2.1. Usefulness of GT1-7 cells in the study of Zn2+-induced neurotoxicity
It has been recognised that abnormalities in Zn homeostasis (e. g., excess or deficiency) are involved in neurological diseases such as VD [
2]. Understanding the molecular mechanism of neuronal cell death induced by Zn
2+, which accounts for a large amount of Zn in the brain, is important for elucidating the pathogenesis of VD and developing therapeutic agents. However, it is difficult to examine Zn
2+-induced neuronal cell death using primary cultured neurones of the cerebral cortex and hippocampus, which are used to study cell death in the brain, or PC-12 cells, a pheochromocytoma cell line. Zn
2+ is released along with glutamate upon glutamatergic neuronal excitation [
20]. As glutamate also causes neurotoxicity, it is difficult to distinguish between the effects of Zn
2+ and glutamate in cells with glutamate receptors. We have shown that Zn
2+ causes cell death in immortalised hypothalamic neurone GT1-7 cells [
21,
22]. In these studies, GT1-7 cells were more vulnerable to Zn
2+ cytotoxicity than other neuronal cells, including primary cultures of rat cortical and hippocampal neurones, PC-12 cells, and B-50 cells. (
Figure 2). GT1-7 cells possess neuronal characteristics such as neurite extension, secretion of gonadotropin-releasing hormone (GnRH), and expression of neurone-specific proteins and receptors, including microtubule-associated protein 2, tau protein, neurofilament, synaptophysin, GABA
A receptors, dopamine receptors, and L-type Ca
2+ channels [
23]. In contrast, GT1-7 cells lack or possess low levels of ionotropic glutamate receptors and exhibit little cytotoxicity to glutamate [
24]. These properties suggest that the GT1-7 cell line is an excellent model for investigating Zn
2+-induced neurotoxicity.
Figure 2.
Apoptotic death of GT1-7 cells after exposure to Zn2+. Viability of various neuronal cells after exposure to Zn2+. Cultured neuronal cells (GT1-7, PC-12, and B-50 cells [a neuroblastoma cell line], primary cultured neurones of the rat cerebral cortex, and primary cultured neurones of the rat hippocampus) were administered 50 μM of ZnCl2. After 24 h, cell viability was analysed by the WST-1 method.
Figure 2.
Apoptotic death of GT1-7 cells after exposure to Zn2+. Viability of various neuronal cells after exposure to Zn2+. Cultured neuronal cells (GT1-7, PC-12, and B-50 cells [a neuroblastoma cell line], primary cultured neurones of the rat cerebral cortex, and primary cultured neurones of the rat hippocampus) were administered 50 μM of ZnCl2. After 24 h, cell viability was analysed by the WST-1 method.
2.2. Molecular mechanism of Zn2+-induced GT1-7 cytotoxicity
Zn
2+-treated GT1-7 cells are positive for transferase-mediated biotinylated UTP nick-end labelling, indicating that Zn
2+ induces apoptosis in GT1-7 cells [
21,
22]. Studies using DNA microarray analysis and real-time PCR (RT-PCR) have revealed that the administration of Zn
2+ to GT1-7 cells induces the expression of various genes, including in metal-related genes (Zn transporter 1 [
ZnT-1], metallothionein [
MT]
1), and
MT2), ER stress-related genes, and Ca
2+ signalling transmission-related genes [
25]. The administration of Zn
2+ to GT1-7 cells also increased intracellular Ca
2+ levels ([Ca
2+]
i). Apoptosis is inhibited by pyruvate, citrate, and Ca
2+ channel antagonists (nifedipine, conotoxin, and Al
3+) [
21,
22,
26]. Using a high-resolution multisite video imaging system with Fura-2 as a fluorescent probe for cytosolic Ca
2+, we observed the changes in [Ca
2+]
i after exposure to Zn
2+ [
22]. This revealed that pretreatment with Al
3+, which acts as various kinds of Ca
2+ channel blocker [
27], suppressed the elevation of the [Ca
2+]
i levels in Zn
2+-treated GT1-7 cells. Al has difficulty entering cells without a membrane-permeable chelator, and Al
3+ does not affect the viability of GT1-7 cells under these experimental conditions [
28]. Zn
2+-induced GT1-7 cell death may be attenuated by Al
3+, which suppress the elevation of the [Ca
2+]
i level. Therefore, Ca
2+ homeostasis is likely to be involved in Zn
2+ neurotoxicity.
We previously showed that the energy substrates pyruvate and citrate salts inhibit Cu
2+- and Zn
2+-induced GT1-7 cell death [
29]. The coexistence of pyruvate and citrate did not affect the [Ca
2+]
i, intracellular Zn
2+ levels ([Zn
2+]
i), or
MT mRNA levels. Therefore, it is unlikely that pyruvate and citrate attenuated Cu/Zn-induced neurotoxicity by chelating Cu
2+ and Zn
2+. Shelline et al. reported that Zn exposure decreased the nicotinamide adenine dinucleotide (NAD
+) and ATP levels in cultured cortical neurones, and treatment with pyruvate restored the NAD
+ levels [
30,
31]. Pyruvate administration also attenuated post-ischaemic neuronal cell death
in vivo [
32]. Imaging studies using Zn
2+-sensitive fluorescent dyes and mitochondrial markers have revealed that Zn
2+ is localised within mitochondria [
33]. Zn
2+ has been reported to inhibit various mitochondrial enzymes and intracellular trafficking of mitochondria. Taken together, these results suggest that energy deficiency and inhibition of mitochondrial glycolysis are involved in Zn
2+ neurotoxicity [
28].
DNA microarray analysis revealed that Zn
2+ markedly increased the expression of ER stress-related genes, including CCAAT enhancer-binding protein homologous protein (
CHOP) and growth arrest and DNA damage-induced gene 34 (
GADD34) [
34,
35]. The ER acts as an intracellular Ca
2+ reservoir and is involved in the regulation of [Ca
2+]
i. Therefore, an increase in the [Ca
2+]
i is associated with a decrease in the Ca
2+ level in the ER. A decreased Ca
2+ level in the ER is thought to cause ER stress because it leads to the decreased function of molecular chaperones and enzymes that bind Ca
2+ [
36]. The increase in the [Ca
2+]
i in GT1-7 cells induced by Zn
2+ administration appears to be closely related to the upregulation of these ER stress-related factors. ER stress is associated with the accumulation of unfolded and misfolded proteins and is involved in various neurological diseases such as cerebral ischaemia, AD, and prion disease (PD) [
37]. Three signalling proteins (ER stress sensors) —inositol-requiring enzyme 1α (IRE1α), protein kinase R-like ER kinase (PERK), and activating transcription factor (ATF) 6— are activated by sensing ER stress [
38]. IRE1α, PERK, and ATF6 activate diverse signalling pathways. Phosphorylation of the α-subunit of eukaryotic translation initiation factor 2α regulates ATF4 translation via PERK. ATF4 is a transcription factor that promotes
CHOP and
GADD34. In addition to these factors, whose expression was confirmed to be induced by Zn
2+ administration in DNA microarray studies, we examined the expression of other ER stress-related genes such as immunoglobulin binding protein (
Bip), ER degradation-enhancing α-mannosidase-like protein (
EDEM), spliced X-box binding protein-1 (
sXBP1), glucose-regulated protein 94 (
GRP94), and protein disulfide isomerase (
PDI) by Zn
2+ using RT-PCR. RT-PCR confirmed that Zn
2+ administration to GT1-7 cells induced the expression of activity-regulated cytoskeleton (
Arc),
CHOP,
GADD34, and
ATF4. The induction of the expression of metal-related genes
ZnT-1,
MT1, and
MT2 was also confirmed, whereas other ER stress-related genes, including
Bip,
EDEM,
sXBP1,
GRP94, and
PDI, showed no significant changes. Furthermore, dantrolene, an inhibitor of ER stress, attenuated Zn
2+-induced GT1-7 cytotoxicity [
25]. These results strongly suggested that PERK-related pathways are involved in Zn
2+-induced ER stress.
In addition to Zn, trace elements such as iron (Fe), copper (Cu), and manganese (Mn) are distributed at different levels in various parts of the brain and maintain the normal structure and function [
39,
40]. Among these metals, we showed that the presence of Cu
2+ markedly exacerbated Zn
2+-induced GT1-7 cytotoxicity [
25]. Cu
2+ alone did not affect the gene expression levels of
Arc,
CHOP, and
GADD34 but significantly enhanced the induction of these factors by Zn
2+. Furthermore, western blotting showed that the co-administration of Zn
2+ and Cu
2+ significantly increased the amount of CHOP protein. CHOP is involved in the initiation of the apoptotic cascade [
41] and the activation of GADD34, which reportedly increases after traumatic brain injury [
42]. We also found that the antioxidant thioredoxin-conjugated human serum albumin (HSA-Trx) attenuated Cu
2+- and Zn
2+-induced neuronal cell death [
43]. Zn exists only as Zn
2+, whereas Cu is a redox-active metal that exists as oxidised Cu
2+ and reduced Cu
+. Cu
2+ administration induces ROS generation in GT1-7 cells, whereas Zn
2+ alone does not induce ROS generation or affect Cu
2+-induced ROS generation [
14]. It is widely known that oxidative stress is involved in various neurodegenerative diseases, and ROS that cause oxidative stress adversely affects many signalling pathways, such as SAPK/JNK-related and ER-related pathways [
44,
45,
46,
47]. The co-administration of Cu
2+ and Zn
2+ to GT1-7 cells activated SAPK/JNK, phospho-c-Jun, and phospho-ATF2 downstream of the SAPK/JNK pathway. Furthermore, SP600125, an inhibitor of the SAPK/JNK signalling pathway, significantly suppresses Cu
2+- and Zn
2+-induced SAPK/JNK signalling pathway activation and neuronal cell death [
14]. In addition, suppression of Cu
2+ and Zn
2+ cytotoxicity by HSA-Trx inhibits SAPK/JNK signalling pathway activation and ROS production [
43]. Furthermore, the endogenous selenium (Se)-containing amino acid selenomethionine (Se-Met) induces glutathione peroxidase, blocks ROS production, significantly inhibited CHOP induction, and inhibites Cu
2+- and Zn
2+-mediated cytotoxicity [
48]. Cu
2+ triggers ROS production, which may enhance Zn
2+ cytotoxicity by inducing the SAPK pathway and ER stress. These findings indicated that Zn
2+ may be significantly involved in the ER stress pathway.
Figure 3 shows our hypotheses regarding Zn
2+-induced neuronal death (and the protective effect of carnosine, which will be discussed later).
Figure 3.
Hypothetical illustration of the molecular mechanism underlying the protective effect of carnosine in preventing zinc-induced neuronal death. Zn2+ is stored in presynaptic vesicles and is secreted into the synaptic cleft during ischaemia. Excess secreted Zn2+ translocates into the cell and can cause disruption of Ca2+ homeostasis, mitochondrial energy failure, endoplasmic reticulum (ER), and oxidative stress and consequently apoptotic neuronal death. Co-exposure to Zn2+ and Cu2+, which are stored in presynaptic vesicles and secreted during ischaemia, potentiates these effects. These pathways are inhibited by Zn2+ chelators (Ca-EDTA), Ca2+ channel blockers (Al3+ and nimodipine), energy substrates (pyruvate and citrate), SAPK/JNK signalling pathway inhibitor (SP600125), and antioxidants (HSA-Trx). Carnosine inhibits the ER stress-related, Arc-related apoptotic, and ROS pathways. Carnosine is synthesised in glial cells, secreted in response to glutamate and Zn2+ stimulation, and protects neurones from Zn2+ neurotoxicity.
Figure 3.
Hypothetical illustration of the molecular mechanism underlying the protective effect of carnosine in preventing zinc-induced neuronal death. Zn2+ is stored in presynaptic vesicles and is secreted into the synaptic cleft during ischaemia. Excess secreted Zn2+ translocates into the cell and can cause disruption of Ca2+ homeostasis, mitochondrial energy failure, endoplasmic reticulum (ER), and oxidative stress and consequently apoptotic neuronal death. Co-exposure to Zn2+ and Cu2+, which are stored in presynaptic vesicles and secreted during ischaemia, potentiates these effects. These pathways are inhibited by Zn2+ chelators (Ca-EDTA), Ca2+ channel blockers (Al3+ and nimodipine), energy substrates (pyruvate and citrate), SAPK/JNK signalling pathway inhibitor (SP600125), and antioxidants (HSA-Trx). Carnosine inhibits the ER stress-related, Arc-related apoptotic, and ROS pathways. Carnosine is synthesised in glial cells, secreted in response to glutamate and Zn2+ stimulation, and protects neurones from Zn2+ neurotoxicity.
2.3. neurodegenerative diseases and Zn
Based on the aforementioned results, the hypotheses regarding Zn
2+-related neurotoxicity are as follows (
Figure 3). Normally, Zn
2+ and Cu
2+ are released into the synaptic cleft upon neuronal excitation and regulate signal transduction [
20]. Secreted Zn
2+ and Cu
2+ undergo rapid reuptake into presynaptic neurones via the Zn transporter or CTR1, thereby maintaining the level of these ions in the synaptic cleft. However, under conditions such as transient global cerebral ischaemia, prolonged neuronal excitation occurs in major parts of the brain, and Zn
2+ and Cu
2+ are released from synaptic vesicles into the synaptic cleft and translocate to the same neurones in large amounts. Increased [Zn
2+]
i inhibit of mitochondrial energy production mechanisms and increases the [Ca
2+]
i. Impaired cellular protein folding due to energy depletion causes the accumulation of defective proteins in the ER. An increase in the [Ca
2+]
i induces ROS generation. This potentiates ER stress and/or the SAPK/JNK pathways leading to apoptotic neuronal cell death.
There is increasing evidence that Zn
2+-mediated postischaemic neuronal cell death is involved in neurodegeneration after stroke or ischaemia [
39,
49,
50,
51]. VD is a disease associated with such neurodegeneration [
8,
52]. After transient global ischaemia, blockage of blood flow and concomitant oxygen and glucose deprivation induce abnormal neuronal excitation in most parts of the brain, followed by an excessive release of glutamate into synaptic clefts. A continuous influx of large amounts of Ca
2+ is triggered, causing delayed cell death of vulnerable neurones in the hippocampus or cerebral cortex, leading to the development of infarcts and, ultimately cognitive impairment and VD. Epidemiological studies have reported that approximately 30% of patients with stroke develop symptoms of dementia 3 years later [
53]. Under ischaemic conditions, significant amounts of Zn
2+ (approximately 300 μM) have been reported to be released into the synaptic cleft along with glutamate after membrane depolarisation [
54]. Furthermore, there is a report showing that Zn accumulates in apoptotic neurones in the hippocampus after ischaemia [
55]. The administration of calcium ethylenediaminetetraacetate (Ca-EDTA), a membrane-impermeable Zn
2+ chelator, protected hippocampal neurones and reduced infarct volume after transient global ischaemia in experimental animals [
56]. Kitamura et al. revealed an increase in the extracellular Zn
2+ levels in rats with transient middle cerebral artery occlusion using microdialysis [
57]. Additionally, Zn
2+ contributes to increased blood-brain barrier (BBB) permeability following ischaemia [
58].
Zn
2+-mediated neuronal cell death may be involved in the pathogenesis of PD [
3]. Dopaminergic neuronal shedding and microglial activation, which are implicated in the pathogenesis of PD, have been observed in animal models of PD established using 6-hydroxydopamine (6-OHDA) or 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine [
59]. ROS derived from 6-OHDA, which are taken up through dopamine transporters, intraneuronal 6-OHDA autoxidation, extracellular 6-OHDA autoxidation, and microglial activation are the molecular mechanisms responsible for 6-OHDA-induced dopaminergic degeneration [
60]. 6-OHDA is readily oxidised and generates several reactive and cytotoxic products, such as quinones, H
2O
2, 5,6-dihydroxyindole, superoxide anion radicals, hydroxyl radicals, and singlet oxygen [
61]. ROS derived from paraquat, an herbicide that is also taken up by dopamine transporters, lead to glutamate exocytosis via transient receptor potential melastatin 2 cation channel activation in the substantia nigra, followed by nigral dopaminergic degeneration via intracellular Zn
2+ dysregulation [
62,
63]. From the intracellular ROS derived from 6-OHDA and PQ, H
2O
2 readily passes through cell membranes via aquaporin channels [
64]. H
2O
2 elevation in the extracellular compartment excites glutamatergic neurone terminals [
65], and Zn
2+ is released along with glutamate. Additionally, we showed that treatment with 6-OHDA induces integrated stress-related genes, such as
CHOP,
GADD34, and
ATF4 in GT1-7 cells [
66]. Increase of Zn
2+ release by 6-OHDA may be related to induction of these factors.
Zn
2+ entry and an increased [Zn
2+]
i in other words ‘Zn translocation’ are key events in Zn
2+-induced neurotoxicity. There are three major pathways for Zn
2+ entry: voltage-gated Ca
2+ channels, N-methyl-D-aspartic acid (NMDA)-type glutamate receptors, and AMPA/kainate-type glutamate receptors (A/K-R) [
28]. Under normal conditions, most hippocampal neurones express AMPA receptors containing GluR2 subunits that are poorly permeable to Ca
2+ and Zn
2+. However, after ischaemia, there is a sharp decrease in GluR2 expression, and neurones express Ca
2+-permeable AMPA receptors (Ca-A/KR). As the permeability of Zn
2+ and Ca
2+ through the Ca-A/KR channels is greater than that through the NMDA receptor channels, the increased expression of Ca-A/KR channels enhances Ca
2+ and Zn
2+ toxicity. Zn
2+ is also thought to be involved in the transcriptional regulation of Ca-A/KR channels, as Ca-EDTA attenuates ischaemia-induced downregulation of the
GluR2 gene [
56]. These Zn
2+-mediated neuronal cell death events can be explained by the Zn
2+-mediated neurotoxicity hypothesis. Considering the involvement of Zn
2+ in transient global ischaemia, substances that prevent Zn
2+-induced neuronal cell death may be potential candidates for the prevention or treatment of postischaemic neurodegeneration and, ultimately, the treatment of VD and PD. We used extracts of various agricultural products, such as vegetables, fruits and fish, and found that extracts of the Japanese eel (
Anguilla japonica), mango fruit (
Mangifera indica L.), and round herring (
Etrumeus teres) protected GT1-7 cells from Zn
2+-induced neurotoxicity. The active fractions were separated from these extracts using high-performance liquid chromatography (HPLC) and the structures of the components were analysed by LC mass spectrometry. The active compounds included carnosine, citric acid, and histidine [
16,
34,
67].