore, no data have been published regarding the effects of developmental exposure to MeHg on the growth of the neural vessels and differentiation of BBB, which constitute essential events for the proper morphofunctional development of the CNS. To gain a better understanding of the damage following developmental exposure of CNS to MeHg, we undertook the present research with the aim of further investigating the changes occurring in the neurons and glial cells and ascertaining whether MeHg causes impairment of the developing neural vessels. Particular attention was paid to BBB differentiation, which takes place progressively during prenatal life and is thought to be controlled by astroglial cells (Akiyama et al., 2000; Bertossi et al., 1993, 1997, 1999, 2002, 2003; Hayashi et al., 1997; Hurwitz et al., 1993; Janzer and Raff, 1987; Kuchler-Bopp et al., 1999; Roncali et al., 1986; Stewart and Wiley, 1981; Virgintino et al., 1998, 2000; Wakai and Hirokawa, 1978). The study was carried out in the cerebellum, which is a well-known target of MeHg poisoning (Hunter and Russell, 1954; Nagashima, 1997).
MATERIALS AND METHODS For this study we utilized an experimental model previously tested in an investigation on the effects of lead acetate on the development of the chick embryo kidney (Errede et al., 2001). In brief, fertilized eggs of White Leghorn chicken were incubated in conditions
of constant humidity and temperature (37 8C). At the 3rd day of incubation, 2–3 cc of albumen were drained away to detach the surface of the chorioallantoic membrane (CAM) from the egg shell; a small window was then made in the shell and closed by a removable glass. A daily concentration of 5 mg/g egg weight of methylmercury (MeHg) chloride was chosen on the basis of literature data in the rat (Sakamoto et al., 1998) and subsequently reduced to 2.5 mg/g per day. In fact, a series of preliminary experiments with progressively lower MeHg concentrations demonstrated that doses higher than 2.5 mg/g per day caused poor survival of the embryos (less than 50% after 1–3 days of treatment), whereas the dose of 2.5 mg/g per day allowed similar survival of the treated embryos to that of the controls (70%). Starting from the 12th day of incubation, seventy embryos were treated daily with 2.5 mg/g egg weight of MeHgCl diluted in saline solution and dropped onto the CAM surface. Ten control embryos were treated daily with saline solution alone. To perform quantitative evaluations of the mercury content in different organs, the CAM, encephalon, liver, and kidney were isolated from the treated and control embryos which survived until hatching time (20–21 days of incubation), sacrificed by decapitation. Soon after sacrifice, three to four cerebellar lamellae were isolated from the encephalon of the embryos and submitted to the routine procedures for light and electron microscopy (see further). To evaluate the permeability of the cerebellar vessels to exogenous protein, three treated and three control embryos received an intracardial injection of horseradish peroxidase (HRP) (see further). Quantitative Evaluation of Mercury Content The glass wares utilized for the following procedures were cleaned with 0.5% nitric acid in an ultrasound bath for 2 h, washed in double distilled water and dried in an oven. The encephala, livers, kidneys and CAMs were separately dried in open glass vials; each sample was kept in an oven for 6 h at 60 2 8C, 10 h at 80 2 8C, 24 h at 110 2 8C, and weighed after cooling at room temperature. The dry samples were placed in screw capped glass vials and digested in 5 ml of an acid mixture of 60% sulphuric acid and 40% nitric acid (Instra-Analized, J.T. Baker, Italy) in a thermostatic bath at 45 0:5 8C for 24 h to obtain the complete dissolution of the tissues. Thereafter, the samples were cooled and reconstituted to 50 ml with double distilled water. The quantitative analysis of the mercury content was performed by cold vapour atomic absorption
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857
spectrophotometry with a Perkin–Elmer 5100 ZL spectrophotometer. The quantification was performed by testing five aliquots of samples (1–2 ml) and adding 0, 20, 30, 40 and 60 ng of Hg/sample, respectively. This method of quantification was chosen instead of quantification by an external standard to reduce the matrix effects on the Hg signal. The instrumental parameters were set as indicated according to the operation manual of the instrument used for the mercury analysis. Mercury standard solution and MeHgCl were purchased from Sigma–Aldrich (Milan, Italy). Light and Electron Microscopy Cerebellar lamellae were fixed in Bouin’s solution and embedded in paraffin; the histological sections were stained with hematoxylin-eosin or with toluidine blue. Small fragments of cerebellar lamellae were fixed in 3% phosphate-buffered glutaraldehyde, embedded in Epon 812 and cut in ultrathin sections (60 nm) with an ultramicrotome. The sections were double-stained with uranyl acetate and lead citrate. HRP Assay Three treated and three control embryos received an intracardial injection of HRP (0.3 mg Sigma type II HRP/g body weight in 0.1–0.3 ml saline solution). The embryos were sacrificed 10 min after the injection and the cerebella were fixed in a sodium-cacodylate buffered mixture of 2% glutaraldehyde and 2% paraformaldehyde. Thick sections (50 mm), cut with a vibratome, were incubated in a 0.05% solution of 30 ,30 diaminobenzidine in 0.05 M Tris–HCl buffer containing 0.01% H2O2. After incubation, the thick sections were dehydrated and mounted on slides; the peroxidase activity was apparent at the light microscope as a brown reaction product (Roncali et al., 1986). Light microscopy sections were photographed with a Vanox-T (Olympus) microscope equipped with a high resolution CCD/KAI-2000 video camera (SPOT Insight Color, Diagnostic Instruments, MI, USA). The ultrathin sections were photographed under the electron microscope EM109 (Carl Zeiss). Morphometric Analysis Image acquisition and measurement were performed by the image analyser VIDAS 2.5 (Kontron Elektronik GmbH, Eching, Germany) connected either via a highresolution (1600 1200 active pixels) CCD/KAI-2000 video camera to an Axioscop 20 (Carl Zeiss) light
851
microscope, or connected to a CCD/RGB video camera (TK-1070E; JVC, Japan) for electron microphotograph analysis. A semiautomatic/interactive morphometric analysis was applied to measure the number of Purkinje neurons affected by degenerative changes. Randomly chosen fields (50 for treated and 30 for control embryos) were acquired at 40 magnification from toluidine blue stained sections of cerebellar lamellae. On the digitised images, both the total number and the number of damaged Purkinje neurons were counted by a ‘mark points’ interactive measurement function and expressed as percent mean value standard deviation. Randomly chosen vascular fields (200 for treated and 160 for control embryos) were digitised from electron microphotographs of the cerebellar lamellae acquired at 18,000 magnification. The total vessel perimeter and the length of the vessel perimeter contacted by astrocytes were calculated by a ‘trace contour’ interactive function and expressed as mean value standard deviation of the percentage of the vessel perimeter lined by astrocytes. The statistical significance of the obtained values was assessed by the t-test.
RESULTS Quantitative Evaluation of Mercury Content No traces of mercury were present in the encephalon, liver, and kidney of the control embryos, while a mercury content of 0:8 0:12 mg/g dry weight was detected in the CAM. The organs of MeHg-poisoned embryos contained a large amount of mercury; the highest value was detected in the CAM (261:9 8:0 mg/g dry weight) and there were virtually similar amounts in the encephalon (75:6 9:8 mg/g dry weight), liver (76:4 9:2 mg/g dry weight), and kidney (79:2 7:8 mg/g dry weight). Light and Electron Microscopy Compared with the controls (Fig. 1a and b), the cerebellar lamellae of MeHg-exposed embryos (Fig. 1c and d) appeared hypotrophic and characterized by widespread cellular damage. In particular, the outer granules were arranged in a discontinuous layer, where a number of cells displayed pyknotic nuclei (Fig. 1d). A statistically significant percentage of Purkinje neurons (59:19 9:20% versus 3:78 0:78 in the controls, P ¼ 0:001) was affected by degenerative changes
852
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857
Fig. 1. Lamellae and details of the cerebellar cortex of control (a and b) and MeHg-treated (c and d) embryos. Note, in the treated embryo, the reduced size of the cerebellar lamella (c compared with a), as well as the disarrangement and irregular thickness of the external granular layer (EGL) (d compared with b). A number of outer (arrows) and inner (double arrow) granules, as well as Purkinje neurons (arrowheads), display nuclear pyknosis (d compared with b). Purkinje neurons featuring chromatolysis and an unstained cytoplasm are also present (d, double arrowheads). In the microvessels of a control embryo cerebellum, the permeability marker HRP has an intravascular localization (e). In a MeHg-treated embryo, HRP is widely diffused throughout the cerebellar neuropil (f). Bars: (a and c) 100 mm; (b and d) 50 mm; (e and f) 25 mm.
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857
853
Fig. 2. Cytoplasmic and nuclear alterations affecting two Purkinje neurons of a MeHg-treated embryo (a). Cerebellar microvessel of a control embryo displaying a regular course of the inner and outer surfaces of the vessel wall and extensive tight junctions (b, double arrowheads) joining the endothelial cells (E). The vessel wall is bordered by a continuous basal lamina (b, arrowheads) and astrocytic end feet (A) linked by gap junctions (b, double arrow). The microvessels of MeHg-treated embryos (c and d) show luminal expansions of the endothelial cells (E, arrows) and pericyte processes (d, P) extending toward the neuropil. The endothelial cells contain numerous vesicles and vacuoles, and are linked by short junctions (c, double arrowhead). The basal lamina enwraps only part of the vessel (c, arrowheads) or is entirely absent (d). The vessel shown in (c) is completely devoid of perivascular glia. Only two small astrocytic end feet (A) are appreciable in (d). Bars: (a) 2 mm and (b–d) 1 mm.
854
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857
consisting of cellular shrinkage, pyknosis, chromatolysis, or a scarce amount of Nissl’s substance (Figs. 1d and 2a). Neurons containing pyknotic nuclei were also recognizable in the inner granular layer (Fig. 1d). Unlike the cerebellum of control embryos, where the permeability marker HRP was retained within the microvessels (Fig. 1e), the cerebellum of all treated embryos displayed massive extravasation of the marker throughout the neuropil (Fig. 1f). By the electron microscope, the cerebellar microvessels of control embryos (Fig. 2b) showed a regular course of the inner and outer profiles of the endothelium-pericyte layer. The endothelial cells were sealed by extensive tight junctions and contained a low number of pinocytotic vesicles. The outer surface of the vessel wall was bordered by a continuous, relatively thick, basal lamina and by an uninterrupted layer of astrocytic end feet, joined by gap junctions and containing clusters of glycogen granules. In the treated embryos, on the contrary, the cerebellar microvessels (Fig. 2c and d) displayed irregular profiles, mainly at the luminal front, a high number of endothelial vesicles and vacuoles, and poorly developed interendothelial junctions; the nucleus and cytoplasmic organelles of endothelial cells and pericytes appeared structurally normal. In a number of microvessels, the basal lamina was discontinuous (Fig. 2c) or entirely lacking (Fig. 2d). Astrocyte bodies and processes were extremely rare, both in the neuropil and near the microvessels (Fig. 2c and d). The quantitative evaluation of the perivascular glia extension showed that astrocytic end feet covered 6:24 1:36% of the endotheliumpericyte layer in the treated embryos versus 96 3:4% in the control embryos (statistical significance P ¼ 0:001). No cellular debris was seen around the vessel walls.
DISCUSSION AND CONCLUSIONS The results of the quantitative evaluations, indicating high mercury concentrations in the brain, liver and kidney, demonstrate the efficacy of our experimental protocol, since these organs are known to be the main sites of mercury storage (Crinnion, 2000; Nagashima, 1997; Opitz et al., 1996; Sakamoto et al., 1998; WHO, 1993). The morphological observations demonstrate severe cellular sufferance in the cerebellum of treated embryos and indicate that the neurons most severely affected by MeHg are the cerebellar granules and Purkinje neurons. MeHg-induced damage to cerebellar
granules has been documented in several in vitro and in vivo studies, one of which suggests that granule cell death depends on MeHg-impairment of cell migration toward the inner granular layer. Instead, the Purkinje neurons of adult humans as well as of neonatal and adult rats appear to be relatively spared, or simply disarranged, under conditions of chronic MeHg exposure (Eto, 1997; Kunimoto, 1994; Kunimoto and Suzuki, 1997; Nagashima, 1997; Sakamoto et al., 1998, 2002; Wakabayashi et al., 1995). The discrepancy between these results and the extensive damage affecting the Purkinje neurons in the chick embryo corroborates the finding that sensitivity to MeHg varies according to both animal species and age (Berlin, 1986; Burbacher et al., 1990; Kakita et al., 2000a,b; Wakabayashi et al., 1995; Willes, 1977; WHO, 1993) and further demonstrates the higher risk of exposure to this compound during development. In this context, it must be remembered that the MeHg treatment in the chick embryo coincided temporally with crucial events of the cerebellar morphohistogenesis, such as differentiation of granule neurons, and their migration toward the inner granular layer, differentiation of Purkinje and astroglial cells, as well as arrangement of astrocytic end-feet around the growing microvessels (Bertossi et al., 1986, 1997; Mugnaini, 1969; Mugnaini and Forstrønen, 1967; Virgintino et al., 1993). Many in vitro studies have proposed different mechanisms through which MeHg exerts cytotoxic effects, such as oxidative stress, apoptosis, necrosis, altered transport of glutamate, and inhibition of mitosis via microtubule disruption (Allen et al., 2002; Amorim et al., 2000; Aschner, 2002; Aschner et al., 2000; Castoldi et al., 2001; Dare´ et al., 2001; Miura et al., 1999). Considering the widespread cellular damage observed in the cerebellum of treated embryos, it is reasonable to assume that MeHg may exert its action on the neuronal and glial populations through one or more of these mechanisms. In particular, the reduction of astrocytes could be related to impairment of proliferation of their precursor cells rather than to astroglia degeneration, since no cellular debris were observed in the MeHg-exposed cerebellum. The strong sensitivity to MeHg displayed by astroglial cells is in agreement with the results of investigations performed in vitro (Allen et al., 2002; Aschner, 2002; Aschner et al., 2000) and suggests that in vivo, and during development, astroglial cells are even more sensitive than neurons to MeHg exposure. It is well known that astrocytes play active roles in the development and function of the CNS. Thanks to the presence of gap junctions, they work as a
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857
functional syncytium that maintains the physiological composition of the extracellular fluid and, among other actions, preserves the neurons from glutamatemediated excitotoxicity (LoPachin and Aschner, 1993; Aschner and Kimelberg, 1996; Barres and Barde, 2000; Ransom et al., 2003). Therefore, it seems conceivable that MeHg-related neuronal sufferance may be secondary to, or at least enhanced by, astrocytic damage (Allen et al., 2002; Aschner, 2002; Brookes, 1992). The ultrastructural features and permeability to the marker HRP shown at hatching by the cerebellar microvessels of MeHg-poisoned embryos are similar to those observed in normally developed embryos at earlier stages (Bertossi et al., 1997, 2003). These findings indicate that MeHg exposure does not cause structural damage to endothelial cells and pericytes, but does lead to significantly delayed maturation of the vessels and imperfect acquirement of their barrier properties. The relationship between absence of perivascular astrocyte end feet and immature morphofunctional features of the cerebellar microvessels likely represents a further indication of the pivotal role played by astroglial cells in the proper differentiation of BBBprovided vessels. Taken as a whole, our observations strongly suggest that astrocyte loss-dependent impairment of the BBB is consistently involved in the developmental neurotoxicity of MeHg.
ACKNOWLEDGEMENTS The work was supported in part by grants from University of Foggia to M. Bertossi (60%, 2001, 2002). We thank Ms M.V.C. Pragnell, BA, for linguistic help and Ms M. Ambrosi and S. Cantatore for their skilful technical assistance.
REFERENCES Akiyama H, Kondoh T, Kokunai T, Nagashima T, Saito M, Tamaki N. Blood-brain barrier formation of grafted human umbilical vein endothelial cells in athymic mouse brain. Brain Res 2000;858:172–6. Allen JW, Shanker G, Tan KH, Aschner M. The consequences of methylmercury exposure on interactive functions between astrocytes and neurons. Neurotoxicology 2002;23:755–9. Amorim MLM, Mergler D, Bahia MO, Dubeau H, Miranda D, Lebel J, Burbano RR, Lucotte M. Cytogenetic damage related to low levels of methylmercury contamination in the Brazilian Amazon. An Acad Bras Ci 2000;72:497–507.
855
Andersen HR, Nielsen JB, Grandjean Ph. Toxicologic evidence of developmental neurotoxicity of environmental chemicals. Toxicology 2000;144:121–7. Aschner M. Neurotoxic mechanism of fish-borne methylmercury. Environ Toxicol Pharmacol 2002;12:101–4. Aschner M, Kimelberg AK, The role of glia in neurotoxicity. Boca Raton: CRC Press; 1996. Aschner M, Yao CP, Allen JW, Tan KH. Methylmercury alters glutamate transport in astrocytes. Neurochem Int 2000;37: 199–206. Barres BA, Barde Y-A. Neuronal and glial cell biology. Curr Opin Neurobiol 2000;10:642–8. Berlin M. Mercury. In: Friberg L, Nordberg GF, Vouk V, editors. Handbook on the toxicology of metals, vol. II. Amsterdam: Elsevier; 1986, p. 387–445. Bertossi M, Roncali L, Mancini L, Ribatti D, Nico B. Process of differentiation of cerebellar Purkinje neurons in the chick embryo. Anat Embryol 1986;175:25–34. Bertossi M, Roncali L, Nico B, Ribatti D, Mancini L, Virgintino D, Fabiani G, Guidazzoli A. Perivascular astrocytes and endothelium in the development of the blood-brain barrier in the optic tectum of the chick embryo. Anat Embryol 1993; 188:21–9. Bertossi M, Virgintino D, Roncali L, Riva A. Embryo cerebellum microvessels under scanning electron microscopy. In: Motta PM, editor. Recent advances in microscopy of cells, tissues and organs. Antonio Delfino Editore: Rome; 1997. p. 215–8 Bertossi M, Virgintino D, Errede M, Roncali L. Immunohistochemical and ultrastructural characterization of cortical plate microvasculature in the human fetus telencephalon. Microvasc Res 1999;58:49–61. Bertossi M, Girolamo F, Errede M, Benagiano V, Virgintino D, Roncali L. Developmental changes of HT7 expression in the microvessels of the chick embryo brain. Anat Embryol 2002; 205:229–33. Bertossi M, Girolamo F, Errede M, Virgintino D, Roncali L. Effects of the 6-aminonicotinamide gliotoxin on the blood-brain barrier differentiation in the chick embryo cerebellum. Anat Embryol 2003;207:209–19. Brookes N. In vitro evidence for the role of glutamate in the CNS toxicity of mercury. Toxicology 1992;76:245–56. Burbacher TM, Rodier PM, Weiss B. Methylmercury developmental neurotoxicity: a comparison of effects in humans and animals. Neurotoxicol Teratol 1990;12:191–202. Castoldi AF, Coccini T, Ceccatelli S, Manzo L. Neurotoxicity and molecular effects of methylmercury. Brain Res Bull 2001;55: 197–203. Charleston JS, Body RL, Bolender RP, Vahter ME, Burbas TM. Changes in the number of astrocytes and microglia in the thalamus of the monkey Macaca fascicularis following longterm subclinical methylmercury exposure. Neurotoxicology 1996;17:127–38. Choi BH, Lapham LW, Amin-Zaki L, Saleem T. Abnormal neuronal migration, deranged cerebral cortex organization, and diffuse white matter astrogliosis of human fetal brain: a major effect of methylmercury poisoning in utero. J Neuropathol Exp Neurol 1978;37:719–33. Crinnion WJ. Environmental medicine. Part 3. Long-term effects of chronic low-dose mercury exposure. Altern Med Rev 2000; 5:209–23.
856
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857
Dare´ E, Li W, Zhivotovsky B, Yuan X, Ceccatelli S. Methylmercury and H2O2 provoke lysosomal damage in human astrocytoma D384 cells followed by apoptosis. Free Radic Biol Med 2001;30:1347–56. Errede M, Elia G, Bertossi M, Mulas ML, Riva A, Virgintino D, Benagiano V, Girolamo F, Roncali L, Ambrosi L. Developmental effects of lead acetate on the chick embryo metanephros. J Submicrosc Cytol Pathol 2001;33:261–73. Eto K. Pathology of Minamata disease. Toxicol Pathol 1997;25: 614–23. Garman RH, Weiss B, Evans HL. Alkylmercurial encephalopathy in the monkey (Saimiri sciureu and Macaca arctoides): a histopathologic and autoradiographic study. Acta Neuropathol (Berl) 1975;32:61–4. Grandjean P, Weihe P, White RF, Debes F, Araki S, Yohoyama K, Murata K, Sorensen N, Dahl R, Jorgensen PJ. Cognitive deficit in 7-year-old-children with prenatal exposure to methylmercury. Neurotoxicol Teratol 1997;19:417–28. Harada M. Congenital Minamata disease: intrauterine methylmercury poisoning. Teratology 1978;18:285–8. Hayashi Y, Nomura M, Yamagishi SL, Harada SJ, Yamashita J, Yamamoto H. Induction of various blood-brain barrier properties in non-neural endothelial cells by close apposition to co-cultured astrocytes. Glia 1997;19:13–26. Hunter D, Russell DS. Focal cerebral and cerebellar atrophy in a human subject due to organic mercury compounds. J Neurol Neurosurg Psychiatr 1954;17:235–41. Hurwitz AA, Berman JW, Rashbaum WK, Lyman WD. Human fetal astrocytes induce the expression of blood-brain barrier specific proteins in autologous endothelial cells. Brain Res 1993;625:238–43. Irukayama K, Kai F, Fujiki M, Kondo T. Study on the origin of causative agent of Minamata disease. III. Industrial waste containing mercury compounds from Minamata factory. Kumamoto Med J 1962;15:57–68. Janzer RC, Raff MC. Astrocytes induce blood-brain barrier properties in endothelial cells. Nature 1987;325:253–7. Jensen S, Jernelov A. Biological methylation of mercury in aquatic organisms. Nature 1969;223:753–4. Kakita A, Wakabayashi K, Su M, Yoneoka Y, Sakamoto M, Ikuta F, Takahashi H. Intrauterine methylmercury intoxication. Consequence of the inherent brain lesions and cognitive dysfunction in maturity. Brain Res 2000a;877:322–30. Kakita A, Wakabayashi K, Su M, Sakamoto M, Ikuta F, Takahashi H. Distinct pattern of neuronal degeneration in the fetal rat brain induced by consecutive transplacental administration of methylmercury. Brain Res 2000;859:233–9. Kjellstrom T, Kennedy P, Wallis S, Mantell C. Physical and mental development of children with prenatal exposure to mercury from fish. Stage I: preliminary tests at age 4. Solna (Sweden): National Swedish Environmental Protection Board Report; 1986. p. 3080. Kjellstrom T, Kennedy P, Wallis S, Stewart A, Friberg L. Physical and mental development of children with prenatal exposure to mercury from fish: Stage II: interviews and psychological tests at age 6. Solna (Sweden): National Swedish Environmental Protection Board Report; 1989. p. 3642. Kuchler-Bopp S, Delaunoy JP, Artault JC, Zaepfel M, Dietrich JB. Astrocytes induce several blood-brain barrier properties in non-neural endothelial cells. Neuroreport 1999;10:1347– 53.
Kunimoto M. Methylmercury induces apoptosis of rat cerebellar neurons in primary culture. Biochem Biophys Res Commun 1994;204:310–7. Kunimoto M, Suzuki T. Migration of granule neurons in cerebellar organotypic cultures is impaired by methylmercury. Neurosci Lett 1997;226:183–6. LoPachin RM, Aschner M. Glial-neuronal interactions: the relevance to neurotoxic mechanisms. Toxicol Appl Pharmacol 1993;118:141–58. Mahaffey KR. Recent advances in recognition of low-level mercury poisoning. Curr Opin Neurol 2000;13:699–707. Mansour MM, Dyer NC, Hoffman LH, Davies J, Brill AB. Placental transfer of mercuric nitrate and methylmercury in the rat. Am J Obstet Gynecol 1974;119:557–62. Miura K, Koide N, Himeno S, Nagakawa I, Imura N. The involvement of microtubular disruption in methylmercury induced apoptosis in neuronal and nonneuronal cell lines. Toxicol Appl Pharmacol 1999;160:279–88. Mugnaini E, Ultrastructural studies on the cerebellar histogenesis. II. Maturation of nerve cell populations and establishment of synaptic connections in the cerebellar cortex of the chick. In: Llina`s RR, editor. Neurobiology of cerebellar evolution and development. Chicago: AMA Educational and Research Foundation; 1969. p. 749–82. Mugnaini E, Forstrønen PF. Ultrastructural studies on the cerebellar histogenesis. I. Differentiation of granule cells and development of glomeruli in the chick embryo. Z Zellforsch 1967;77:115–43. Myers GJ, Davidson PhW. Does methylmercury have a role in causing developmental disabilities in children? Environ Health Perspect 2000;108:413–20. Nagashima K. A review of experimental methylmercury toxicity in rats: Neuropathology and evidence for apoptosis. Toxicol Pathol 1997;25:624–31. Opitz H, Schweinsberg F, Grossmann T, Wendt-Gallitelli MF, Meyermann R. Demonstration of mercury in the human brain and other organs 17 years after metallic mercury exposure. Clin Neuropathol 1996;15:139–44. Pirrone N. Mercury research in Europe: towards the preparation of the EU air quality directive. Atmos Environ 2001;35: 2979–86. Ransom B, Behar T, Nedergaard M. New roles for astrocytes (stars at last). Trends Neurosciences 2003;26:620–2. Roncali L, Nico B, Ribatti D, Bertossi M, Mancini L. Microscopical and ultrastructural investigations on the development of the blood-brain barrier in the chick embryo optic tectum. Acta Neuropathol (Berl) 1986;70:193–201. Sakamoto M, Wakabayashi K, Kakita A, Takahashi H, Adachi T, Nakano A. Widespread neuronal degeneration in rats following oral administration of methylmercury during the postnatal developing phase: a model of fetal-type Minamata disease. Brain Res 1998;784:351–4. Sakamoto M, Kakita A, Wakabayashi K, Takahashi H, Nakano A, Akagi H. Evaluation of changes in methylmercury accumulation in the developing rat brain and its effects: a study with consecutive and moderate dose exposure throughout gestation and lactation periods. Brain Res 2002;949:51–9. Steuerwald U, Weihe P, Jorgensen PJ, Bjerve K, Brock J, Heinzow B, Budtz-Jorgensen E, Grandjean P. Maternal seafood diet, methylmercury exposure, and neonatal neurologic function. J Pediatr 2000;136:599–605.
M. Bertossi et al. / NeuroToxicology 25 (2004) 849–857 Stewart PA, Wiley MJ. Developing nervous tissue induces formation of blood-brain barrier characteristics in invading endothelial cells. Dev Biol 1981;84:183–92. Takeuchi T, Morikawa N, Matsumoto H, Shiraishi A. A pathological study of Minamata disease. Acta Neuropathol 1962;2:40–57. Virgintino D, Maiorano E, Bertossi M, Pollice L, Ambrosi G, Roncali L. Vimentin- and GFAP-immunoreactivity in developing and mature neural microvessels. Study in the chicken tectum and cerebellum. Eur J Histochem 1993;37:353–62. Virgintino D, Maiorano E, Errede M, Vimercati A, Greco P, Selvaggi L, Roncali L, Bertossi M. Astroglia-microvessel relationship in the developing human telencephalon. Int J Dev Biol 1998;42:1165–8. Virgintino D, Robertson D, Benagiano V, Errede M, Bertossi M, Ambrosi G, Roncali L. Immunogold cytochemistry of the blood-brain barrier glucose transporter GLUT1 and endogenous albumin in the developing human brain. Dev Brain Res 2000;123:95–101.
857
Wakabayashi K, Kakita A, Sakamoto M, Su M, Iwanaga K, Ikuta F. Variability of brain lesions in rat administered methylmercury at various postnatal developmental phases. Brain Res 1995;705:267–72. Wakai S, Hirokawa N. Development of the blood-brain barrier to horseradish peroxidase in the chick embryo. Cell Tiss Res 1978;195:195–203. Watanabe C, Sato H. Evolution of our understanding of methylmercury as a health threat. Environ Health Perspect 1996;104:367–79. Willes RF. Tissue distribution as a factor in species susceptibility to toxicity and hazard assessment. Example: methylmercury. J Environ Pathol Toxicol 1977;1:135–46. WHO. Evaluation of carcinogenic risks to humans, vol. 58; 1993. p. 2339–45 Wood JM, Kennedy FS, Rosen CE. Synthesis of methylmercury compounds by extracts of methanogenic bacterium. Nature 1968;220:173–4.