2022 Volume 47 Issue 5 Pages 211-219
Methylmercury (MeHg), an environmentally toxic substance, causes site-specific neuronal cell death; while MeHg exposure causes death in cerebrocortical neurons, interestingly, it does not in hippocampal neurons, which are generally considered to be vulnerable to toxic substances. This phenomenon of site-specific neuronal cell death can be reproduced in animal experiments; however, the mechanism underlying the resistance of hippocampal neurons to MeHg toxicity has not been clarified. In this study, we comparatively analyzed the response to MeHg exposure in terms of viability and the expression characteristics of primary cultured cerebrocortical neurons and hippocampal neurons derived from fetal rat brain. Neuronal differentiated hippocampal neurons were more resistant to MeHg toxicity than cerebrocortical neurons, as indicated by a 2‒3 fold higher half-maximal inhibitory concentration (IC50; 3.3 μM vs. 1.2 μM), despite similar intracellular mercury concentrations in both neuronal cell types. Comprehensive RNA sequencing-based gene expression analysis of non-MeHg-exposed cells revealed that 80 out of 15,208 genes showed at least 10-fold higher expression in hippocampal neurons than in cerebrocortical neurons, whereas six genes showed at least 10-fold higher expression in cerebrocortical neurons than in hippocampal neurons. In particular, genes related to neuronal function, including those encoding transthyretin and brain-derived neurotrophic factor, showed approximately 50-fold higher expression in hippocampal neurons than in cerebrocortical neurons. In conclusion, the resistance of hippocampal neurons to MeHg toxicity may be related to the high expression of neuronal function-related proteins.
Methylmercury (MeHg) is an environmental toxicant that causes neurological and developmental impairments (Bakir et al., 1973; Takeuchi, 1982). MeHg intoxication in human manifests in two characteristic clinical forms: fetal-type and adult-type Minamata disease. Fetal-type Minamata disease caused by exposure to MeHg in utero is characterized by extensive brain lesions, whereas the adult-type caused by MeHg intoxication during adulthood causes site-specific brain lesions in the cerebrum and cerebellum. Autopsy studies of human cerebrum have revealed that the lesions are localized in the deep layer of cerebrocortical neurons and cerebellar granule cells, respectively (Eto and Takeuchi, 1978; Eto et al., 1999). Such site-specific cerebral and cerebellar lesions have also been observed in MeHg-intoxicated animal models (Sager et al., 1984; Wakabayashi et al., 1995; Fujimura and Usuki, 2017a, 2017b; Fujimura et al., 2019). However, MeHg exposure also induces some damage to other parts including the shallow layers of cerebrocortical neurons, cerebellar Purkinje cells, and peripheral motor nerves etc. The adult hippocampus, as part of the cerebrum, is protected from the effects of MeHg intoxication, although it is known to be affected by various insults, such as ischemia, inflammation, hypoglycemia, or excessive metabolic demand during epileptic activity, resulting in functional and structural neuronal integrity impairment (Schmidt-Kastner and Freund, 1991; Haces et al., 2010; Sloviter, 1989). Although hippocampal neurons are highly vulnerable to various external stimuli (Newell et al., 1990; Vornov et al., 1991; Wilde et al., 1997; Wang and Michaelis, 2010), MeHg exposure has no impact on this area (Fujimura et al., 2009; Fujimura and Usuki, 2017a, 2017b) in MeHg-intoxicated animal models. However, the molecular mechanisms underlying the exclusion of adult hippocampus from MeHg intoxication remain to be elucidated. We believe that exploring this characteristic property will lead to preventive medical research on MeHg toxicity. Based on the above, in this study, we focused on the vulnerability of hippocampal neurons.
RNA sequencing (RNA-seq) is a next-generation sequencing technology that can be used to detect and quantify RNA in a biological sample at a given time point to analyze the continuously changing cellular transcriptome (Wang et al., 2009; Chu and Corey, 2012). Specifically, RNA-seq facilitates the study of alternatively spliced transcripts, post-transcriptional modifications, gene fusions, mutations/single-nucleotide polymorphisms, and changes in gene expression over time, as well as differential gene expression among groups or treatments (Maher et al., 2009). Prior to RNA-seq, gene expression was mostly studied using hybridization-based microarrays. However, microarrays have limitations, including cross-hybridization artifacts, poor quantification of lowly and highly expressed genes, and the need for a priori sequence knowledge (Kukurba and Montgomery, 2015). Therefore, RNA-seq is increasingly being used in gene expression studies.
In the present study, we prepared primary cultures of rat cerebrocortical or hippocampal neurons isolated from rat embryo and used these to compare their vulnerability/resistance to MeHg exposure and for a comprehensive gene expression analysis of both neuronal cell types, using RNA-seq, with the aim to preliminary unravel the mechanism underlying site-specific neuronal cell death.
Primary cultures of rat cerebrocortical and hippocampal neurons were established using a modified version of the protocol described by Unoki et al. (2012) and Fujimura and Usuki (2015). In brief, the cerebral cortex and hippocampus were dissected from 20-day-old Sprague-Dawley rat embryos (Clear Japan, Tokyo, Japan). The tissues were digested with papain (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) at 37°C for 30 min and dissociated by gentle pipetting. The suspensions were filtered and centrifuged at 110 × g at room temperature for 5 min, and the cells were cultured in polyethyleneimine (Merck, Darmstadt, Germany)-coated wells in tissue-culture plates at a density of 4 × 104 cells/cm2 in neurobasal medium supplemented with B27 and 2 mM GlutaMAX (Thermo Fisher Scientific, Waltham, MA, USA). Culture in this serum-free medium reportedly suppresses the growth of glial cells by more than 99.5%. Seven days after plating, the cells were exposed to MeHg chloride (Tokyo Chemical Industry, Tokyo, Japan). MeHg chloride was dissolved in dimethyl sulfoxide (DMSO; Nacalai Tesque, Kyoto, Japan) and added to the medium at a DMSO concentration of 0.1% (Fujimura and Usuki, 2018).
All experiments were carried out in accordance with the Guide for the Care and Use of Laboratory Animals issued by the National Institute for Minamata Disease.
Cell viability assayCell viability was assayed using a CellTiter 96® Aqueous One Solution Cell Proliferation Assay kit (Promega, Madison, WI, USA) containing 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl-2H-tetrazolium, inner salt) (MTS), which is a substrate for intracellular dehydrogenases. The measurement was performed using a microplate reader (Bio-Rad Laboratories, Hercules, CA, USA) with an absorbance of 490 nm. We confirmed that MeHg did not directly interfere with the assay (data not shown).
Measurement of intracellular mercury concentrationsFor the measurement of intracellular mercury (Hg) concentrations, cells were prepared using a modified version of the protocol described by (Usuki et al., 2017). In brief, cerebrocortical and hippocampal neurons were washed twice with Ca2+- and Mg2+-free phosphate-buffered saline and incubated in cell lysis buffer M (FUJIFILM Wako Pure Chemical Corporation) for 10 min after MeHg exposure for the indicated times. Total Hg concentrations in cell lysates were determined by the oxygen combustion-gold amalgamation method using a mercury analyzer (MA2000; Nippon Instruments, Tokyo, Japan) as previously described (Fujimura et al., 2012). The protein content was measured using a DC Protein Assay kit (Bio-Rad Laboratories). The measurement was performed using a microplate reader (Bio-Rad Laboratories) with an absorbance of 620 nm.
RNA isolationRNA isolation and sequencing were conducted using a modified version of the protocol described by Martin and Wang (2011). In brief, total RNA was isolated from cerebrocortical or hippocampal neurons with genomic DNA removal using a RNeasy Plus Mini kit (Qiagen, Venlo, Netherlands) according to the manufacturer’s instructions. RNA quality and quantity were measured using a NanoDrop ND-2000 spectrophotometer (Thermo Fisher Scientific).
RNA seqAn mRNA-seq library was constructed using the TruSeqTM RNA Sample Prep Kit (Illumina Inc., San Diego, CA, USA). In brief, the poly-A-containing mRNA molecules were isolated using poly-T oligo-attached magnetic beads, followed by thermal mRNA fragmentation. The mRNA fragments were then transcribed into first-strand cDNA using reverse transcriptase (Thermo Fisher Scientific) and random primers. Second-strand cDNA was synthesized using DNA Polymerase I and RNase H (Illumina Inc.). After end-repair, single ‘A’ bases were added to the fragments and adapters were ligated to prepare the cDNA for hybridization onto a flow cell. Finally, the products were purified and enriched by PCR to create the cDNA library (Macrogen, Seoul, Korea) using NovaSeq 6000 (Illumina Inc.). Transcripts per million (TPM) values were calculated and normalized by quantile normalization.
Statistical analysisStatistical significance was determined by one-way analysis of variance, followed by Dunnett’s multiple comparison test. Data are expressed as the mean ± SEM. Differences were considered statistically significant when p < 0.05.
The viability of cerebrocortical and hippocampal neurons was analyzed 24 hr after exposure to MeHg at concentrations ranging from 0.25 to 10 μM (Fig. 1). The hippocampal neurons were more resistant to MeHg than the cerebrocortical neurons (half-maximal inhibitory concentration (IC50) for cerebrocortical neurons: 1.2 μM, IC50 for hippocampal neurons: 3.3 μM).
Cell viability at 24 hr after treatment with a range of MeHg concentrations as analyzed by the MTS assay. Data are relative to those for untreated cells and represent the mean ± SEM (n = 6). *p < 0.05, **p < 0.01 vs untreated cells.
Total mercury concentrations in cerebrocortical and hippocampal neurons 24 hr after exposure to MeHg at 1 and 2.5 μM) are shown in Fig. 2. The intracellular concentrations of total mercury were nearly similar in both neuron types (cerebrocortical neurons: 102 and 208 mg/g protein for MeHg exposure at 1 and 2.5 μM, respectively, hippocampal neurons: 107 and 216 mg/g protein, respectively).
Intracellular mercury concentration in primary cultured cerebrocortical and hippocampal neurons at 24 hr after MeHg exposure. Data represent the mean ± SEM (n = 3).
To explore the resistance factors to MeHg toxicity in hippocampal neurons, we performed RNA-seq of both cell types without exposure to MeHg. The RNA-seq results revealed that 80 out of 15,208 genes, excluding unknown genes, showed at least 10-fold higher expression in hippocampal neurons than in cerebrocortical neurons (Table 1). Various genes, including genes related to nerve function, chemokines, cell growth, ion channels, cytokines, and the complement system, were specifically expressed in hippocampal neurons. Among them, the genes encoding transthyretin and brain-derived neurotrophic factor (BDNF) were expressed approximately 50-fold higher in hippocampal neurons than in cerebrocortical neurons. Six genes, including genes related to autophagy, nucleation, glycoproteins, and the cytoskeleton, showed at least 10-fold higher expression in cerebrocortical neurons than in hippocampal neurons (Table 2). None of these genes were related to nerve fragility. All RNA-seq analysis data are attached to this paper (Supplement file 1).
| Gene name | Hi (TPM) | CC (TPM) | Hi/CC | Fuction |
|---|---|---|---|---|
| Transthyretin | 39 | 0.5 | 77 | Biological antioxidant |
| Brain-derived neurotrophic factor | 62.9 | 1.3 | 48.3 | Nerve cells maintenance |
| C-C motif chemokine ligand-7 | 14 | 0.3 | 46.6 | Inflammatory chemokine |
| Tetraspanin-18 | 20.4 | 0.5 | 43.8 | Growth factor related protein |
| Transient RC potential cation channel | 8.5 | 0.2 | 42.4 | Potential channel |
| Sosondowah ankyrin repeat domain F, M-B | 3.2 | 0.1 | 40.8 | Ion channel related protein |
| Interleukin-16 | 6.6 | 0.2 | 36.2 | Inflammatory cytokine |
| Complement CN-1, q SCN, C-chain | 16.6 | 0.5 | 34.8 | Complement component |
| FRAS1 relted extracellular matrix-3 | 1.8 | 0.1 | 34.1 | Extracellular matrix protein |
| Family with sequence similarity-150, M-B | 14.1 | 0.4 | 32.4 | Pseudogene |
| Complement CN-1, q SCN, A-chain | 12.2 | 0.4 | 31 | Complement component |
| C-C-motif chemokine ligand-2 | 86.6 | 2.8 | 30.6 | Inflammatory chemokine |
| Colony stimulating factor-1 RC | 4.1 | 0.1 | 30.6 | Inflammatory cytokine receptor |
| Complement CN 1, q SCN, B chain | 16 | 0.5 | 30 | Complement component |
| Serpin F-G, M-1 | 9.4 | 0.3 | 29.4 | Protease inhibitor |
| Fibrinogen C domain containing-1 like-1 | 8.4 | 0.3 | 29.1 | Hemostasis factor |
| Fibrinogen C domain containing-1 | 9.2 | 0.3 | 29 | Hemostasis factor |
| Neuronal PAS domain protein-4 | 261.8 | 9 | 29 | Regulating neuronal plasticity |
| Frizzled-related protein | 27.1 | 1 | 27.3 | Nerve synapse formation |
| Lysosomal protein trans-membrane-5 | 3.6 | 0.1 | 27 | Lysosome component |
| Interleukin-20 RC subunit-beta | 28.3 | 1.1 | 24.9 | Inflammatory cytokine |
| Adhesion G protein-coupled RC-E1 | 2.7 | 0.1 | 24.1 | Adhesion protein receptor |
| Kinase insert domain RC | 2.9 | 0.1 | 23.6 | VEGF receptor |
| Sterile alpha motif domain containing-15 | 4 | 0.2 | 23 | RNA-binding protein |
| Family with sequence similarity-46, M-B | 1.6 | 0.1 | 20.8 | Pseudogene |
| Claudin-1 | 2.2 | 0.1 | 20.6 | Membrane protein |
| G protein-coupled RC, Class-C, G-5, M-A | 6.4 | 0.3 | 20.5 | Amino acid binding G protein |
| Solute carrier F-9 M-A4 | 5.1 | 0.3 | 19.3 | Membrane transport protein |
| Lymphoid enhancer binding factor-1 | 3.6 | 0.2 | 18.9 | Lymphoid enhancer |
| Toll-like RC-2 | 1 | 0.1 | 17.4 | Immune system receptor |
| Armadillo repeat containing-3 | 0.7 | 0.1 | 16.8 | Interactions between proteins |
| Collagen T-1, alpha-2, transcript variant-XI | 2.9 | 0.2 | 16.3 | Skin component |
| Carbonic anhydrase-3 | 4.7 | 0.3 | 16.2 | Carbonic anhydrase |
| Chemokine (C-X-C motif) ligand-1 | 6.3 | 0.4 | 16.2 | Inflammatory cytokine |
| Plakophilin-2 | 5.1 | 0.3 | 16 | Cytoskeleton |
| Plasminogen activator, urokinase ret proto-oncogene | 7 | 0.7 | 16 | Hemostasis factor |
| Collagen T-III, alpha-2 chain | 17.6 | 1.1 | 16 | Endocrine system receptor |
| Ret proto-oncogene | 3.3 | 0.2 | 15.9 | Skin component |
| Nuclear RC SF-4, G-4, M-1 | 412.7 | 25.9 | 15.9 | Nuclear receptor |
| Cytochrome P450, F-26, SF-b, poly-peptide-1 | 16.7 | 1.1 | 15.5 | Metabolic enzyme |
| Cadherin-related F M-4, transcript varient-XI | 6 | 0.4 | 15.3 | Cell adhesion molecule |
| Macrophage expressed-1 | 1.5 | 0.1 | 15 | Macrophage |
| Neuritin-1 | 100 | 6.8 | 14.7 | Neuro-development |
| Prospero homeobox-1 | 26.8 | 1.9 | 14.2 | Marker for lymphatic endothelium |
| Basic helix-loop-helix F, M-e23 | 12.8 | 0.9 | 13.9 | DNA binding protein |
| GTP binding protein over-expressed in skeletal muscle | 19.7 | 1.4 | 13.7 | GTP binding protein |
| Nuclear RC SF-4, G-A, M-3 | 229.5 | 17 | 13.5 | Nuclear receptor |
| Alpha-2-macroglobulin-like | 22.2 | 1.6 | 13.4 | Autoimmune disease related protein |
| Solute carrier F-2 M-9 | 1.4 | 0.1 | 13.4 | Membrane transport protein |
| Chloride intracellular channel-6 | 3.4 | 0.3 | 13.3 | Ion channel |
| MAM domain containing-2, transcript variant-XI | 2.5 | 0.2 | 13.3 | Cancer related protein |
| Asparaginase | 3 | 0.2 | 13.2 | Converting enzyme |
| Interleukine 1 RC T-1 | 2 | 0.2 | 13.2 | Inflammatory cytokine receptor |
| Ras homolog F, M-J | 13.3 | 1 | 13 | Cytoskeletal dynamics regulator |
| Rho GTPase activating protein-8 | 6.6 | 0.5 | 13 | Cytoskeletal dynamics regulator |
| Nuclear RC SF-4, G-A, M-2 | 57.4 | 4.4 | 12.9 | Nuclear receptor |
| Neuronal pentraxin-1 | 95.4 | 7.5 | 12.8 | Excitatory synapse formation |
| FOS like 1, AP-1 transcription factor SU | 24.2 | 2 | 12.4 | Transcription factor |
| Neuronal pentraxin-2 | 304.4 | 25.1 | 12.1 | Excitatory synapse formation |
| Carbonic anhydrase-12 | 16.8 | 1.4 | 12.1 | Metallo-enzyme |
| Retinal degeneration-3-like | 8.3 | 0.7 | 12 | Retinal protein |
| Calmodulin-like-4 | 11.5 | 1 | 11.6 | Calcium binding protein |
| Similar to guanylate binding protein F, M-6, transcript variant-XI | 5.7 | 0.5 | 11.6 | Pseudogene |
| Radial spoke head-4 homolog-A | 1.1 | 0.1 | 11.5 | Flagellum/cilium related protein |
| Matrix Gla protein | 148.8 | 12.9 | 11.5 | Calcium binding protein |
| Tumor necrosis factor RC superF, M-25 | 1.7 | 0.2 | 11.5 | Cytokine receptor |
| Cd4 molecule | 1.5 | 0.1 | 11.4 | Immune receptor |
| Oncostatin M RC | 7.6 | 0.7 | 11.3 | Cytokine receptor |
| Serpine F-E, M-1 | 1.4 | 0.1 | 11.3 | Protease inhibitor |
| CollagenT-IV alpha-6-chain, transcript variant-XI | 1.9 | 0.2 | 11.2 | Skin component |
| Ring finger protein-39 | 28.7 | 2.6 | 11.1 | Ligase component |
| Collagen T-III, alpha-1 chain | 3.8 | 0.3 | 11.1 | Skin component |
| Interferon induced trans-membrane protein-3 | 47 | 4.3 | 11 | Antiviral protein |
| Oxytocin RC | 2.9 | 0.3 | 10.9 | Peptide hormone |
| Msh homeobox-1 | 8.3 | 0.8 | 10.9 | Transcriptional repressor |
| Prolactin releasing hormone RC | 5.3 | 0.5 | 10.9 | Hormone releasing factor |
| Cysteine and serine rich nuclear protein-1 | 39.3 | 3.6 | 10.9 | Nuclear protein |
| Similar to protein C8orf4 | 8.3 | 0.8 | 10.8 | Pseudogene |
| Otoferlin | 5 | 0.5 | 10.4 | Calcium binding protein |
| RAS-like, estrogen-regulated, growth-inhibitor, transcript variant-XII | 28.8 | 2.8 | 10.3 | Cytoskeletal dynamics regulator |
TPM, transcripts per million; CC, cerebrocortical neurons; Hi, Hippocampal neurons; F, family; M, member; CN, component; SCN, subcomponent; RC, receptor; G, group; SF, subfamily; T, type.
| Gene name | CC (TPM) | Hi (TPM) | CC/Hi | Fuction |
|---|---|---|---|---|
| SCAN domain containing-3, pseudogene 1 | 74.2 | 0.9 | 82.4 | Pseudogene |
| SATB homeobox-2 | 48.7 | 3.3 | 14.8 | DNA-binding protein |
| Heparan sulfate-glucosamine-3 | 20.2 | 1.5 | 13.5 | Bone component |
| Collagen, T-XI, alpha-2 | 26.7 | 2 | 13.4 | Skin component |
| Ribosomal protein S-2, pseudogene-1 | 1.2 | 0.1 | 12 | Pseudogene |
| G protein-coupled RC-6 | 22.2 | 1.9 | 11.4 | Neurite outgrowth |
TPM, transcripts per million; CC, cerebrocortical neurons; Hi, Hippocampal neurons; T, type; RC, receptor.
Next, we performed RNA-seq of both cell types with exposure to MeHg at concentrations ranging from 0.5 to 2.5 μM for 1 hr. In this evaluation system, MeHg exposure to cultured neurons showed little drastic change in the gene expression. In both cell types, 7 out of 15,208 genes, excluding unknown genes, were found to be upregulated at least 5-fold by exposure to MeHg (Table 3). In contrast, only one pseudogene (SCAN domain containing 3, pseudogene 1) decreased by 5 times or more. However, these genes were not directly related to neural function, except for transthyretin. In genes related to nerve function, the expression of BDNF gene was not altered by MeHg exposure in both cell types (Table 3). On the other hand, the transthyretin gene was elevated only in the hippocampal neurons by exposure to MeHg. All RNA-seq analysis data are attached to this paper (Supplement file 1).
| Gene name | Hi MeHg 0 nM |
Hi MeHg 500 nM |
Hi MeHg 1000 nM |
Hi MeHg 2500 nM |
CC MeHg 0 nM |
CC MeHg 500 nM |
CC MeHg 1000 nM |
CC MeHg 2500 nM |
Function | |
|---|---|---|---|---|---|---|---|---|---|---|
| Transthyretin | TPM (% of basal) |
39 | 39.7 | 57.8 | 39 | 0.5 | 0.3 | 1.6 | 3.5 | Biological antioxidant |
| 100 | 101.8 | 148.2 | 100 | 100 | 58.5 | 307.7 | 684.7 | |||
| UDP glucuronosyl-transferase 1F, polypeptide A7C | TPM (% of basal) |
0.2 | 0.7 | 1.7 | 0.4 | 0.6 | 2 | 0.7 | 0.6 | Chemical defense system |
| 100 | 420.8 | 952 | 254.1 | 100 | 349.8 | 115.8 | 91.4 | |||
| Coiled-coil-helix-coiled-coil-helix domain containing 2 | TPM (% of basal) |
61.2 | 15.4 | 67 | 65.2 | 11.4 | 65.9 | 58.2 | 60.3 | Mitochondorial protein |
| 100 | 25.2 | 109.4 | 106.5 | 100 | 578.1 | 510.5 | 528.9 | |||
| Ribosomal protein L37a, pseudogene 1 | TPM (% of basal) |
0.1 | 1.2 | 0.6 | 1 | 1.2 | 0.7 | 1.1 | 1.4 | Pseudogene |
| 100 | 1558.4 | 831.1 | 1307.8 | 100 | 61 | 93.9 | 116.9 | |||
| SCAN domain containing 3, pseudogene 1 | TPM (% of basal) |
0.9 | 71.5 | 6.7 | 7.3 | 74.2 | 1.1 | 8.4 | 1 | Pseudogene |
| 100 | 8372.4 | 786.9 | 859.4 | 100 | 1.5 | 11.3 | 1.3 | |||
| MicroRNA 3593 | TPM (% of basal) |
9.7 | 6.8 | 4.9 | 6.5 | 2.1 | 7.8 | 11.4 | 9 | MicroRNA |
| 100 | 69.8 | 50.5 | 67.2 | 100 | 368.1 | 534.3 | 423.3 | |||
| C-type lectin domain F3, MB | TPM (% of basal) |
0.8 | 0.8 | 0.7 | 0.7 | 0.1 | 0.1 | 0.6 | 0.3 | Glycoprotein |
| 100 | 102.3 | 89.5 | 89.6 | 100 | 116.7 | 609 | 293.9 | |||
| Brain-derived neurotrophic factor | TPM (% of basal) |
62.9 | 58.5 | 60.4 | 62.5 | 1.3 | 1.8 | 1.7 | 1.8 | Nerve cells maintenance |
| 100 | 93 | 96 | 99.4 | 3.3 | 138.5 | 130.8 | 139.5 |
TPM, transcripts per million; CC, cerebrocortical neurons; Hi, Hippocampal neurons.
It is well known that hippocampal neurons are not damaged in adult Minamata disease. The resistance of hippocampal neurons to MeHg toxicity has been reproduced in MeHg-intoxicated animal models (Fujimura and Usuki, 2017a, 2017b; Fujimura et al., 2019). Although various studies have used animal models to investigate the mechanism underlying the resistance phenomenon, it has not been clarified to date. In the present study, we demonstrated, for the first time, that hippocampal neurons are more resistant to MeHg toxicity than cerebrocortical neurons using isolated cultured primary neurons (Fig. 1). As the intracellular mercury concentrations were similar in both neuronal cell types, the difference in resistance is not due to intracellular translocation of MeHg (Fig. 2).
We evaluated the possible contribution of differential gene expression to the site-specific neurotoxicity using RNA-seq of isolated and cultured cerebrocortical and hippocampal neurons. Numerous gene expression analyses using cerebral cortex and hippocampus from experimental animals have been performed; however, some studies pointed out that there are limited differences in gene expression between both regions in the basal state (Large et al., 1986; Rall et al., 2003). It has been suggested that this is because when a piece of animal brain tissue is used for analysis, it not only contains nerve cells, but also glial cells and stromal cells. To solve this problem, we cultured primary neurons in a medium suppressing the growth of glial cells and stromal cells and used these for genetic analysis. Our analysis of isolated cultured neurons revealed that genes related to neural functional maintenance were specifically expressed in hippocampal neurons, which may be resistant to MeHg toxicity (Table 1). The most notably differentially expressed genes were those encoding transthyretin, which is a biological antioxidant, and BDNF, which is a neurotrophic factor.
Furthermore, gene expression analysis exposed to MeHg indicated that transthyretin gene was only elevated in cerebrocortical neurons, but not in hippocampal neurons (Table 3). Since cerebrocortical neurons are vulnerable to MeHg toxicity, it is considered that the elevation of this gene does not play a major role in the vulnerability/resistance to MeHg toxicity. On the other hand, BDNF gene showed no drastic changes in both cells. In our previous animal studies (Fujimura and Usuki, 2017a), exposure to MeHg increased BDNF in the cerebral cortex. It was suggested that BDNF increased by MeHg in animal experiments was derived from glial cells other than nerve cells. In addition, almost no dramatic changes in gene expression due to MeHg were observed in both cultured neurons (Table 3). Since this analysis used only nerve cells, it is considered that no inflammatory factors increased by exposure to MeHg were observed.
Transthyretin is a transport protein found in the serum and cerebrospinal fluid that carries the thyroid hormone thyroxine and retinol-binding protein bound to retinol (this is how transthyretin received its name; transports thyroxine and retinol). The liver secretes transthyretin into the blood, and the choroid plexus secretes transthyretin into the cerebrospinal fluid (Dickson et al., 1985). It has been demonstrated that transthyretin has protective effects on nerve tissues. Continuous administration of the ω3 system to Alzheimer’s disease patients that had induced memory impairment promoted transthyretin expression mainly in the hippocampus and improved memory impairment (Faxén-Irving et al., 2013). Transthyretin is rarely expressed in normal neurons that have lost their proliferative capacity during the differentiation process. However, it is considered that it remains present in the hippocampal neurons that retain their proliferative ability (Zhao et al., 2008).
BDNF acts on certain neurons of the central and the peripheral nervous systems, supporting the survival of existing neurons and promoting the growth and differentiation of new neurons and synapses (Acheson et al., 1995; Huang and Reichardt, 2001). In the brain, it is active in the hippocampus, cortex, and basal forebrain—areas vital to learning, memory, and higher thinking (Yamada and Nabeshima, 2003). The protective effects of BDNF or its pathway activation on glutamate-induced neurotoxicity in rat hippocampal neurons has been reported (Almeida et al., 2005; Chu and Corey, 2012). Further, BDNF treatment ameliorates cell loss in the cerebral cortex of Alzheimer’s disease model mice (Nagahara et al., 2013). Previous papers reported that MeHg exposure changes the expression levels of BDNF in cerebral cortex of mice (Fujimura and Usuki, 2017a), and that forced expression of BDNF strongly inhibited MeHg-induced cell death in cerebrocortical neurons (Guida et al., 2017). BDNF expression is enhanced upon nerve cell activation (Hofer et al., 1990). Hippocampal neurons are considered to have higher BDNF expression because they have higher neural activity than cerebral cortical cells.
We plan to evaluate the functions of the MeHg neurotoxicity resistance/vulnerability factors identified in this study at the protein level in a future study.
We are grateful to Michiko Fuchigami and Shiori Fukuhama for their excellent technical assistance.
Conflict of interestThe authors declare that there is no conflict of interest.
