Pathogenesis of selective damage of granule cell layer in cerebellum of rats exposed to methylmercury

Abstract

Granule cell-selective toxicity of methylmercury in the cerebellum is one of the main unresolved issues in the pathogenesis of Minamata disease. Rats were orally administered methylmercury chloride (10 mg/kg/day) for 5 consecutive days, and their brains were harvested on days 1, 7, 14, 21, or 28 after the last administration for histological examination of the cerebellum. It was found that methylmercury caused a marked degenerative change to the granule cell layers but not to the Purkinje cell layers. The generative change of the granule cell layer was due to cell death, including apoptosis, which occurred at day 21 and beyond after the methylmercury administration. Meanwhile, cytotoxic T-lymphocytes and macrophages had infiltrated the granule cell layer. Additionally, granule cells are shown to be a cell type susceptible to TNF-α. Taken together, these results suggest that methylmercury causes small-scale damage to granule cells, triggering the infiltration of cytotoxic T-lymphocytes and macrophages into the granule cell layer, which secrete tumor necrosis factor-α (TNF-α) to induce apoptosis in granule cells. This chain is established based on the susceptibility of granule cells to methylmercury, the ability of cytotoxic T lymphocytes and macrophages to synthesize and secrete TNF-α, and the sensitivity of granule cells to TNF-α and methylmercury. We propose to call the pathology of methylmercury-induced cerebellar damage the “inflammation hypothesis.”

INTRODUCTION

Minamata disease, also known as Hunter-Russell syndrome, is a toxic neurological disease caused by the daily ingestion of methylmercury, which accumulates in high concentrations in fish and shellfish through the food chain. Methylmercury has been unintentionally produced during the industrial synthesis of acetaldehyde from acetylene at the Chisso Minamata Plant since the early 1950s (Clarkson and Strain, 2003). Minamata disease is medically characterized by the following two features: first, methylmercury toxicity is highly specific to the central nervous system; second, the degree of brain damage varies greatly depending on the timing and amount of methylmercury exposure. Minamata disease is classified into adult, pediatric, and fetal. The symptoms of adult and pediatric Minamata disease include sensory disturbances in the extremities, motor disturbance, balance dysfunction, bilateral centripetal visual field constriction, gait disturbance, muscle weakness, and hearing impairment. Taste disorders, odor disorders, and psychiatric symptoms were also observed, depending on individual sensitivity. However, the detailed mechanisms underlying nervous system-specific toxicity of methylmercury remain largely unknown.

The main unresolved issues can be summarized in two questions. The first question concerns the localization of lesions in the cerebral tissue. In adult patients with Minamata disease, lesions in the cerebral tissue are localized around the deep sulci, such as the precentral and postcentral gyri around the central sulcus, the postcentral and transverse temporal gyri around the sulcus of Sylvius, and the thoracic area around the thoracic sulcus of the occipital lobe, where the visual cortex is located. In contrast, in the brains of patients with fetal Minamata disease, methylmercury toxicity is less localized, and degenerative changes occur throughout the brain (Eto, 1997). The pathogenesis of such site-specific injuries in the cerebrum, especially the mechanism of injury site localization in adults, appears to be important. In a detailed pathological study of the brains of common marmosets exposed to methylmercury, it was found that methylmercury exposure caused brain edema as an initial lesion in tissues around the deep cerebral sulcus, followed by selective injury to tissues around the brain edema, and methylmercury-induced site-specific injury to the cerebral cortex (Eto et al., 2001). This “Edema hypothesis” is promising as a pathogenesis of site-specific cerebral injury caused by methylmercury.

The second question is the mechanisms by which the specific injury of the granule cell layer that causes cerebellar ataxia occurs in the cerebellum. In methylmercury-induced cerebellar lesions, the loss or reduction of Purkinje cells, which are large neurons and small granule cells, occurs in response to the amount of methylmercury. In the mild to moderate stages, Purkinje cells are relatively preserved, whereas granule cells just under the Purkinje cell layer are selectively injured, resulting in central cerebellar atrophy (Eto et al., 1999). Although edema formation is observed in the cerebellum and cerebrum in the early stages of methylmercury intoxication, it is difficult to explain specific lesion formation in the granule cell layer due to edema. However, there is no alternative hypothesis regarding the pathogenesis of the severe damage to granule cells in the cerebellum of patients with Minamata disease.

Previous studies on methylmercury-induced cerebellar injury have focused on the mechanisms underlying the differential susceptibility of granule cells, Purkinje cells, and glial cells, which constitute the cerebellar tissue, to methylmercury toxicity in an attempt to explain the granule cell-specific exhibition of methylmercury toxicity (Edwards et al., 2005; Herden et al., 2008; Allen et al., 2002). Although such differences in susceptibility are important for understanding the direct effects of methylmercury on neurons, it is difficult to rationally explain the mechanisms underlying the rapid and massive granule cell layer injury caused by methylmercury in the cerebellar tissue. The purpose of this study was to investigate the pathogenesis of methylmercury-induced lesions specific to the granule cell layer of the cerebellum through the histopathological examination of the cerebellum of rats in which methylmercury administration induced typical toxic symptoms of hindlimb crossing.

MATERIALS AND METHODS

Materials

Nine-week-old male Wistar rats were purchased from CLEA (Tokyo, Japan). Bovine aortic endothelial and smooth muscle cells were purchased from Cell Applications (San Diego, CA, USA). Primary cultured rat cerebellar granule cells were obtained from ScienCell Research Laboratories (Carlsbad, CA, USA). Dulbecco’s modified Eagle medium was purchased from Nissui Pharmaceutical Co., Ltd. (Tokyo, Japan). Methylmercury chloride was purchased from Tokyo Kasei (Tokyo, Japan). Meyer’s hematoxylin and Eosin Y were purchased from Mutoh Chemical (Tokyo, Japan). Methyl green and DAB substrate kits were purchased from DaKo Cytomation (Carpinteria, CA, USA). Phosphomolybdate, chloroform, and crystal violet were purchased from Nacalai Tesque Inc. (Kyoto, Japan). NOE cover glass (0.12~0.17 mm), APS-coated glass slides (0.9~1.2 mm), and specimen sealant M480 were from Matsunami Glass Industry (Osaka, Japan). Bovine serum albumin (BSA) and in situ apoptosis detection kits were purchased from Sigma-Aldrich (St. Louis, MO, USA) and Takara Bio (Otsu, Japan), respectively. The Histofine SAB-PO (M) kit (mouse), Histofine SAB-PO (G) kit (goat), and Histofine SAB-PO (R) kit (rabbit) were purchased from Nichirei Bioscience, Tokyo, Japan. Mouse anti-rat CD 3 (MCA772), mouse anti-rat CD 8 alpha (MCA48GA), and mouse anti-rat CD 68 antibody (MCA341R) were purchased from AbD Serotec (Kidington, UK). The CytoTox96 non-radioactive cytotoxicity assay kit for determining lactate dehydrogenase activity was obtained from Promega (Madison, WI, USA). Glutathione, citric acid monohydrate, trisodium citrate dihydrate, hydrogen peroxide (hydrogen peroxide solution [30%]), phosphate buffer solution, fetal bovine serum, and other reagents were purchased from FUJIFILM Wako Pure Chemical Industries (Osaka, Japan).

Animals

Rats were orally administered methylmercury chloride (10 mg/kg/day) for five consecutive days, and brains were harvested on days 1, 7, 14, 21, or 28 after the last administration.

Histological techniques

The cerebellae were removed from the brains and fixed in 10% neutral formalin. They were embedded in paraffin in a conventional way and sectioned serially to 4 or 5 µm thickness and stained with Mayer’s hematoxylin and eosin. The cerebellum was examined histologically. Additionally, the number of Purkinje cells in the cerebellum was measured using stained images at 200 × magnification, and the number of granule cells per arbitrary 15 cm² of the cerebellar granule cell layer was measured using stained images at 400 × magnification.

Detection of apoptosis

Cerebellar sections from rats exposed to methylmercury were treated with Proteinase K for 15 min at room temperature and then washed three times with PBS for 5 min. They were then immersed in an appropriate volume of 3% H₂O₂ solution for 5 min at room temperature to inactivate endogenous peroxidase (POD), followed by three washes with PBS for 5 min. Apoptotic cells were detected using an in-situ apoptosis detection kit. The sections were stained with a 3% methyl green solution for 5 min at room temperature, rinsed, dehydrated, permeabilized with xylene, and sealed with a non-aqueous sealant.

Immunohistochemistry

After deparaffinisation, the hydrated sections were washed with distilled water. The deparaffinized and hydrated sections were immersed in 0.01 M citrate buffer (pH 6.0) and heat-treated in an autoclave at 121°C for 15 min to inactivate the antigen. The sections were then cooled to room temperature and washed thrice with phosphate-buffered saline for 5 min. The sections were immersed in blocking reagent I (3% hydrogen peroxide in methanol) for 15 min at room temperature, washed three times with phosphate-buffered saline for 5 min, and then immersed in blocking reagent II (10% normal animal serum from the Histofine SAB-PO kit, corresponding to the animal immunized with the primary antibody) for 10 min at room temperature. After carefully wiping the area around the sections on the slides with filter paper, the sections were completely covered with CD 3 (1:40), CD 8 (1:40), or CD 68 (1:50) as primary antibodies and placed in a wet box to react overnight at 4°C, and then washed three times with phosphate-buffered saline for 5 min. After carefully wiping the area around the sections on the slides with filter paper, secondary antibodies (using the secondary antibody of the Histofine SAB-PO kit according to the primary antibody) were added to completely cover the sections, which were allowed to react for 10 min at room temperature, followed by washing three times for 5 min in phosphate-buffered saline. The enzyme reagent was placed on all the slides to cover the sections completely, allowed to react for 5 min at room temperature (15-25°C), and then washed three times with phosphate-buffered saline for 5 min. The DAB reaction solution was added to all slides so that the sections were completely covered, and the slides were incubated at room temperature until the stained images were clearly visible under a microscope. Sections were stained with Mayer’s hematoxylin.

Cell culture

Rat cerebellar granules, bovine aortic endothelial cells, and bovine aortic smooth muscle cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum in 24-well plates until confluence. The cells were then treated with recombinant rat tumor necrosis factor- α (TNF-α) (0.5, 1, 2, and 10 ng/mL) for 24 hr. The activity of lactate dehydrogenase that leaked from the cells into the conditioned medium was measured using a CytoTox96 non-radioactive cytotoxicity assay kit. The cytotoxicity index (%) was calculated as follows:

where Tc represents the lactate dehydrogenase activity in the medium of the cells treated with methylmercury and Cc represents the lactate dehydrogenase activity in the medium of the cells treated without methylmercury. The value of Cc was determined as the mean of three biological replicates of the experiment.

Statistical analysis

The data were statistically analyzed using a One-way ANOVA and post hoc Tukey–Kramer’s test. Differences were considered statistically significant at p < 0.05.

RESULTS

Methylmercury causes damage specifically in the cerebellar granule cell layer

First, we examined the cerebellar cortex of rats after exposure to methylmercury. As shown in Fig. 1A, although there were no abnormalities in the Purkinje cell morphology, the number of granule cells appeared to decrease on days 21 and 28 after the administration of methylmercury.

Fig. 1

Morphological observations of the cerebellar cortex of rats treated with methylmercury. [A] No regressive changes are observed in the Purkinje cell layer. In contrast, the granule cell layer was damaged after 21 days. Hematoxylin-Eosin stain (×100). P: Purkinje cells (arrows); M, Molecular layer; G, Granular layer; W, White matter. [B] A marked decrease in granule cells is observed on days 21 and 28 after the administration of methylmercury, and the infiltration of lymphocytes into the granule cell layer is observed after 14 day. Hematoxylin-Eosin stain (×400). M: molecular layer; P: Purkinje cells (blue arrowheads); G, Granule cells (orange arrowheads); L, Lymphocytes (white arrowheads). [C] Detection of apoptotic cells in the cerebellum of methylmercury-treated rats. Red arrowheads indicate TUNEL-positive (apoptotic) cells. Tunel stain (×40). M, molecular layer; G, granular layer; W, white matter.

We examined the cerebellar granule cell layer under a high-magnification field to view the granule cell layer in detail (Fig. 1B). A decrease in the number of stained granule cells was observed in the cerebellar granule cell layer on days 21 and 28 after methylmercury administration. Notably, infiltration of lymphocytes into the granule cell layer was observed from day 14 onward. When the number of cells per unit area was measured using these images, no change due to methylmercury was observed in the Purkinje cells, but the number of granule cells decreased by approximately half after days 21 and 28 when the number of cells in five randomly selected areas was counted. This granule cell-specific damage is consistent with the observation in the cerebellum of patients with Minamata disease, namely, the loss of granule cells in the Purkinje cell layer in the cerebellar hemispheres (Eto et al., 2010). Additionally, we observed lymphocyte infiltration into the granule cell layer of the cerebellum at 14 days and beyond after the administration of methylmercury.

The possibility that the decrease in granule cells induced by methylmercury was due to the induction of apoptosis was investigated using TUNEL staining. In this staining method, apoptotic cells were stained brown. Apoptotic cells largely occurred in the cerebellar granule cell layers at 21 and 28 days after methylmercury administration (Fig. 1C). In other words, methylmercury induced apoptosis specifically in the granule cells of the rat cerebellum; at that time, lymphocyte infiltration had occurred.

Infiltration of granule cell layer by inflammatory cells

Because the infiltration of T lymphocytes into the cerebellar granule cell layer was morphologically suggested, immunostaining with a CD 3 antibody was performed to detect T lymphocytes. As shown in Fig. 2A, the infiltration of CD 3-positive cells (T-lymphocytes), indicated by arrowheads, was observed in the cerebellar granule cell layer at 14, 21, and 28 days after methylmercury administration, and the number of T-lymphocytes increased rapidly in a time-dependent manner.

Fig. 2

Infiltration of granule cell layer of cerebellar cortex of rats treated with methylmercury by cytotoxic T-lymphocytes. [A] Immunostaining with a CD 3 antibody to detect T-lymphocyte infiltration into the granule cell layer of rats treated with methylmercury (×100). Arrowheads indicate CD 3-positive cells (T lymphocytes). T lymphocytes were detected at days 14, 21, and 28 after treatment with methylmercury. M, molecular layer; G, granular layer; W, white matter. [B] Immunostaining with a CD 8 antibody to detect cytotoxic T lymphocytes infiltrating the granule cell layer in rats treated with methylmercury (×200). Rat spleen tissue was used as a positive control. Arrowheads indicate CD 8-positive cells (cytotoxic T lymphocytes). T lymphocytes were detected on days 14, 21, and 28 after treatment with methylmercury.

Among T lymphocytes, cytotoxic T-lymphocytes are a cell-type that induces apoptosis by directly acting on target cells during the inflammatory response. Therefore, we investigated the infiltration of cytotoxic T-lymphocytes into the cerebellar granule cell layer by immunostaining with a CD 8 antibody. Rat spleen tissue sections were used as positive controls. Similar to CD3-positive cells, infiltration of CD 8-positive cytotoxic T-lymphocytes into the cerebellar granule cell layer was confirmed on days 14, 21, and 28 after methylmercury administration, and the number of cytotoxic T lymphocytes increased rapidly in a time-dependent manner (Fig. 2B). These results suggest that the T lymphocytes infiltrating the granule cell layer of the methylmercury-treated rat cerebellum were cytotoxic.

Since the infiltration of cytotoxic T-lymphocytes was confirmed in the cerebellar granule cell layer, we examined the infiltration of macrophages by immunostaining with the CD 68 antibody (Fig. 3). CD 68-positive cells (macrophages) infiltrated the cerebellar granule cell layer on days 21 and 28 after methylmercury administration.

Fig. 3

Infiltration of granule cell layer of cerebellar cortex of rats treated with methylmercury by macrophages. Immunostaining with a CD 68 antibody to detect macrophages infiltrating the granule cell layer of the rats treated with methylmercury (×200). The arrowheads indicate CD 68-positive cells (macrophages). Macrophages were detected on days 21 and 28 after methylmercury treatment. M, molecular layer; P, Purkinje cell layer; G, granular layer; W, white matter.

Cerebellar granule cells are susceptible to TNF-α

Macrophages and lymphocytes have the ability to secrete cytokines such as TNF-α. It is known that there are some cell types, including vascular endothelial cells, which are susceptible to TNF-α (Polunovsky et al., 1994). We considered that granule cells may be a cell-type susceptible to TNF-α. Figure 4 shows the susceptibility of rat cerebellar granule cells to TNF-α, compared to that of either bovine aortic endothelial cells (susceptible to TNF-α) or bovine aortic smooth muscle cells (resistant to TNF-α). It was shown that cerebellar granule cells are as susceptible to TNF-α as vascular endothelial cells, although vascular smooth muscle cells are resistant to TNF-α.

Fig. 4

Susceptibility of rat cerebellar granule cells, bovine vascular endothelial cells, and bovine vascular smooth muscle cells to TNF-α. [A] Morphological appearance of-treated cells. Subconfluent cultures of rat cerebellar granule cells, bovine vascular endothelial cells, and bovine vascular smooth muscle cells were treated with TNF-α (10 ng/mL) for 24 hr. Giemsa stain (× 50). [B] The cytotoxicity of methylmercury was evaluated based on the leakage of lactate dehydrogenase from the cells into the conditioned medium. Subconfluent cultures of rat cerebellar granule cells, bovine vascular endothelial cells, and bovine vascular smooth muscle cells were treated with TNF-α (0.5, 1, 2, 5, or 10 ng/mL) for 24 hr. Values are means ± S.E of three samples. ** Significantly different from the corresponding control, p < 0.01. RCGC, rat cerebellar granule cell; BAEC, bovine vascular endothelial cell; BSMC, bovine vascular smooth muscle cell.

DISCUSSION

Minamata disease is a central nervous system disease, called Hunter-Russell syndrome, caused by the ingestion of large amounts of methylmercury via seafood (Clarkson and Strain, 2003). In this study, we examined the cerebella of rats histologically to investigate the pathogenesis of methylmercury-induced cerebellar lesions. The following results were obtained: (1) methylmercury caused a marked degenerative change in the granule cell layers but not in the Purkinje cell layers; (2) the degenerative changes in the granule cell layer were due to cell death, including apoptosis; (3) cell death occurred 21 days and later following the administration of methylmercury following the infiltration of the granule cell layer by cytotoxic T lymphocytes and macrophages; (4) granule cells are a cell-type susceptible to TNF-α. These results led us to hypothesize the selective damage of granule cells in the cerebellum exposed to methylmercury as follows: first, methylmercury caused cell death to a small degree in granule cells susceptible to methylmercury (Limke et al., 2004; Bellum et al., 2007) in the early stage of methylmercury exposure; secondly, cell death triggers cytotoxic T lymphocytes and macrophages to invade the granule cell layer; third, T-lymphocytes and/or macrophages secrete TNF-α which is highly toxic to the granule cells. Thus, TNF-α secretion may be promoted by methylmercury; fourth, a “TNF-α storm” occurred to induce apoptosis of granule cells on a large scale. The reason for the 21-day time lag between methylmercury exposure and the occurrence of extensive death of granule cells is that this cell death requires time for events such as infiltration of cytotoxic T-lymphocytes and macrophages into the granule cell layers, activation by methylmercury of the lymphocytes and macrophages to secrete TNF-α, and the occurrence of apoptosis of granule cells by TNF-α. Since the involvement of inflammatory cell types, T-lymphocytes and macrophages, in the pathogenesis of cerebellar damage by methylmercury, we propose to call the pathology of methylmercury-induced cerebellar damage the “inflammation hypothesis.” This hypothesis is summarized in Fig. 5.

Fig. 5

Summary of the “inflammation hypothesis” as a pathological hypothesis on the selective and severe damage of the granule cell layer in the cerebellum of rats administered methylmercury. (1) Methylmercury causes a small degree of cell death in granular cells susceptible to methylmercury. (2) The cell death triggers cytotoxic T lymphocytes and macrophages to invade the granule cell layer. (3) T-lymphocytes and/or macrophages secrete TNF-α that is highly toxic to the granule cells. Thus, TNF-α secretion may be promoted by methylmercury. (4) A “TNF-α storm” occurred to induce apoptosis of granule cells on a large scale. As a result, granule cells, a cell-type susceptible to TNF-α are dead extensively by apoptosis. Inflammatory cell involvement is definitely implicated, but mechanisms other than the “TNF-α storm” such as activation of perforin-granzyme B system in cytotoxic T-lymphocytes cannot be excluded.

The granule cell layer was strongly and specifically injured in the cerebellum of rats treated with methylmercury; however, no such injury was observed in the Purkinje cell layer. Such selective injury to granule cells has also been observed in the cerebellum of Minamata disease patients (Eto, 1997). There are several possible mechanisms that define susceptibility to methylmercury. Because methylmercury is transported intracellularly via the L-type neutral amino acid transporter (LAT1) (Simmons-Willis et al., 2002) and extracellularly via multidrug resistance-associated protein 1 (MRP1) (Straka et al., 2016), a higher expression of LAT1 and/or a lower expression of MRP1 may be a mechanism of higher accumulation of intracellular methylmercury to induce susceptibility to methylmercury (Hirooka et al., 2010a; Hirooka et al., 2010b). Glutathione has been reported to reduce intracellular oxidative stress induced by methylmercury (Kromidas et al., 1990; Gatti et al., 2004; Kaur et al., 2006). The expression level of γ-glutamylcysteine synthetase, the rate-limiting enzyme for glutathione synthesis, is lower in granule cells than in Purkinje cells (Li et al., 1996), suggesting that the higher susceptibility of granule cells to methylmercury is due to lower levels of glutathione, which protects cells against oxidative stress induced by methylmercury. Because glutathione promotes the efflux of methylmercury from the intracellular space to the extracellular space (Miura and Clarkson, 1993; Fujiyama et al., 1994), a lower expression level of glutathione may increase the intracellular accumulation of methylmercury by decreasing the transport of intracellular methylmercury to the extracellular space. Metallothionein is a cysteine-rich metal-binding peptide synthesized in all tissues and cells. Its synthesis is induced by oxidative stress as well as by metals, such as cadmium, lead, and inorganic mercury (Sato and Kondoh, 2002). Metallothionein also reduces methylmercury cytotoxicity (Yao et al., 2000). Leyshon-Sørland et al. (1994) investigated the distribution of inorganic mercury and metallothionein in the cerebellum of methylmercury-treated rats and found that inorganic mercury and metallothionein were detected mainly in Purkinje cells and neighboring Hagmann glial cells, but not in granule cells, suggesting that a lower ability of metallothionein induction by oxidative stress may be a mechanism of the higher susceptibility of granule cell to methylmercury. Additionally, the expression levels of the calcium-binding protein calbindin D28 (Celio, 1990; Kadowaki et al., 1993; Amenta et al., 1994) and calcium channel subtypes (Tanaka et al., 1995; Randall and Tsien, 1995) can influence the susceptibility of neural cells to methylmercury. These results are useful for understanding the higher susceptibility of granule cells to methylmercury; however, granule cell-selective damage by methylmercury cannot be explained because the damage quickly and extensively occurred on day 21 and beyond, after the administration of methylmercury. These results are useful for understanding the higher susceptibility of granule cells to methylmercury; however, granule cell-selective damage by methylmercury cannot be explained by them because the damage quickly and extensively occurred on day 21 and beyond, after the administration of methylmercury. If the difference in susceptibility between granule cells and Purkinje cells was important for granule cell-selective damage, the damage would increase in a time-dependent manner. As stated in the “inflammation hypothesis,” we believe that the high sensitivity of granule cells to methylmercury is a factor that triggers the subsequent severe injury of granule cells by cytotoxic T-lymphocytes and macrophages.

In the “inflammation hypothesis,” the mechanisms by which cytotoxic T-lymphocytes and macrophages cause severe damages selectively to granule cells are thought to include a TNF-α storm, because granule cells are susceptible to TNF-α and both cell types have the ability to synthesize and secrete TNF-α (Vilcek and Lee, 1991; Idriss and Naismith, 2000). TNF-α is a cytokine that induces apoptosis. It is likely that methylmercury induces the synthesis and secretion of TNF-α in cytotoxic T-lymphocytes and macrophages. We confirmed this induction by methylmercury in cultured macrophage-like cells. The perforin-granzyme B system may be important for understanding the mechanisms underlying cell death of cerebellar granule cells by cytotoxic T lymphocytes. Upon contact with target cells, cytotoxic T-lymphocytes secrete perforin monomers that are stored in intracytoplasmic granules. The secreted perforin monomers aggregate on the target cells to form oligomers, which then form small pores on the target cells. Through these pores, cytotoxic T lymphocytes inject granzyme B into target cells, inducing apoptosis in the target cells (de Vries et al., 2007; Li, 2007). Another well-known apoptosis-inducing system involving cytotoxic T lymphocytes is the Fas-Fas L system (Andersen et al., 2006). Fas acts as a self-destructive switch in cells, and its stimulation induces apoptosis in them (Timmer et al., 2002). Cytotoxic T lymphocytes recognize abnormal cells and activate Fas molecules on the cell surface along with Fas L to induce apoptosis and death of target cells. Methyl mercury may activate this system.

As far as we know, our study is the first to suggest that inflammatory changes are involved in the pathogenesis of methylmercury-induced damage in the cerebellar granule cell layer. Specifically, it was suggested that in the cerebellum, methylmercury first causes small-scale damage to granule cells, which triggers the infiltration of cytotoxic T-lymphocytes and macrophages into the granule cell layer, which secrete TNF-α to induce apoptosis in granule cells. This chain is established by the susceptibility of granule cells to methylmercury, the ability of cytotoxic T lymphocytes and macrophages to synthesize and secrete TNF-α, and the sensitivity of granule cells to TNF-α and methylmercury. There is need for further research based the “inflammation hypothesis.” In particular, studies on the effects of methylmercury on the TNF-α synthesis in macrophages and the perforin-granzyme B system in cytotoxic T-lymphocytes appear to be important to clarify the molecular mechanisms underlying the “inflammation hypothesis.”

ACKNOWLEDGMENTS

This research was funded by the Study of the Health Effects of Heavy Metals, organized by the Ministry of the Environment, Japan (to Y.S. and T.K.).

Conflict of interest

The authors declare that there is no conflict of interest.

REFERENCES

•  Allen,  J.W.,  Shanker,  G.,  Tan,  K.H. and  Aschner,  M. (2002): The consequences of methylmercury exposure on interactive functions between astrocytes and neurons. Neurotoxicology, 23, 755-759.
•  Amenta,  F.,  Cavalotta,  D.,  Del Valle,  M.E.,  Mancini,  M.,  Sabbatini,  M.,  Torres,  J.M. and  Vega,  J.A. (1994): Calbindin D-28k immunoreactivity in the rat cerebellar cortex: age-related changes. Neurosci. Lett., 178, 131-134.
•  Andersen,  M.H.,  Schrama,  D.,  Thor Straten,  P. and  Becker,  J.C. (2006): Cytotoxic T cells. J. Invest. Dermatol., 126, 32-41.
•  Bellum,  S.,  Bawa,  B.,  Thuett,  K.A.,  Stoica,  G. and  Abbott,  L.C. (2007): Changes in biochemical processes in cerebellar granule cells of mice exposed to methylmercury. Int. J. Toxicol., 26, 261-269.
•  Celio,  M.R. (1990): Calbindin D-28k and parvalbumin in the rat nervous system. Neuroscience, 35, 375-475.
•  Clarkson,  T.W. and  Strain,  J.J. (2003): Nutritional factors may modify the toxic action of methyl mercury in fish-eating populations. J. Nutr., 133 (Suppl 1), 1539S-1543S.
•  de Vries,  J.F.,  von dem Borne,  P.A.,  van Luxemburg-Heijs,  S.A.,  Heemskerk,  M.H.,  Willemze,  R.,  Falkenburg,  J.H. and  Barge,  R.M. (2007): Differential activation of the death receptor pathway in human target cells induced by cytotoxic T lymphocytes showing different kinetics of killing. Haematologica, 92, 1671-1678.
•  Edwards,  J.R.,  Marty,  M.S. and  Atchison,  W.D. (2005): Comparative sensitivity of rat cerebellar neurons to dysregulation of divalent cation homeostasis and cytotoxicity caused by methylmercury. Toxicol. Appl. Pharmacol., 208, 222-232.
•  Eto,  K. (1997): Pathology of Minamata disease. Toxicol. Pathol., 25, 614-623.
•  Eto,  K.,  Takizawa,  Y.,  Akagi,  H.,  Haraguchi,  K.,  Asano,  S.,  Takahata,  N. and  Tokunaga,  H. (1999): Differential diagnosis between organic and inorganic mercury poisoning in human cases--the pathologic point of view. Toxicol. Pathol., 27, 664-671.
•  Eto,  K.,  Yasutake,  A.,  Kuwana,  T.,  Korogi,  Y.,  Akima,  M.,  Shimozeki,  T.,  Tokunaga,  H. and  Kaneko,  Y. (2001): Methylmercury poisoning in common marmosets--a study of selective vulnerability within the cerebral cortex. Toxicol. Pathol., 29, 565-573.
•  Eto,  K.,  Marumoto,  M. and  Takeya,  M. (2010): The pathology of methylmercury poisoning (Minamata disease): The 50th Anniversary of Japanese Society of Neuropathology. Neuropathology, 30, 471-479.
•  Fujiyama,  J.,  Hirayama,  K. and  Yasutake,  A. (1994): Mechanism of methylmercury efflux from cultured astrocytes. Biochem. Pharmacol., 47, 1525-1530.
•  Gatti,  R.,  Belletti,  S.,  Uggeri,  J.,  Vettori,  M.V.,  Mutti,  A.,  Scandroglio,  R. and  Orlandini,  G. (2004): Methylmercury cytotoxicity in PC12 cells is mediated by primary glutathione depletion independent of excess reactive oxygen species generation. Toxicology, 204, 175-185.
•  Herden,  C.J.,  Pardo,  N.E.,  Hajela,  R.K.,  Yuan,  Y. and  Atchison,  W.D. (2008): Differential effects of methylmercury on gamma-aminobutyric acid type A receptor currents in rat cerebellar granule and cerebral cortical neurons in culture. J. Pharmacol. Exp. Ther., 324, 517-528.
•  Hirooka,  T.,  Fujiwara,  Y.,  Minami,  Y.,  Ishii,  A.,  Ishigooka,  M.,  Shinkai,  Y.,  Yamamoto,  C.,  Satoh,  M.,  Yasutake,  A.,  Eto,  K. and  Kaji,  T. (2010a): Cell-density-dependent methylmercury susceptibility of cultured human brain microvascular pericytes. Toxicol. In Vitro, 24, 835-841.
•  Hirooka,  T.,  Fujiwara,  Y.,  Shinkai,  Y.,  Yamamoto,  C.,  Yasutake,  A.,  Satoh,  M.,  Eto,  K. and  Kaji,  T. (2010b): Resistance of human brain microvascular endothelial cells in culture to methylmercury: cell-density-dependent defense mechanisms. J. Toxicol. Sci., 35, 287-294.
•  Idriss,  H.T. and  Naismith,  J.H. (2000): TNF alpha and the TNF receptor superfamily: structure-function relationship(s). Microsc. Res. Tech., 50, 184-195.
•  Kadowaki,  K.,  McGowan,  E.,  Mock,  G.,  Chandler,  S. and  Emson,  P.C. (1993): Distribution of calcium binding protein mRNAs in rat cerebellar cortex. Neurosci. Lett., 153, 80-84.
•  Kaur,  P.,  Aschner,  M. and  Syversen,  T. (2006): Glutathione modulation influences methyl mercury induced neurotoxicity in primary cell cultures of neurons and astrocytes. Neurotoxicology, 27, 492-500.
•  Kromidas,  L.,  Trombetta,  L.D. and  Jamall,  I.S. (1990): The protective effects of glutathione against methylmercury cytotoxicity. Toxicol. Lett., 51, 67-80.
•  Leyshon-Sørland,  K.,  Jasani,  B. and  Morgan,  A.J. (1994): The localization of mercury and metallothionein in the cerebellum of rats experimentally exposed to methylmercury. Histochem. J., 26, 161-169.
•  Li,  Q. (2007): New mechanism of organophosphorus pesticide-induced immunotoxicity. J. Nippon Med. Sch., 74, 92-105.
•  Li,  S.,  Thompson,  S.A. and  Woods,  J.S. (1996): Localization of gamma-glutamylcysteine synthetase mRNA expression in mouse brain following methylmercury treatment using reverse transcription in situ PCR amplification. Toxicol. Appl. Pharmacol., 140, 180-187.
•  Limke,  T.L.,  Bearss,  J.J. and  Atchison,  W.D. (2004): Acute exposure to methylmercury causes Ca²⁺ dysregulation and neuronal death in rat cerebellar granule cells through an M3 muscarinic receptor-linked pathway. Toxicol. Sci., 80, 60-68.
•  Miura,  K. and  Clarkson,  T.W. (1993): Reduced methylmercury accumulation in a methylmercury-resistant rat pheochromocytoma PC12 cell line. Toxicol. Appl. Pharmacol., 118, 39-45.
•  Polunovsky,  V.A.,  Wendt,  C.H.,  Ingbar,  D.H.,  Peterson,  M.S. and  Bitterman,  P.B. (1994): Induction of endothelial cell apoptosis by TNF α: modulation by inhibitors of protein synthesis. Exp. Cell Res., 214, 584-594.
•  Randall,  A. and  Tsien,  R.W. (1995): Pharmacological dissection of multiple types of Ca²⁺ channel currents in rat cerebellar granule neurons. J. Neurosci., 15, 2995-3012.
•  Sato,  M. and  Kondoh,  M. (2002): Recent studies on metallothionein: protection against toxicity of heavy metals and oxygen free radicals. Tohoku J. Exp. Med., 196, 9-22.
•  Simmons-Willis,  T.A.,  Koh,  A.S.,  Clarkson,  T.W. and  Ballatori,  N. (2002): Transport of a neurotoxicant by molecular mimicry: the methylmercury-L-cysteine complex is a substrate for human L-type large neutral amino acid transporter (LAT) 1 and LAT2. Biochem. J., 367, 239-246.
•  Straka,  E.,  Ellinger,  I.,  Balthasar,  C.,  Scheinast,  M.,  Schatz,  J.,  Szattler,  T.,  Bleichert,  S.,  Saleh,  L.,  Knöfler,  M.,  Zeisler,  H.,  Hengstschläger,  M.,  Rosner,  M.,  Salzer,  H. and  Gundacker,  C. (2016): Mercury toxicokinetics of the healthy human term placenta involve amino acid transporters and ABC transporters. Toxicology, 340, 34-42.
•  Tanaka,  O.,  Sakagami,  H. and  Kondo,  H. (1995): Localization of mRNAs of voltage-dependent Ca(²⁺)-channels: four subtypes of alpha 1- and beta-subunits in developing and mature rat brain. Brain Res. Mol. Brain Res., 30, 1-16.
•  Timmer,  T.,  de Vries,  E.G. and  de Jong,  S. (2002): Fas receptor-mediated apoptosis: a clinical application? J. Pathol., 196, 125-134.
•  Vilcek,  J. and  Lee,  T.H. (1991): Tumor necrosis factor. New insights into the molecular mechanisms of its multiple actions. J. Biol. Chem., 266, 7313-7316.
•  Yao,  C.P.,  Allen,  J.W.,  Mutkus,  L.A.,  Xu,  S.B.,  Tan,  K.H. and  Aschner,  M. (2000): Foreign metallothionein-I expression by transient transfection in MT-I and MT-II null astrocytes confers increased protection against acute methylmercury cytotoxicity. Brain Res., 855, 32-38.