- Email: [email protected]
Mercury Distribution in the Neonatal and Adult Cerebellum after Mercury Vapor Exposure of Pregnant Squirrel Monkeys
Environmental Research Section A 83, 93}101 (2000) Article ID enrs.1999.4013, available online at http://www.idealibrary.com on
Mercury Distribution in the Neonatal and Adult Cerebellum after Mercury Vapor Exposure of Pregnant Squirrel Monkeys1 Karin Warfvinge2 Department of Ophthalmology, University Hospital of Lund, S-221 85 Lund, Sweden, and University College of Health Sciences, S-551 11 JoK nkoK ping, Sweden Received July 2, 1999
amounts of mercury, but the cerebellar nuclei accumulated the highest amounts of mercury. No correlation was found between cellular accumulation and maturity of the brain; that is, the cellular localization of mercury did not differ between adult and neonatal brain, except for the amount of visualized mercury. This pattern corresponded well to the mercury concentrations found in the cerebral occipital pole. The differences found in mercury accumulation were instead considered to be doserelated. The results demonstrate that the distribution of mercury in the cerebellum after mercury vapor exposure is similar to the distribution pattern obtained after methyl mercury exposure and that mercury is trapped in the cerebellum over a long period of time. 2000 Academic Press Key Words: squirrel monkeys; mercury accumulation; developmental toxicity; mercury vapor; cerebellum.
The objectives of the study were (1) to map the detailed localization of mercury in the monkey cerebellum after mercury vapour exposure; (2) to investigate whether there is any difference in mercury distribution between neonatal and adult cerebellum after mercury vapor exposure; (3) to investigate the ability of mercury to accumulate in the cerebellum years after the end of exposure. Pregnant squirrel monkeys were exposed 5 days/ week to mercury vapor at a concentration of 0.5 mg Hg/m3 air 4 or 7 h/day or 1 mg Hg/m3 air for 4 or 7 h/day. Mercury concentration in the offspring and maternal brains was examined by cold vapor, Bameless atomic absorption spectrophotometry. Mercury distribution was examined by processing cerebellar sections for autometallographic (AMG) silver enhancement. Mercury concentration in the offspring cerebral occipital pole ranged between 0.20 and 0.70 lg Hg/g tissue, and in the maternal between 0.80 and 2.58 lg/Hg tissue in animals killed immediately after the end of exposure. AMG revealed that the external granule cell layer of offspring cerebellar tissue contained small amounts of mercury. The molecular layer contained mercury in some of the mercury-exposed monkeys. In the Purkinje cell layer, the Bergmann glial cells together with the Purkinje cells contained mercury. The granule cells and the Golgi cells contained small amounts of mercury. The astrocytes of the medullary layer, identiAed by immunohistochemistry, contained considerable
INTRODUCTION
The toxic properties of mercury vapor are mainly due to the fact that mercury accumulates in the brain, causing neurological signs at high exposure. Mercury vapor also penetrates the placental barrier and enters the fetus and its brain (Clarkson et al., 1972; Warfvinge et al., 1994) and retina (Warfvinge and Bruun, 2000). At short-term mercury vapor exposure of rats (500 lg Hg/m3 air for 4 h), mercury is found in only a few areas of the brain, including the cerebellar deep nuclei (MoK ller-Madsen, 1992). Other studies have also shown that the cerebellar nuclei accumulate mercury after mercury vapor exposure (Cassano et al., 1966, 1969; Berlin et al., 1969; Warfvinge et al., 1992; Warfvinge, 1995). Moreover, the Purkinje cells accumulate mercury when the dose is increased. Since the outbreaks of methyl mercury neuropoisoning in Japan in the 1950s (Harada, 1968) and
This paper was presented at Mercury as a Global Pollutant: 5th International Conference, Rio de Janeiro, Brazil, May 23}28, 1999. 1 This work was supported by the NIH, the Swedish Medical Research Council, the Scienti7c Board of JoK nkoK ping County Council and the Foundations of Crafoord, Maggie Stephens, and Vas rdal, Sweden. The experiments were conducted in accordance with national and institutional guidelines for the protection of animal welfare. 2 Address for correspondence: Department of Ophthalmology, University Hospital of Lund, S-221 85 Lund, Sweden. Fax: (46) 46 1 72721. E-mail: [email protected]. 93
0013-9351/00 $35.00 Copyright 2000 by Academic Press All rights of reproduction in any form reserved.
94
KARIN WARFVINGE
in Iraq in the 1970s (Clarkson et al., 1981), the brain has been known to be most vulnerable during its prenatal development. Case reports from these epidemics indicate that the fetal cerebral cortex and the cerebellum are the main targets for mercury injury, with degeneration of cerebral cortical cells and granule and Purkinje cells of cerebellum. A study of the effects of mercury vapor exposure on the rat cerebellum has shown a degeneration of the Purkinje cell layer combined with gliosis (Hua et al., 1995). This indicates that the cerebellar involvement seen after organic mercury exposure might be seen also after mercury vapor exposure. In addition, mercury vapor has proven to be a potent neuroteratogenic substance (Berlin et al., 1992); prenatal exposure results in migration disturbances similar to those found after prenatal methyl mercury exposure, and, in addition, it results in behavioral disturbances (Newland et al., 1996). However, the brain mercury concentration found is 10 times lower after mercury vapor exposure than after methyl mercury exposure. The objectives of the present study were the following: (1) to map the detailed localization of mercury in the monkey cerebellum after mercury vapor exposure; (2) to investigate whether there is any difference in mercury distribution between neonatal and adult cerebellum after mercury vapor exposure; (3) to investigate the ability of mercury to accumulate in the cerebellum years after the end of exposure. MATERIALS AND METHODS
Animals. Female monkeys with good reproductive records were selected from our breeding colony of squirrel monkeys (Saimiri boliviensis boliviensis; subspecies type veri7ed by karyotype). The females were mated with a male for 1 week every third week to produce timed pregnancies. A few weeks after mating diagnostic abdominal palpations and measurements of fundus height were performed. The pregnant monkeys were then exposed to mercury vapor during the rest of gestation. Brains from unexposed newborn offspring and adult monkeys from our breeding colony served as controls. Details of the housing and diet have been described elsewhere (LoK gdberg et al., 1987). Experimental procedures. The pregnant monkeys were enclosed in a chamber of stainless-steel and glass walls of Rochester type. The volume of the chamber was approximately 2 m3 and it was ventilated with an air 8ow of 2 m3 per hour with temperature (23@C) and humidity (about 60%) controlled.
Before entering the chamber, a part of the air passed a mercury vapor generator. Air was also continuously drawn through a mercury meter (Milton Roy P0408, sensitivity 7 lg/m3) at the outlet of the chamber recording the mercury vapor concentration. Thus, mercury vapor concentration in the chamber was continuously recorded. The calibration was carried out by sampling 10 liters of chamber air into mercury-absorbing acid solution (0.4 M KMnO4 and 2% H2SO4 ). Mercury contents were assayed with the acid of cold vapor atomic absorption spectrophotometry. The method for exposure has earlier been described by Warfvinge et al. (1994). Mercury vapor generation. Mercury vapor was generated by stirring metallic mercury in a 10-liter glass 8ask. The stirring was accomplished with the aid of a magnetic stirrer. The mercury concentration in air was adjusted by adapting the turning speed of the stirring magnet. Further adjustment could be obtained by changing the amount of incoming air through the mercury generator. Mercury exposure. From the groups of pregnant monkeys, exposed to different levels of mercury, described above, the following exposure levels were used in the present study: Group B was exposed to a mercury concentration of approximately 1 mg Hg/m3 air for 7 h/day and Group C for 4 h/day, Group D to a mercury concentration of 0.5 mg Hg/m3 air for 7 h/day and Group E for 4 h/day 5 days/week during gestation. From these groups of exposed monkeys, two offspring (Bo2 and Bo3) brains from group B, three offspring (Co1, Co3, and Co4) brains and three maternal (C2, C3, and C4) brains from group C, four maternal (D1, D2, D5, and D6) brains from group D, and two brains (E3 and E5) from Group E were used for the present study (Table 1). Two monkeys (C4 and D1) died and one monkey (D2) was sacri7ced at the end of exposure. Four monkeys (D5, D6, E3, and E5) were sacri7ced 1 year and two monkeys (C2 and C3) 3 years after the end of exposure. The ages of the offspring ranged between 18 weeks gestational age and 2 days. Mercury analysis. The total mercury content in the cerebral occipital pole was determined by a modi7ed method described by Einarsson et al. (1984). Brie8y, homogenized brain samples from the right occipital pole were digested with a mixture of HClO and HNO3 overnight at 68@C. The procedure was modi7ed for higher sensitivity by enrichment (amalgamation) of the mercury vapor on a gold-wire 7lter. Chemical blanks and reference samples (Seronorm Trace Elements, Batch 904, Nycomed AS, Oslo,
95
Hg IN ADULT AND NEONATAL MONKEY CEREBELLUM
TABLE 1 Exposure Indices for Squirrel Monkeys Exposed to Mercury Vapor during Pregnancy
Exposure duration
Calculated Hg absorption
External Hg dose
Monkey No. (Time lapse between exposure and 7xation)
Weeks of gestational age
No. of days with exposure
Approx. concentration (mg/m3)
Hours per day, with exposure
Total hours of exposure
B2 B3 C1 C2 (3 years) C3 (3 years) C4 (directly) D1 (directly) D2 (1 month) D5 (1 year) D6 (1 year) E3 (1 year) E5 (1 year)
5}18 5}22 7}21 7}21 7}20 7}23 5}9 6}21 5}10 5}8 7}20 5}21
66 84 65 69 59 72 20 65 38 18 65 79
1 1 1 1 1 1 0.5 0.5 0.5 0.5 0.5 0.5
7 7 4 4 4 4 7 7 7 7 4 4
422 496 263 283 239 287 137 400 251 125 245 323
Norway) were included in the sample series. Mercury was determined by cold vapor, 8ameless atomic absorption spectrophotometry. The mercury concentrations in the maternal and offspring brains and the calculated mercury absorption are presented in Table 1. Mercury absorption was calculated as follows: (mercury concentration in the chamber);(time of exposure);(alveolar ventilation of squirrel monkeys);80%. The mercury concentration was measured by the mercury meter as described above, the alveolar ventilation of monkeys is 13 liters/h (Rosenblum and Cooper, 1968) and 80% absorption is assumed, as has been shown for man (Hursh et al., 1976). Preparation of tissue. Three of the offspring (Bo2, Co1, and Co4) and seven adult monkeys (C2, C3, D2, D5, D6, E3, and E5) were anesthetized with barbiturate (Mebumal) and the CNS was 7xed by perfusion with 4% formaldehyde in PBS. The two other mercury-exposed offspring (Bo3 and Co3) were found dead and their brains were removed and placed in 4% formaldehyde in PBS. Two mercuryexposed adult monkeys (C4 and D1) died and the brains from these monkeys were also placed in 4% formaldehyde in PBS. Perfusion-7xed brains from two unexposed offspring, immersion-7xed brains from three unexposed offspring and eight unexposed adult monkeys were used as control brains. Two-to
Per day Total (lg) (lg/day) 4564 5402 2693 2901 2492 3020 768 2497 1462 685 1378 1585
69 64 41 42 42 42 38 38 38 38 21 20
Hg concentration in the occipital pole
Mother (lg/g tissue)
Offspring (lg/g tissue) 0.21 0.70 0.20
2.58 0.80 1.29
0.24 0.30
three-millimeter-coronal slices of the cerebellum were then dissected out. The slices were dehydrated and embedded in paraf7n. Five-micrometer-thick sections were cut and placed on glass slides and further processed for visualization of mercury using the autometallographic method described by Danscher and MoK llerMadsen (1985). Brie8y, the slides were deparaf7nized, coated with gelatin, and developed at 26@C fr 1 h in a dark box (the developer contained gum arabic, sodium citrate buffer, hydroquinone, and silver lactate). The gelatin was washed out in 40@C tap water followed by distilled water, and the slides were then exposed to 5% sodium thiosulfate and counterstained with hematoxylineosin. The method has been shown, in several reports by Danscher and MoK ller-Madsen (Danscher and MoK ller-Madsen, 1985; MoK ller-Madsen and Danscher, 1986, 1991; MoK ller-Madsen, 1990, 1991, 1992), to be speci7c for metallic gold and inorganic mercury. To exclude the possibility that other metals could have acted as catalysts for the reduction, sections from all monkeys were treated with potassium cyanide or acid according to Danscher and MoK ller-Madsen (1985) and Danscher and Rungby (1986). Additionally, immunohistochemistry was carried out by an avidin-biotin complex technique on one section from each monkey. A bovine polyclonal antibody raised against human glial 7brillary acidic
96
KARIN WARFVINGE
protein (GFAP) was purchased from Dakopatts AB in Stockholm. At microscopy, at least 10 sections from each block were examined for mercury content. The amount of visualized mercury for each cell type was estimated according to a three-grade scale: #, a few grains in the cytoplasm; ##, a moderate amount of grains in the cytoplasm; ###, the cytoplasm was loaded with grains. RESULTS
Mercury exposure. The duration of exposure and the estimated absorbed amount of mercury are presented in Table 1. The exposure started between 5 and 7 weeks of gestational age and ended between 18 and 23 weeks of gestational age (normal gestational duration: 152 days), except for monkeys D1, D5, and D6 which were exposed 4}7 weeks. The estimated mercury absorption ranged between 20 and 69 lg Hg/day. Mercury concentration. Concentrations of mercury in the cerebral occipital pole of some of the animals used in the present study are presented in Table 1. The maternal brains always showed higher mercury concentration (0.80}2.58 lg/g tissue) than the offspring brains (0.20}0.70 lg/g tissue), although the external mercury dose was lower for monkeys D1 and D2 than for offspring Bo2, Bo3, Co1, and Co4. Mercury distribution. No signi7cant difference in the results was found between perfusion-7xed and TABLE 2 Mercury Content Estimated on a Three-Grade Scale
Monkey No. External granule cell layer Molecular layer Purkinje cell layer Purkinje cells Bergmann glial cells Granule cell layer Granule cells Golgi cells Medullary layer Astrocytes Other glial cells Deep cerebellar nuclei Neurons Glial cells a
D5, D6, Bo2, C2, C3 E3, E5 Bo3
Co1, Co3, Co4
C4, D2
D1
;
;
;
;
#a
#a
##
##
!
!
#
#a
## ##
# #
## ##
# #
! !
! !
#a #
! #a
# !
# !
## #a
#a !
### ## ### ## # #
! !
### ## # !
### ### ## ### ## #
Some cells contained mercury.
# #
### ## ## #a
FIG. 1. (a) An adult control monkey. No silver enhancement is observed. Large arrows point at Purkinje cells and small arrows at Bergmann glial cells. Scale bar 30 lm. (b) A cerebellar nucleus of an adult control monkey. No visualized mercury is found. Arrows point at large neurons. Scale bar, 30 lm. Abbreviations for all 7gures: ext, external layer; mol, molecular layer; Pur, Purkinje cell layer; gran, granule cell layer.
immersion-7xed material, nor in the potassium cyanide- and acid-treated sections. No silver-enhanced mercury was detected in the brains of the control animals. The results of the examination of cerebellum from 7ve prenatally mercury-exposed monkey offspring and nine mercury-exposed adult monkeys are presented in Table 2 and Figs. 1}3. Immunohistochemistry using the GFAP antibody revealed that the positive cells of the medullary layer were similar in numbers and morphology to those that contained silver grains. The microscopical examination of sections stained on at least 10 different occasions revealed a uniform staining intensity within each cell type and each animal.
Hg IN ADULT AND NEONATAL MONKEY CEREBELLUM
97
FIG. 2. (a) The external granule cell layer from monkey offspring Co4. Mercury deposits are seen in the epithelium covering the folia (large arrows) and in the wall of a vessel (small arrows). Scale bar, 30 lm. (b) Offspring Bo3. This micrograph demonstrate mercury accumulation in the molecular layer (mol), the Bergmann glial cells (small arrows), the Purkinje cells (large arrows), and some cells of the granule cell layer (gran). Arrowhead points at a Golgi cell containing a few silver grains. Scale bar, 30 lm. (c) The medulla of offspring Bo2. Arrows point at mercury-containing astrocytes, which were identi7ed with GFAP immunohistochemistry. Scale bar, 30 lm. (d) A cerebellar nucleus of monkey Co4. Large arrows point at mercury-containing neurons, small arrows at glial cells, and arrowheads at dilated blood vessels that have accumulated mercury in their walls. Scale bar, 30 lm.
DISCUSSION
Cerebellar mercury distribution revealed by autoradiography after administration of mercury compounds has been studied in mice (Berlin and Ullberg, 1963a, b, c; Cassano et al., 1966, 1969; Rodier and Kates, 1988), rats (Berlin et al., 1969; Cassano et al., 1969; Chang and Hartmann 1972a, b), guinea pigs (Nordberg and Serenius, 1969), rabbits (Berlin et al., 1969), and monkeys (Berlin et al., 1969). In a series of reports on rats, MoK ller-Madsen (1990, 1991, 1992) and MoK ller-Madsen and Danscher (1986, 1991) have demonstrated at the cellular level
that the deep cerebellar nuclei accumulate the earliest amount of mercury, irrespective of the mercury compound administered. After mercury vapor exposure, the Purkinje cells and the granule cells are the next target for mercury accumulation (MoK llerMadsen, 1992; Warfvinge et al., 1992). In monkeys, Berlin et al. (1969) have described mercury accumulation in nucleus dentatus, Purkinje cells, and granule cells after mercury vapor exposure. However, the autoradiographic study did not reveal the exact cellular distribution. The present investigation demonstrates the cellular distribution of mercury in the cerebellum of the neonatal and adult monkey brain.
98
KARIN WARFVINGE
FIG. 3. (a) Monkey C4. The molecular layer (mol) contains a moderate amount of mercury. Arrowhead points at a cell which has accumulated mercury. Large arrows point at Purkinje cells and small arrows at Bergmann glial cells in the Purkinje cell layer (Pur). A few silver grains are found in the granule cell layer (gran). Scale bar, 30 lm. (b). The medullary layer of monkey C4. Arrows point at mercury-containing astrocytes, which were identi7ed with GFAP immunohistochemistry. Scale bar, 30 lm. (c) A cerebellar nuclei of monkey D1. Arrows point at large, mercury-containing neurons. Scale bar, 30 lm.
In the present study, neurons of the deep cerebellar nuclei accumulated mercury in all mercury-exposed monkeys examined. This is in accordance with previous studies on mercury vapor exposure (Berlin et al., 1969; Cassano et al., 1966, 1969; MoK ller-Madsen, 1992; Warfvinge et al., 1992; Warfvinge, 1995). These nuclei belong to the motor system, and a causal relationship between the localization and the neurological symptoms induced by mercury intoxication has been suggested by MoK ller-Madsen (1990). In the present study, the Purkinje cells and the Bergmann glial cells contained visualized mercury
in all monkeys examined. The Purkinje cells have previously been described as being targets for mercury accumulation after mercury vapor exposure (MoK ller-Madsen, 1992; Warfvinge et al., 1992; Warfvinge, 1995), in monkeys after only a 4-h exposure to 1 mg Hg/m3 (Berlin et al., 1969). Accumulation of mercury in the Bergmann glial cells after mercury vapor exposure has not been described, although proliferation of Bergmann glial cells together with Purkinje cell degeneration exists in rats (Hua et al., 1995). Accumulation of mercury in the Bergmann glial cells has, however, been described after exposure to organic mercury (Hargreaves et al., 1985;
Hg IN ADULT AND NEONATAL MONKEY CEREBELLUM
MoK ller-Madsen, 1991; MoK ller-Madsen and Danscher, 1991; Suda et al., 1989). In the study by Hargreaves et al. (1985), it was found that the Bergmann glial cells contained mercury (mercury concentration in cerebellum was 12 lg/g tissue; the inorganic mercury fraction represented less than 1%) during the latent phase, that is, before functional disturbances appeared in the methyl mercury-intoxicated rats. In the symptomatic phases, mercury was also visualized in the Purkinje cells. Similar 7ndings were reported by Suda et al. [brain concentrations: 8.87 (inorganic mercury: 0.21) and 5.54 (0.23) lg/g tissue]. Hence, the distribution of mercury in the Purkinje cells and the Bergmann glial cells of the monkey cerebellum after long-term mercury vapor exposure (brain concentration range: 0.20} 2.58 lg/g tissue) corresponds to the distribution of inorganic mercury after methyl mercury exposure. The organic and inorganic forms of mercury are not transported into the CNS (and its cells) in the same way. Mercury vapor easily diffuses across the blood-brain barrier into the CNS because of its high lipophilicity (Magos, 1967). Within the CNS, it is oxidized by an endogenous hydrogen peroxide catalase system to the divalent cation Hg2>. This oxidation results in 7xation of mercury within the cell by its binding to -SH-containing ligands (Magos, 1967). The transport model for the organic mercury compound;methyl mercury;differs from that of mercury vapor. In the plasma, the high af7nity of methyl mercury to -SH-containing compounds, such as cysteine, results in the methyl mercury-cysteine conjugates that are transported across the bloodbrain barrier into the CNS (Aschner and Clarkson, 1988). Despite the fact that the transport model differs, Purkinje cells and Bergmann glia cells accumulate mercury after both methyl mercury and mercury vapor exposure. Another resemblance between the present study and the one by Hargreaves et al. (1985) is the intense staining of the astrocytes of the medullary layer. However, the mercury deposits in the glial cells were largely absent at 31 days after termination of methyl mercury exposure, indicating that these cells serve as ‘‘buffer function’’. Mercury accumulation in the glial cells of the medullary layer has been reported for rats (Warfvinge et al., 1992) and mice (Warfvinge, 1995) after long-term mercury vapor exposure. The most active glial cell in this respect seemed to be the 7brillous astocyte, as was indicated by the GFAP reaction. The granule cells contained little or no visualized mercury in the monkey brains examined. This
99
is in accordance with previous studies, mentioned above, on mercury vapor exposure. The susceptibility of the granular layer to mercury vapor (Davis et al., 1974) and methyl mercury (Matsumoto et al., 1965) exposure has been reported in humans. We have recently reported on the distribution of mercury in the eye after direct exposure and where the monkeys survived years after the end of exposure (Warfvinge and Bruun, 1996). In that study, it was revealed that mercury was trapped predominantly in the vessel walls and retinal pigment epithelium over a long period of time. This heavy accumulation of mercury in the vessel walls is not obvious in the brain. This indicates that one or more retinal trapping mechanisms exist, not present to nearly the same extent in the brain capillaries, since the monkeys that survived years after the exposure did not show that heavy accumulation in the vessel walls. In conclusion, no correlation was found between cellular accumulation and maturity of the brain in the present study; that is, the cellular localization of mercury did not differ between adult and neonatal brain, except for the amount of visualized mercury. This pattern corresponds well to the mercury concentrations found in the cerebral occipital pole. The reason for the neonatal brains having a lower concentration than the maternal brains might be explained by the fact that the offspring liver contains more mercury than the maternal (unpublished observations). Much of the mercury is accumulated in the neonatal liver, protecting the brain from severe exposure. The differences found in mercury accumulation were instead considered to be doserelated. The results demonstrate that the distribution of mercury in the cerebellum after mercury vapor exposure is similar to the distribution pattern obtained after methyl mercury exposure and that mercury is trapped in the cerebellum over a long period of time. However, the trapping mechanism may not be present in the developing brain since results from retinal accumulation after in utero exposure suggest that mercury is excreted rather rapidly after the end of exposure (Warfvinge and Bruun, 2000).
ACKNOWLEDGMENT Thanks are due to Dr. Andrejs SchuK tz, Department of Occupational and Environmental Medicine, Lund University, Lund, for help with the mercury analyses.
100
KARIN WARFVINGE
REFERENCES Aschner, M., and Clarkson, T. W. (1988). Uptake of methyl mercury in the rat brain: Effects of amino acid. Brain Res. 462, 31}39. Berlin, M., and Ullberg, S. (1963a). Accumulation and retention of mercury in the mouse. I. An autoradiographic study after a single intravenous injection of mercuric chloride. Arch. Environ. Health 6, 589}601. Berlin, M., and Ullberg, S. (1963b). Accumulation and retention of mercury in the mouse. II. An autoradiographic comparison of phenylmercuric acetate with inorganic mercury. Arch. Environ. Health 6, 602}609. Berlin, M., and Ullberg, S. (1963c). Accumulation and retention of mercury in the mouse. An autoradiographic comparison of methyl mercuric dicyandiamide with inorganic mercury. Arch. Environ. Health 6, 610}616. Berlin, M., Fazackerley, J., and Nordberg, G. (1969). The uptake of mercury in the brains of mammals exposed to mercury vapor and to mercuric salt. Arch. Environ. Health 18, 719}729. Berlin, M., Hua, J., LoK gdberg, B., and Warfvinge, K. (1992). Prenatal exposure to mercury vapor: Effects on brain development. Fund. Appl. Toxicol. 19, 320}321. Cassano, G. B., Amaducci, L., and Viola, P. L. (1966). Distribution of mercury (Hg203) in the brain of chronically intoxicated mice (autoradiographic study). Riv. Patol. Nerv. Ment. 87, 214}225. Cassano, G. B., Viola, P. L., Ghetti, B., and Amaducci, L. (1969). The distribution of inhaled mercury (Hg203 ) vapors in the brain of rats and mice. J. Neuropathol. Exp. Neurol. 29, 308}320. Chang, L. W., and Hartmann, H. A. (1972a). Electron microscopic histochemical study on the localization and distribution of mercury in the nervous system after mercury intoxication. Exp. Neurol. 35, 122}137. Chang, L. W., and Hartmann, H. A. (1972b). Ultrastructural studies of the nervous system after mercury intoxication. I. Pathological changes in the nerve cell bodies. Acta Neuropathol. 20, 122}138. Clarkson, T. W., Magos, L., and Greenwood, M. R. (1972). The transport of elemental mercury into fetal tissues. Biol. Neonate 21, 239}244. Clarkson, T. W., Cox, F., Marsh, D. O., Myers, G. J., Al-Tikriti, S., Amin-Saki, L., and Dabbagh, A. R. (1981). Dose-response relationships for adult and prenatal exposures to methylmercury. In ‘‘Measurement of Risk’’ (G. Berg, and H. Maillie, Eds.), pp. 111}130. Plenum, New York. Danscher, G., MoK ller-Madsen, B. (1985). Silver ampli7cation of mercury sul7de and selenide: A histochemical method for light and electron microscopic localization of mercury in tissue. J. Histochem. Cytochem. 33, 219}228. Danscher, G., and Rungby, J. (1986). Differentiation of histochemically visualized mercury and silver. Histochemistry 18, 109}114. Davis, L. E., Wands, J. R., Weiss, S. A., Price, D. L., and Girling, E. F. (1974). Central nervous system intoxication from mercurous chloride laxatives. Quantitative, histochemical and ultrastructural studies. Arch. Neurol. 30, 428}431. Einarsson, OG ., Lindstedt, G., and BergstroK m, B. (1984). A computerized automatic apparatus for determination of mercury in biological samples. J. Autom. Chem. 6, 74}79.
Harada, Y. (1968). Congenital (or fetal) Minamata Bay disease. In ‘‘Minamata Disease, Study Group of Minamata Disease’’ (M. Kutsuma, Ed.), Kumamoto, Japan. Hargreaves, R. J., Foster, J. R., Pelling, D., Moorhouse, S. R., Gangolli, S. D., and Rowland, I. R. (1985). Changes in the distribution of histochemically localized mercury in the CNS and in tissue levels of organic and inorganic mercury during the development of intoxication in methyl mercury treated rats. Neuropathol. Appl. Neurobiol. 11, 383}401. Hua, J., Brun, A., and Berlin, M. (1995). Pathological changes in the Brown Norway rat cerebellum after mercury vapor exposure. Toxicology 104, 83}90. Hursh, J. B., Clarkson, T. W., Cherian, M. G., Vostal, J. V., and Mallie, R. V. (1976). Clearance of mercury (Hg197, Hg203 ) vapor inhaled by human subjects. Arch. Environ. Health 31, 302}309. LoK gdberg, B., Berlin, M., and SchuK tz, A. (1987). Effects of lead exposure on pregnancy outcome and the fetal brain of squirrel monkeys. Scand. J. Work Environ. Health 13, 135}145. Magos, L. (1967). Mercury blood interaction and mercury uptake by brain. Environ. Res. 1, 323}327. Matsumoto, H., Koya, G., and Takeuchi, T. (1965). Fetal Minamata disease: A neuropathological study of two cases of intrauterine intoxication by a methyl mercury compound. J. Neuropathol. Exp. Neurol. 24, 563}574. MoK ller-Madsen, B., and Danscher, G. (1986). Localization of mercury in CNS of the rat. I. Mercuric chloride (HgCl2) per os. Environ. Res. 41, 29}43. MoK ller-Madsen, B. (1990). Localization of mercury in CNS of the rat. II. Intraperitoneal injection of methyl mercuric chloride (CH3HgCl) and mercuric chloride (HgCl2). Toxicol. Appl. Pharmacol. 103, 303}323. MoK ller-Madsen, B. (1991). Localization of mercury in CNS of the rat. III. Oral administration of methyl mercuric chloride (CH3HgCl). Fund. Appl. Toxicol. 16, 172}187. MoK ller-Madsen, B., and Danscher, G. (1991). Localization of mercury in CNS of the rat IV. The effect of selenium on orally administered organic and inorganic mercury. Toxicol. Appl. Pharmacol. 108, 457}473. MoK ller-Madsen, B. (1992). Localization of mercury in CNS of the rat. V. Inhalation exposure to metallic mercury. Arch. Toxicol. 66, 79}89. Newland, M. C., Warfvinge, K., and Berlin, M. (1996). Behavioral consequences of in utero exposure to mercury vapor: Alterations in lever-press durations and learning in squirrel monkeys. Toxicol. Appl. Pharmacol. 139, 374}386. Nordberg, G. F., and Serenius, F. (1969). Distribution of inorganic mercury in the guinea pig brain. Acta Pharmacol. Toxicol. 27, 269}283. Rodier, P. M., and Kates, B. (1988). Histoligical localization of methyl mercury in mouse brain and kidney by emulsion autoradiography of 203Hg. Toxicol. Appl. Pharmacol. 92, 224}234. Rosenblum, L. A., and Cooper, R. W. (1968). ‘‘The Squirrel Monkey.’’ Academic Press, New York. Suda, I., Eto, K., Tokunaga, H., Furusawa, R., Suetomi, K., and Takahashi, H. (1989). Different histochemical 7ndings in the brain produced by mercuric chloride and methyl mercury chloride in rats. Neurotoxicology 10, 113}126. Warfvinge, K., Hua, J., and Berlin, M. (1992). Mercury distribution in the rat brain after mercury vapor exposure. Toxicol. Appl. Pharmacol. 117, 46}52.
Hg IN ADULT AND NEONATAL MONKEY CEREBELLUM Warfvinge, K., Hua, J., and LoK gdberg, B. (1994). Mercury distribution in cortical areas and 7ber systems of the neonatal and maternal adult cerebrum after exposure of pregnant squirrel monkeys to mercury vapor. Environ. Res. 67, 196}208. Warfvinge, K. (1995). Mercury distribution in the mouse brain after mercury vapor exposure. Int. J. Exp. Pathol. 76, 29}35.
101
Warfvinge, K., and Bruun, A. (1996). Mercury accumulation in the squirrel monkey eye after mercury vapor exposure. Toxicology 107, 189}200. Warfvinge, K., and Bruun, A. (2000). Mercury distribution in the Squirrel monkey retina after in utero exposure to mercury vapor.
Mercury Distribution in the Neonatal and Adult Cerebellum after Mercury Vapor Exposure of Pregnant Squirrel Monkeys1 Karin Warfvinge2 Department of Ophthalmology, University Hospital of Lund, S-221 85 Lund, Sweden, and University College of Health Sciences, S-551 11 JoK nkoK ping, Sweden Received July 2, 1999
amounts of mercury, but the cerebellar nuclei accumulated the highest amounts of mercury. No correlation was found between cellular accumulation and maturity of the brain; that is, the cellular localization of mercury did not differ between adult and neonatal brain, except for the amount of visualized mercury. This pattern corresponded well to the mercury concentrations found in the cerebral occipital pole. The differences found in mercury accumulation were instead considered to be doserelated. The results demonstrate that the distribution of mercury in the cerebellum after mercury vapor exposure is similar to the distribution pattern obtained after methyl mercury exposure and that mercury is trapped in the cerebellum over a long period of time. 2000 Academic Press Key Words: squirrel monkeys; mercury accumulation; developmental toxicity; mercury vapor; cerebellum.
The objectives of the study were (1) to map the detailed localization of mercury in the monkey cerebellum after mercury vapour exposure; (2) to investigate whether there is any difference in mercury distribution between neonatal and adult cerebellum after mercury vapor exposure; (3) to investigate the ability of mercury to accumulate in the cerebellum years after the end of exposure. Pregnant squirrel monkeys were exposed 5 days/ week to mercury vapor at a concentration of 0.5 mg Hg/m3 air 4 or 7 h/day or 1 mg Hg/m3 air for 4 or 7 h/day. Mercury concentration in the offspring and maternal brains was examined by cold vapor, Bameless atomic absorption spectrophotometry. Mercury distribution was examined by processing cerebellar sections for autometallographic (AMG) silver enhancement. Mercury concentration in the offspring cerebral occipital pole ranged between 0.20 and 0.70 lg Hg/g tissue, and in the maternal between 0.80 and 2.58 lg/Hg tissue in animals killed immediately after the end of exposure. AMG revealed that the external granule cell layer of offspring cerebellar tissue contained small amounts of mercury. The molecular layer contained mercury in some of the mercury-exposed monkeys. In the Purkinje cell layer, the Bergmann glial cells together with the Purkinje cells contained mercury. The granule cells and the Golgi cells contained small amounts of mercury. The astrocytes of the medullary layer, identiAed by immunohistochemistry, contained considerable
INTRODUCTION
The toxic properties of mercury vapor are mainly due to the fact that mercury accumulates in the brain, causing neurological signs at high exposure. Mercury vapor also penetrates the placental barrier and enters the fetus and its brain (Clarkson et al., 1972; Warfvinge et al., 1994) and retina (Warfvinge and Bruun, 2000). At short-term mercury vapor exposure of rats (500 lg Hg/m3 air for 4 h), mercury is found in only a few areas of the brain, including the cerebellar deep nuclei (MoK ller-Madsen, 1992). Other studies have also shown that the cerebellar nuclei accumulate mercury after mercury vapor exposure (Cassano et al., 1966, 1969; Berlin et al., 1969; Warfvinge et al., 1992; Warfvinge, 1995). Moreover, the Purkinje cells accumulate mercury when the dose is increased. Since the outbreaks of methyl mercury neuropoisoning in Japan in the 1950s (Harada, 1968) and
This paper was presented at Mercury as a Global Pollutant: 5th International Conference, Rio de Janeiro, Brazil, May 23}28, 1999. 1 This work was supported by the NIH, the Swedish Medical Research Council, the Scienti7c Board of JoK nkoK ping County Council and the Foundations of Crafoord, Maggie Stephens, and Vas rdal, Sweden. The experiments were conducted in accordance with national and institutional guidelines for the protection of animal welfare. 2 Address for correspondence: Department of Ophthalmology, University Hospital of Lund, S-221 85 Lund, Sweden. Fax: (46) 46 1 72721. E-mail: [email protected]. 93
0013-9351/00 $35.00 Copyright 2000 by Academic Press All rights of reproduction in any form reserved.
94
KARIN WARFVINGE
in Iraq in the 1970s (Clarkson et al., 1981), the brain has been known to be most vulnerable during its prenatal development. Case reports from these epidemics indicate that the fetal cerebral cortex and the cerebellum are the main targets for mercury injury, with degeneration of cerebral cortical cells and granule and Purkinje cells of cerebellum. A study of the effects of mercury vapor exposure on the rat cerebellum has shown a degeneration of the Purkinje cell layer combined with gliosis (Hua et al., 1995). This indicates that the cerebellar involvement seen after organic mercury exposure might be seen also after mercury vapor exposure. In addition, mercury vapor has proven to be a potent neuroteratogenic substance (Berlin et al., 1992); prenatal exposure results in migration disturbances similar to those found after prenatal methyl mercury exposure, and, in addition, it results in behavioral disturbances (Newland et al., 1996). However, the brain mercury concentration found is 10 times lower after mercury vapor exposure than after methyl mercury exposure. The objectives of the present study were the following: (1) to map the detailed localization of mercury in the monkey cerebellum after mercury vapor exposure; (2) to investigate whether there is any difference in mercury distribution between neonatal and adult cerebellum after mercury vapor exposure; (3) to investigate the ability of mercury to accumulate in the cerebellum years after the end of exposure. MATERIALS AND METHODS
Animals. Female monkeys with good reproductive records were selected from our breeding colony of squirrel monkeys (Saimiri boliviensis boliviensis; subspecies type veri7ed by karyotype). The females were mated with a male for 1 week every third week to produce timed pregnancies. A few weeks after mating diagnostic abdominal palpations and measurements of fundus height were performed. The pregnant monkeys were then exposed to mercury vapor during the rest of gestation. Brains from unexposed newborn offspring and adult monkeys from our breeding colony served as controls. Details of the housing and diet have been described elsewhere (LoK gdberg et al., 1987). Experimental procedures. The pregnant monkeys were enclosed in a chamber of stainless-steel and glass walls of Rochester type. The volume of the chamber was approximately 2 m3 and it was ventilated with an air 8ow of 2 m3 per hour with temperature (23@C) and humidity (about 60%) controlled.
Before entering the chamber, a part of the air passed a mercury vapor generator. Air was also continuously drawn through a mercury meter (Milton Roy P0408, sensitivity 7 lg/m3) at the outlet of the chamber recording the mercury vapor concentration. Thus, mercury vapor concentration in the chamber was continuously recorded. The calibration was carried out by sampling 10 liters of chamber air into mercury-absorbing acid solution (0.4 M KMnO4 and 2% H2SO4 ). Mercury contents were assayed with the acid of cold vapor atomic absorption spectrophotometry. The method for exposure has earlier been described by Warfvinge et al. (1994). Mercury vapor generation. Mercury vapor was generated by stirring metallic mercury in a 10-liter glass 8ask. The stirring was accomplished with the aid of a magnetic stirrer. The mercury concentration in air was adjusted by adapting the turning speed of the stirring magnet. Further adjustment could be obtained by changing the amount of incoming air through the mercury generator. Mercury exposure. From the groups of pregnant monkeys, exposed to different levels of mercury, described above, the following exposure levels were used in the present study: Group B was exposed to a mercury concentration of approximately 1 mg Hg/m3 air for 7 h/day and Group C for 4 h/day, Group D to a mercury concentration of 0.5 mg Hg/m3 air for 7 h/day and Group E for 4 h/day 5 days/week during gestation. From these groups of exposed monkeys, two offspring (Bo2 and Bo3) brains from group B, three offspring (Co1, Co3, and Co4) brains and three maternal (C2, C3, and C4) brains from group C, four maternal (D1, D2, D5, and D6) brains from group D, and two brains (E3 and E5) from Group E were used for the present study (Table 1). Two monkeys (C4 and D1) died and one monkey (D2) was sacri7ced at the end of exposure. Four monkeys (D5, D6, E3, and E5) were sacri7ced 1 year and two monkeys (C2 and C3) 3 years after the end of exposure. The ages of the offspring ranged between 18 weeks gestational age and 2 days. Mercury analysis. The total mercury content in the cerebral occipital pole was determined by a modi7ed method described by Einarsson et al. (1984). Brie8y, homogenized brain samples from the right occipital pole were digested with a mixture of HClO and HNO3 overnight at 68@C. The procedure was modi7ed for higher sensitivity by enrichment (amalgamation) of the mercury vapor on a gold-wire 7lter. Chemical blanks and reference samples (Seronorm Trace Elements, Batch 904, Nycomed AS, Oslo,
95
Hg IN ADULT AND NEONATAL MONKEY CEREBELLUM
TABLE 1 Exposure Indices for Squirrel Monkeys Exposed to Mercury Vapor during Pregnancy
Exposure duration
Calculated Hg absorption
External Hg dose
Monkey No. (Time lapse between exposure and 7xation)
Weeks of gestational age
No. of days with exposure
Approx. concentration (mg/m3)
Hours per day, with exposure
Total hours of exposure
B2 B3 C1 C2 (3 years) C3 (3 years) C4 (directly) D1 (directly) D2 (1 month) D5 (1 year) D6 (1 year) E3 (1 year) E5 (1 year)
5}18 5}22 7}21 7}21 7}20 7}23 5}9 6}21 5}10 5}8 7}20 5}21
66 84 65 69 59 72 20 65 38 18 65 79
1 1 1 1 1 1 0.5 0.5 0.5 0.5 0.5 0.5
7 7 4 4 4 4 7 7 7 7 4 4
422 496 263 283 239 287 137 400 251 125 245 323
Norway) were included in the sample series. Mercury was determined by cold vapor, 8ameless atomic absorption spectrophotometry. The mercury concentrations in the maternal and offspring brains and the calculated mercury absorption are presented in Table 1. Mercury absorption was calculated as follows: (mercury concentration in the chamber);(time of exposure);(alveolar ventilation of squirrel monkeys);80%. The mercury concentration was measured by the mercury meter as described above, the alveolar ventilation of monkeys is 13 liters/h (Rosenblum and Cooper, 1968) and 80% absorption is assumed, as has been shown for man (Hursh et al., 1976). Preparation of tissue. Three of the offspring (Bo2, Co1, and Co4) and seven adult monkeys (C2, C3, D2, D5, D6, E3, and E5) were anesthetized with barbiturate (Mebumal) and the CNS was 7xed by perfusion with 4% formaldehyde in PBS. The two other mercury-exposed offspring (Bo3 and Co3) were found dead and their brains were removed and placed in 4% formaldehyde in PBS. Two mercuryexposed adult monkeys (C4 and D1) died and the brains from these monkeys were also placed in 4% formaldehyde in PBS. Perfusion-7xed brains from two unexposed offspring, immersion-7xed brains from three unexposed offspring and eight unexposed adult monkeys were used as control brains. Two-to
Per day Total (lg) (lg/day) 4564 5402 2693 2901 2492 3020 768 2497 1462 685 1378 1585
69 64 41 42 42 42 38 38 38 38 21 20
Hg concentration in the occipital pole
Mother (lg/g tissue)
Offspring (lg/g tissue) 0.21 0.70 0.20
2.58 0.80 1.29
0.24 0.30
three-millimeter-coronal slices of the cerebellum were then dissected out. The slices were dehydrated and embedded in paraf7n. Five-micrometer-thick sections were cut and placed on glass slides and further processed for visualization of mercury using the autometallographic method described by Danscher and MoK llerMadsen (1985). Brie8y, the slides were deparaf7nized, coated with gelatin, and developed at 26@C fr 1 h in a dark box (the developer contained gum arabic, sodium citrate buffer, hydroquinone, and silver lactate). The gelatin was washed out in 40@C tap water followed by distilled water, and the slides were then exposed to 5% sodium thiosulfate and counterstained with hematoxylineosin. The method has been shown, in several reports by Danscher and MoK ller-Madsen (Danscher and MoK ller-Madsen, 1985; MoK ller-Madsen and Danscher, 1986, 1991; MoK ller-Madsen, 1990, 1991, 1992), to be speci7c for metallic gold and inorganic mercury. To exclude the possibility that other metals could have acted as catalysts for the reduction, sections from all monkeys were treated with potassium cyanide or acid according to Danscher and MoK ller-Madsen (1985) and Danscher and Rungby (1986). Additionally, immunohistochemistry was carried out by an avidin-biotin complex technique on one section from each monkey. A bovine polyclonal antibody raised against human glial 7brillary acidic
96
KARIN WARFVINGE
protein (GFAP) was purchased from Dakopatts AB in Stockholm. At microscopy, at least 10 sections from each block were examined for mercury content. The amount of visualized mercury for each cell type was estimated according to a three-grade scale: #, a few grains in the cytoplasm; ##, a moderate amount of grains in the cytoplasm; ###, the cytoplasm was loaded with grains. RESULTS
Mercury exposure. The duration of exposure and the estimated absorbed amount of mercury are presented in Table 1. The exposure started between 5 and 7 weeks of gestational age and ended between 18 and 23 weeks of gestational age (normal gestational duration: 152 days), except for monkeys D1, D5, and D6 which were exposed 4}7 weeks. The estimated mercury absorption ranged between 20 and 69 lg Hg/day. Mercury concentration. Concentrations of mercury in the cerebral occipital pole of some of the animals used in the present study are presented in Table 1. The maternal brains always showed higher mercury concentration (0.80}2.58 lg/g tissue) than the offspring brains (0.20}0.70 lg/g tissue), although the external mercury dose was lower for monkeys D1 and D2 than for offspring Bo2, Bo3, Co1, and Co4. Mercury distribution. No signi7cant difference in the results was found between perfusion-7xed and TABLE 2 Mercury Content Estimated on a Three-Grade Scale
Monkey No. External granule cell layer Molecular layer Purkinje cell layer Purkinje cells Bergmann glial cells Granule cell layer Granule cells Golgi cells Medullary layer Astrocytes Other glial cells Deep cerebellar nuclei Neurons Glial cells a
D5, D6, Bo2, C2, C3 E3, E5 Bo3
Co1, Co3, Co4
C4, D2
D1
;
;
;
;
#a
#a
##
##
!
!
#
#a
## ##
# #
## ##
# #
! !
! !
#a #
! #a
# !
# !
## #a
#a !
### ## ### ## # #
! !
### ## # !
### ### ## ### ## #
Some cells contained mercury.
# #
### ## ## #a
FIG. 1. (a) An adult control monkey. No silver enhancement is observed. Large arrows point at Purkinje cells and small arrows at Bergmann glial cells. Scale bar 30 lm. (b) A cerebellar nucleus of an adult control monkey. No visualized mercury is found. Arrows point at large neurons. Scale bar, 30 lm. Abbreviations for all 7gures: ext, external layer; mol, molecular layer; Pur, Purkinje cell layer; gran, granule cell layer.
immersion-7xed material, nor in the potassium cyanide- and acid-treated sections. No silver-enhanced mercury was detected in the brains of the control animals. The results of the examination of cerebellum from 7ve prenatally mercury-exposed monkey offspring and nine mercury-exposed adult monkeys are presented in Table 2 and Figs. 1}3. Immunohistochemistry using the GFAP antibody revealed that the positive cells of the medullary layer were similar in numbers and morphology to those that contained silver grains. The microscopical examination of sections stained on at least 10 different occasions revealed a uniform staining intensity within each cell type and each animal.
Hg IN ADULT AND NEONATAL MONKEY CEREBELLUM
97
FIG. 2. (a) The external granule cell layer from monkey offspring Co4. Mercury deposits are seen in the epithelium covering the folia (large arrows) and in the wall of a vessel (small arrows). Scale bar, 30 lm. (b) Offspring Bo3. This micrograph demonstrate mercury accumulation in the molecular layer (mol), the Bergmann glial cells (small arrows), the Purkinje cells (large arrows), and some cells of the granule cell layer (gran). Arrowhead points at a Golgi cell containing a few silver grains. Scale bar, 30 lm. (c) The medulla of offspring Bo2. Arrows point at mercury-containing astrocytes, which were identi7ed with GFAP immunohistochemistry. Scale bar, 30 lm. (d) A cerebellar nucleus of monkey Co4. Large arrows point at mercury-containing neurons, small arrows at glial cells, and arrowheads at dilated blood vessels that have accumulated mercury in their walls. Scale bar, 30 lm.
DISCUSSION
Cerebellar mercury distribution revealed by autoradiography after administration of mercury compounds has been studied in mice (Berlin and Ullberg, 1963a, b, c; Cassano et al., 1966, 1969; Rodier and Kates, 1988), rats (Berlin et al., 1969; Cassano et al., 1969; Chang and Hartmann 1972a, b), guinea pigs (Nordberg and Serenius, 1969), rabbits (Berlin et al., 1969), and monkeys (Berlin et al., 1969). In a series of reports on rats, MoK ller-Madsen (1990, 1991, 1992) and MoK ller-Madsen and Danscher (1986, 1991) have demonstrated at the cellular level
that the deep cerebellar nuclei accumulate the earliest amount of mercury, irrespective of the mercury compound administered. After mercury vapor exposure, the Purkinje cells and the granule cells are the next target for mercury accumulation (MoK llerMadsen, 1992; Warfvinge et al., 1992). In monkeys, Berlin et al. (1969) have described mercury accumulation in nucleus dentatus, Purkinje cells, and granule cells after mercury vapor exposure. However, the autoradiographic study did not reveal the exact cellular distribution. The present investigation demonstrates the cellular distribution of mercury in the cerebellum of the neonatal and adult monkey brain.
98
KARIN WARFVINGE
FIG. 3. (a) Monkey C4. The molecular layer (mol) contains a moderate amount of mercury. Arrowhead points at a cell which has accumulated mercury. Large arrows point at Purkinje cells and small arrows at Bergmann glial cells in the Purkinje cell layer (Pur). A few silver grains are found in the granule cell layer (gran). Scale bar, 30 lm. (b). The medullary layer of monkey C4. Arrows point at mercury-containing astrocytes, which were identi7ed with GFAP immunohistochemistry. Scale bar, 30 lm. (c) A cerebellar nuclei of monkey D1. Arrows point at large, mercury-containing neurons. Scale bar, 30 lm.
In the present study, neurons of the deep cerebellar nuclei accumulated mercury in all mercury-exposed monkeys examined. This is in accordance with previous studies on mercury vapor exposure (Berlin et al., 1969; Cassano et al., 1966, 1969; MoK ller-Madsen, 1992; Warfvinge et al., 1992; Warfvinge, 1995). These nuclei belong to the motor system, and a causal relationship between the localization and the neurological symptoms induced by mercury intoxication has been suggested by MoK ller-Madsen (1990). In the present study, the Purkinje cells and the Bergmann glial cells contained visualized mercury
in all monkeys examined. The Purkinje cells have previously been described as being targets for mercury accumulation after mercury vapor exposure (MoK ller-Madsen, 1992; Warfvinge et al., 1992; Warfvinge, 1995), in monkeys after only a 4-h exposure to 1 mg Hg/m3 (Berlin et al., 1969). Accumulation of mercury in the Bergmann glial cells after mercury vapor exposure has not been described, although proliferation of Bergmann glial cells together with Purkinje cell degeneration exists in rats (Hua et al., 1995). Accumulation of mercury in the Bergmann glial cells has, however, been described after exposure to organic mercury (Hargreaves et al., 1985;
Hg IN ADULT AND NEONATAL MONKEY CEREBELLUM
MoK ller-Madsen, 1991; MoK ller-Madsen and Danscher, 1991; Suda et al., 1989). In the study by Hargreaves et al. (1985), it was found that the Bergmann glial cells contained mercury (mercury concentration in cerebellum was 12 lg/g tissue; the inorganic mercury fraction represented less than 1%) during the latent phase, that is, before functional disturbances appeared in the methyl mercury-intoxicated rats. In the symptomatic phases, mercury was also visualized in the Purkinje cells. Similar 7ndings were reported by Suda et al. [brain concentrations: 8.87 (inorganic mercury: 0.21) and 5.54 (0.23) lg/g tissue]. Hence, the distribution of mercury in the Purkinje cells and the Bergmann glial cells of the monkey cerebellum after long-term mercury vapor exposure (brain concentration range: 0.20} 2.58 lg/g tissue) corresponds to the distribution of inorganic mercury after methyl mercury exposure. The organic and inorganic forms of mercury are not transported into the CNS (and its cells) in the same way. Mercury vapor easily diffuses across the blood-brain barrier into the CNS because of its high lipophilicity (Magos, 1967). Within the CNS, it is oxidized by an endogenous hydrogen peroxide catalase system to the divalent cation Hg2>. This oxidation results in 7xation of mercury within the cell by its binding to -SH-containing ligands (Magos, 1967). The transport model for the organic mercury compound;methyl mercury;differs from that of mercury vapor. In the plasma, the high af7nity of methyl mercury to -SH-containing compounds, such as cysteine, results in the methyl mercury-cysteine conjugates that are transported across the bloodbrain barrier into the CNS (Aschner and Clarkson, 1988). Despite the fact that the transport model differs, Purkinje cells and Bergmann glia cells accumulate mercury after both methyl mercury and mercury vapor exposure. Another resemblance between the present study and the one by Hargreaves et al. (1985) is the intense staining of the astrocytes of the medullary layer. However, the mercury deposits in the glial cells were largely absent at 31 days after termination of methyl mercury exposure, indicating that these cells serve as ‘‘buffer function’’. Mercury accumulation in the glial cells of the medullary layer has been reported for rats (Warfvinge et al., 1992) and mice (Warfvinge, 1995) after long-term mercury vapor exposure. The most active glial cell in this respect seemed to be the 7brillous astocyte, as was indicated by the GFAP reaction. The granule cells contained little or no visualized mercury in the monkey brains examined. This
99
is in accordance with previous studies, mentioned above, on mercury vapor exposure. The susceptibility of the granular layer to mercury vapor (Davis et al., 1974) and methyl mercury (Matsumoto et al., 1965) exposure has been reported in humans. We have recently reported on the distribution of mercury in the eye after direct exposure and where the monkeys survived years after the end of exposure (Warfvinge and Bruun, 1996). In that study, it was revealed that mercury was trapped predominantly in the vessel walls and retinal pigment epithelium over a long period of time. This heavy accumulation of mercury in the vessel walls is not obvious in the brain. This indicates that one or more retinal trapping mechanisms exist, not present to nearly the same extent in the brain capillaries, since the monkeys that survived years after the exposure did not show that heavy accumulation in the vessel walls. In conclusion, no correlation was found between cellular accumulation and maturity of the brain in the present study; that is, the cellular localization of mercury did not differ between adult and neonatal brain, except for the amount of visualized mercury. This pattern corresponds well to the mercury concentrations found in the cerebral occipital pole. The reason for the neonatal brains having a lower concentration than the maternal brains might be explained by the fact that the offspring liver contains more mercury than the maternal (unpublished observations). Much of the mercury is accumulated in the neonatal liver, protecting the brain from severe exposure. The differences found in mercury accumulation were instead considered to be doserelated. The results demonstrate that the distribution of mercury in the cerebellum after mercury vapor exposure is similar to the distribution pattern obtained after methyl mercury exposure and that mercury is trapped in the cerebellum over a long period of time. However, the trapping mechanism may not be present in the developing brain since results from retinal accumulation after in utero exposure suggest that mercury is excreted rather rapidly after the end of exposure (Warfvinge and Bruun, 2000).
ACKNOWLEDGMENT Thanks are due to Dr. Andrejs SchuK tz, Department of Occupational and Environmental Medicine, Lund University, Lund, for help with the mercury analyses.
100
KARIN WARFVINGE
REFERENCES Aschner, M., and Clarkson, T. W. (1988). Uptake of methyl mercury in the rat brain: Effects of amino acid. Brain Res. 462, 31}39. Berlin, M., and Ullberg, S. (1963a). Accumulation and retention of mercury in the mouse. I. An autoradiographic study after a single intravenous injection of mercuric chloride. Arch. Environ. Health 6, 589}601. Berlin, M., and Ullberg, S. (1963b). Accumulation and retention of mercury in the mouse. II. An autoradiographic comparison of phenylmercuric acetate with inorganic mercury. Arch. Environ. Health 6, 602}609. Berlin, M., and Ullberg, S. (1963c). Accumulation and retention of mercury in the mouse. An autoradiographic comparison of methyl mercuric dicyandiamide with inorganic mercury. Arch. Environ. Health 6, 610}616. Berlin, M., Fazackerley, J., and Nordberg, G. (1969). The uptake of mercury in the brains of mammals exposed to mercury vapor and to mercuric salt. Arch. Environ. Health 18, 719}729. Berlin, M., Hua, J., LoK gdberg, B., and Warfvinge, K. (1992). Prenatal exposure to mercury vapor: Effects on brain development. Fund. Appl. Toxicol. 19, 320}321. Cassano, G. B., Amaducci, L., and Viola, P. L. (1966). Distribution of mercury (Hg203) in the brain of chronically intoxicated mice (autoradiographic study). Riv. Patol. Nerv. Ment. 87, 214}225. Cassano, G. B., Viola, P. L., Ghetti, B., and Amaducci, L. (1969). The distribution of inhaled mercury (Hg203 ) vapors in the brain of rats and mice. J. Neuropathol. Exp. Neurol. 29, 308}320. Chang, L. W., and Hartmann, H. A. (1972a). Electron microscopic histochemical study on the localization and distribution of mercury in the nervous system after mercury intoxication. Exp. Neurol. 35, 122}137. Chang, L. W., and Hartmann, H. A. (1972b). Ultrastructural studies of the nervous system after mercury intoxication. I. Pathological changes in the nerve cell bodies. Acta Neuropathol. 20, 122}138. Clarkson, T. W., Magos, L., and Greenwood, M. R. (1972). The transport of elemental mercury into fetal tissues. Biol. Neonate 21, 239}244. Clarkson, T. W., Cox, F., Marsh, D. O., Myers, G. J., Al-Tikriti, S., Amin-Saki, L., and Dabbagh, A. R. (1981). Dose-response relationships for adult and prenatal exposures to methylmercury. In ‘‘Measurement of Risk’’ (G. Berg, and H. Maillie, Eds.), pp. 111}130. Plenum, New York. Danscher, G., MoK ller-Madsen, B. (1985). Silver ampli7cation of mercury sul7de and selenide: A histochemical method for light and electron microscopic localization of mercury in tissue. J. Histochem. Cytochem. 33, 219}228. Danscher, G., and Rungby, J. (1986). Differentiation of histochemically visualized mercury and silver. Histochemistry 18, 109}114. Davis, L. E., Wands, J. R., Weiss, S. A., Price, D. L., and Girling, E. F. (1974). Central nervous system intoxication from mercurous chloride laxatives. Quantitative, histochemical and ultrastructural studies. Arch. Neurol. 30, 428}431. Einarsson, OG ., Lindstedt, G., and BergstroK m, B. (1984). A computerized automatic apparatus for determination of mercury in biological samples. J. Autom. Chem. 6, 74}79.
Harada, Y. (1968). Congenital (or fetal) Minamata Bay disease. In ‘‘Minamata Disease, Study Group of Minamata Disease’’ (M. Kutsuma, Ed.), Kumamoto, Japan. Hargreaves, R. J., Foster, J. R., Pelling, D., Moorhouse, S. R., Gangolli, S. D., and Rowland, I. R. (1985). Changes in the distribution of histochemically localized mercury in the CNS and in tissue levels of organic and inorganic mercury during the development of intoxication in methyl mercury treated rats. Neuropathol. Appl. Neurobiol. 11, 383}401. Hua, J., Brun, A., and Berlin, M. (1995). Pathological changes in the Brown Norway rat cerebellum after mercury vapor exposure. Toxicology 104, 83}90. Hursh, J. B., Clarkson, T. W., Cherian, M. G., Vostal, J. V., and Mallie, R. V. (1976). Clearance of mercury (Hg197, Hg203 ) vapor inhaled by human subjects. Arch. Environ. Health 31, 302}309. LoK gdberg, B., Berlin, M., and SchuK tz, A. (1987). Effects of lead exposure on pregnancy outcome and the fetal brain of squirrel monkeys. Scand. J. Work Environ. Health 13, 135}145. Magos, L. (1967). Mercury blood interaction and mercury uptake by brain. Environ. Res. 1, 323}327. Matsumoto, H., Koya, G., and Takeuchi, T. (1965). Fetal Minamata disease: A neuropathological study of two cases of intrauterine intoxication by a methyl mercury compound. J. Neuropathol. Exp. Neurol. 24, 563}574. MoK ller-Madsen, B., and Danscher, G. (1986). Localization of mercury in CNS of the rat. I. Mercuric chloride (HgCl2) per os. Environ. Res. 41, 29}43. MoK ller-Madsen, B. (1990). Localization of mercury in CNS of the rat. II. Intraperitoneal injection of methyl mercuric chloride (CH3HgCl) and mercuric chloride (HgCl2). Toxicol. Appl. Pharmacol. 103, 303}323. MoK ller-Madsen, B. (1991). Localization of mercury in CNS of the rat. III. Oral administration of methyl mercuric chloride (CH3HgCl). Fund. Appl. Toxicol. 16, 172}187. MoK ller-Madsen, B., and Danscher, G. (1991). Localization of mercury in CNS of the rat IV. The effect of selenium on orally administered organic and inorganic mercury. Toxicol. Appl. Pharmacol. 108, 457}473. MoK ller-Madsen, B. (1992). Localization of mercury in CNS of the rat. V. Inhalation exposure to metallic mercury. Arch. Toxicol. 66, 79}89. Newland, M. C., Warfvinge, K., and Berlin, M. (1996). Behavioral consequences of in utero exposure to mercury vapor: Alterations in lever-press durations and learning in squirrel monkeys. Toxicol. Appl. Pharmacol. 139, 374}386. Nordberg, G. F., and Serenius, F. (1969). Distribution of inorganic mercury in the guinea pig brain. Acta Pharmacol. Toxicol. 27, 269}283. Rodier, P. M., and Kates, B. (1988). Histoligical localization of methyl mercury in mouse brain and kidney by emulsion autoradiography of 203Hg. Toxicol. Appl. Pharmacol. 92, 224}234. Rosenblum, L. A., and Cooper, R. W. (1968). ‘‘The Squirrel Monkey.’’ Academic Press, New York. Suda, I., Eto, K., Tokunaga, H., Furusawa, R., Suetomi, K., and Takahashi, H. (1989). Different histochemical 7ndings in the brain produced by mercuric chloride and methyl mercury chloride in rats. Neurotoxicology 10, 113}126. Warfvinge, K., Hua, J., and Berlin, M. (1992). Mercury distribution in the rat brain after mercury vapor exposure. Toxicol. Appl. Pharmacol. 117, 46}52.
Hg IN ADULT AND NEONATAL MONKEY CEREBELLUM Warfvinge, K., Hua, J., and LoK gdberg, B. (1994). Mercury distribution in cortical areas and 7ber systems of the neonatal and maternal adult cerebrum after exposure of pregnant squirrel monkeys to mercury vapor. Environ. Res. 67, 196}208. Warfvinge, K. (1995). Mercury distribution in the mouse brain after mercury vapor exposure. Int. J. Exp. Pathol. 76, 29}35.
101
Warfvinge, K., and Bruun, A. (1996). Mercury accumulation in the squirrel monkey eye after mercury vapor exposure. Toxicology 107, 189}200. Warfvinge, K., and Bruun, A. (2000). Mercury distribution in the Squirrel monkey retina after in utero exposure to mercury vapor.