NEUROTOXICITY OF METAL MIXTURES

Abstract

Environmental exposures and/or alterations in the homeostasis of essential transition metals (ETM), such as Fe, Cu, Zn or Mn, are known to contribute to neurodegenerative diseases (ND), such as Alzheimer’s Disease (AD) and Parkinson’s Disease (PD). Aberrant ETM homeostasis leads to altered distributions, as significant amounts may accumulate in specific brain areas, while causing metal deficiency in others. The disruption of processes reliant on the interplay between these ETM, may lead to loss of metal balance and the ensuing neurotoxicity via shared mechanisms, such as the induction of oxidative stress (OS). Both ETM imbalance and OS may play a role, via complex positive loop processes, in primary neuropathological signatures of AD, such as the accumulation of amyloid plaques and neurofibrillary tangles (NTF), and in PD, α-Syn aggregation and loss of dopamine(DA)rgic neurons. The association between ETM imbalance and ND is rarely approached under the view that metals such as Fe, Cu, Zn and Mn, can act as dangerous endogenous neurotoxic mixtures when their control mechanisms became disrupted. In fact, their presence as mixtures implies intricacies, which should be kept in mind when developing therapies for complex disorders of metal dyshomeostasis, which commonly occur in ND.

1. INTRODUCTION

Most elements in the periodic table are metals, sharing their propensity to lose electrons and react with molecular oxygen to form oxides (Garza-Lombó et al., 2018). Among them, heavy metals are found naturally in the earth crust and have been exploited from many years for various industrial and economic purposes (Engwa et al., 2019). In their normal state, heavy metals have a specific gravity of more than about 5 g cm⁻³, i.e. their densities are five times greater than water (Garza-Lombó et al., 2018). Unfortunately, since various industries produce and discharge wastes with different heavy metals into the environment, excessive exposure is as common source of toxicity in multiple populations around the world (Mahmood et al., 2012; Wright and Baccarelli, 2007). In biological systems, metals can be broadly divided into: xenobiotic heavy metals such as mercury (Hg), lead (Pb), and cadmium (Cd); alkali and alkaline-earth metals, such as sodium (Na), potassium (K), magnesium (Mg), and calcium (Ca); and essential transition metals (ETM), such as iron (Fe), copper (Cu), zinc (Zn) and manganese (Mn) (Garza-Lombó et al., 2018). While, toxic heavy metals do not have any function in the body and in minute quantities can have deleterious effects, some ETM (and some alkali and alkaline-earth metals) have functional roles in the organism, which are essential for various diverse physiological and biochemical activities (Engwa et al., 2019; Garza-Lombó et al., 2018). The biological effects of metals are related to their chemical properties. Toxic heavy metals can induce oxidative toxicity, and bind to proteins changing their function, which can occur either by forming a complex with functional side-chain groups or by displacing ETM in metalloproteins; they can also interfere with other metals transport, including the transport of ETM (Garza-Lombó et al., 2018; Tamás et al., 2104). In this manner, toxic heavy metals exposures can influence ETM homeostasis in the human body (Garza-Lombó et al., 2018; Karri et al., 2016). By its turn, ETM in biological systems are particularly adept at catalyzing important redox reactions, when present in proper levels (Garza-Lombó et al., 2018). However, though the presence of ETM in the body is vital, there is a very thin line between their beneficial and toxic effect on cellular processes, since in fact, even ETM are toxic when surplus (Wojtunik-Kulesza et al., 2019). Hence, environmental exposures to ETM may also result in their accumulation in the body, leading to detrimental changes with adverse health consequences (Adedara et al., 2015).

Among several injurious effects potentially induced by metals, neurotoxicity is a common health endpoint for excessive metal levels in the organism (Mitra et al., 2014, Wright and Baccarelli, 2007). Toxic heavy metals are broadly known to induce neurotoxicity (Nava-Ruíz and Méndez-Armenta, 2011), and even ETM may play a sinister part in the brain. Truly, Fe, Cu, Zn or Mn dyshomeostasis results in their abnormal distribution, leading to excessive/deficient amounts, which is condition that may elicit oxidative stress (OS) and molecular damage impeding brain cellular functions (Bonda et al., 2011). Today, it is well recognized that there is an association between increased environmental/ occupational exposure to toxic heavy metals or alterations in the homeostasis of ETM, and several neurological disorders (Bonda et al., 2011; Garza-Lombó et al., 2018). These disorders include neurodegenerative diseases (ND), such as Alzheimer’s Disease (AD) and Parkinson’s Disease (PD) (Mitra et al., 2014; Trist et al., 2018; Weekley and He, 2017), to name a few. Accordingly, and with respect to environmental exposures, high incidences of AD and PD have been observed in employees in the automobile and paint industries as well as in other metal-utilizing factories (Mitra et al., 2014), where there is a widespread use of toxic heavy metals and ETM (Risopatron, 2014). Regarding ETM, other examples pertaining to environmental exposures include ingestion of excessive Fe through drinking water or food, which might contribute to the evolution and progression of AD (Li et al., 2019a); and the accumulated evidence that chronic exposure to Cu is implicated in PD (Lan et al., 2016). Additionally, it is established that the biggest risk factor for ND is age, which directly correlates with metal dyshomeostasis. In essence, biometal imbalance is more causative of AD pathogenesis than toxic metal exposure (Sharma et al., 2018). This, when considered together, there is a clear recognition that abnormal accumulation and distribution of ETM is involved in the pathogenesis and progression of several brain disorders, such as AD and PD (Mitra et al., 2014; Trist et al., 2018; Weekley and He, 2017). This issue prompted widespread interest by multiple research groups (Bonda et al., 2011), and while many of these diseases torment the affected persons, effective treatments are still absent.

Most relevant is the fact that real-life scenarios include exposures to multiple metals simultaneously, with biological systems exposed to a complex environmental/ endogenous system where chemical species could be considered a “chemical cocktail” (Rodríguez-Moro et al., 2020; Wright and Baccarelli, 2007). Metal mixture toxic interactions can be dose additive, interactive (synergistic or antagonistic) or independent of each other, generating high level biochemical changes in sensitive targets, such as the brain. In this sense, metal mixtures can pose a bigger risk for cognitive, motor or behavior dysfunction than individual metals (Karri et al., 2016, Zhou et al., 2019). As components of mixtures, metals can have competitive interactions with macromolecule and transporters due to their functional similarities. For instance, some xenobiotic metals and ETM have affinity for the same metal ion transporters, such as the divalent metal transporter (DMT)-1, leading to toxic metal interactions at the blood-brain barrier (BBB) and brain tissue. Furthermore, when the toxicity of the components of these mixtures is exerted via mutual/simultaneous mechanisms, there is an increased risk of neurotoxicity. The induction of OS is an example, constituting a most relevant convergent point on the mechanisms of metal toxicity (Whittaker et al, 2010), and representing a pathway that leads to the destruction of cells, including neurons and vascular cells in the brain (Chong et al, 2005); OS is a common event for ND, such as AD and PD (Pimentel et al., 2012).

In a previous volume of “Advances in Neurobiology”, neurotoxicity induced by mixtures of Pb, Hg, Cd, Arsenic (As), Fe, Cu, Zn and/or Mn was illustrated in the perspective of environmental exposures; Pb’s toxicity and its interaction with other metals constituted the major core. This chapter is focused in the ETM Fe, Cu, Zn and Mn from the view of endogenous neurotoxic mixtures, and it is intended to explore its interactions in vivo, the shared toxic mechanisms and subsequent damage, stemming from their aberrant homeostasis; the contribution of these events to the most relevant AD and PD hallmarks will be depicted.

2. ESSENTIAL TRANSITION METALS IN THE BRAIN

The brain contains some of the highest concentrations of Fe, Cu, Zn and Mn in the human body (Genoud et al., 2017). These ETM are essential for numerous cellular functions and their high reactivity is essential for life; however, as cofactors for many enzymes that catalyze redox reactions, they can also be involved in uncontrolled redox reactions associated with OS and cellular damage. Hence, a highly conserved network of proteins/transporters strictly regulates the homeostasis of these metals, controlling their uptake, intracellular distribution, storage, and export; these processes assure that the metals are stored, and released only in response to metabolic needs (Garza-Lombó et al., 2018; Mitra et al., 2014; Pokusa and Trančíková, 2017).

2.1. Iron distribution, functions, and homeostasis

The metal Fe is heterogeneously distributed in the brain, being highly concentrated in the substantia nigra (SN), hippocampus, striatum, interpeduncular nuclei areas, and in myelin (Garza-Lombó et al., 2018). Fe is essential for the synthesis and metabolism of neurotransmitters including dopamine (DA), norepinephrine, epinephrine and serotonin, Each one of these monoamine neurotransmitters are synthesized by Fe-dependent enzymes, which include phenylalanine hydroxylase, tyrosine hydroxylase and tryptophan hydroxylase (Lan et al, 2016, Zucca et al., 2017). Fe have also a role in other steps of neurotransmitter metabolism, involving uptake, extracellular concentration, interaction with receptors and catabolism (Zucca et al., 2017). It is also a required cofactor for the synthesis of lipid components of myelin, with myelination requiring adequate levels of Fe; this metal is also a cofactor for a large number of metalloproteins involved in metabolism and signal transduction (Li et al., 2017; Zucca et al., 2017). DNA synthesis and repair, oxygen activation, mitochondrial electron transport and metabolism are other processes for which Fe is required (Lan et al, 2016; Pokusa and Trančíková, 2017).

Fe homeostasis is regulated through the communication between the BBB and astrocytes. The uptake into the BBB can occur through two pathways: (a) direct transport of Fe²⁺ into the cytosol via DMT1; or (b) endocytosis of transferrin (Tf)-bound Fe³⁺ via a Tf receptor, where the low pH of the endosome causes the release of Fe³⁺ from Tf (Garza-Lombó et al., 2018). Since excess Fe is toxic and Fe²⁺ reacts with oxygen to produce ROS, living organisms have developed mechanisms to sequester Fe in a non-toxic form until it is required for metabolic needs. In this way, the Fe storage protein, ferritin (Ft), plays a crucial role in protecting cells from Fe catalyzed ROS formation (Orihuela et al., 2011). The efflux of Fe from endothelial cells to the interstitial space occurs via the coordinated activity of the Fe²⁺ transporter ferroportin (Fpn) and the Cu-dependent ferroxidases, hephaestin (Hp) or soluble ceruloplasmin (Cp). By its turn, astrocytes regulate the release of Fe from the BBB by either secretion of Cp, which stimulates Fe release, or inversely by production of hepcidin decreasing Fe efflux. Although in neurons the mechanisms involved in Fe uptake are still unclear, it is known that DMT1 is prominent in human neurons, with higher expression in the basal ganglia, including the SN, globus palidus (GB), hypothalamic nucleus and striatum (Chen et al., 2015, Garza-Lombó et al., 2018; Skjørringe et al., 2015). Truly, DMT1 is involved in the transport of several metals including Fe, Cu, Zn and Mn (Espinoza et al., 2011); in the case of Fe, DMT1 is the major transporter contributing for non-heme Fe uptake in most types of cells. DMT1 expression is regulated by Fe through translational and degradation pathways, ensuring Fe homeostasis. Additionally, a recent report demonstrated also that ZIP8, a member of the ZIP family of metal transporters, is the primary transporter involved in non-Tf-bound Fe into neurons (Garza-Lombó et al., 2018).

2.2. Copper distribution, functions, and homeostasis

The third-most abundant ETM in the brain is Cu, with uneven distribution in their different regions and variations reported in different age groups. Labile brain Cu stores have been detected in the soma of cerebellar granular and cortical pyramidal neurons, as well as in neuropil within the cerebral cortex, cerebellum, hippocampus and spinal cord. As an ETM, owing to its redox activity, Cu serves as a cofactor for key metabolic enzymes that mediate various cellular processes, including mitochondrial energy generation (cytochrome c oxidase), free radical detoxification (Cu/Zn superoxide dismutase) and as mentioned, Fe homeostasis (Cp) (Pal and Prasad, 2016). Cu is also required as a cofactor of several other brain enzymes, such as peptidylglycine monooxygenase and tyrosinase (Garza-Lombó et al., 2018). In the same way as described for Fe, Cu is involved oxygen activation for reduction, and in neurotransmitter biosynthesis and metabolism, with Cu serving as a co factor of DA β-hydroxylase and DA b-monooxygenase (Garza-Lombó et al., 2018; Pal and Prasad, 2016).

The control of Cu homeostasis in the brain requires a close interrelationship between the BBB, neurons, and astrocytes, with astrocytes regulating the properties of the BBB, which is the entry point for Cu into the brain from the blood stream. The transport of extracellular Cu into brain cells is primarily via the Cu transport protein CTR1 (Garza-Lombó et al., 2018), although an in vitro study using Caco-2 cells showed that hCTR1 is also able to transport Fe and Zn (Espinoza et al., 2011). Other molecules have also been proposed to mediate Cu uptake, such as the Zn-regulated transporter (Zrt) or the Irt (Fe-regulated transporter)-like protein 4 (ZIP4) (Garza-Lombó et al., 2018). As already noted, the uptake of Cu by the brain barriers may be also mediated by DMT1, and (similarly to Fe) Cu levels regulate the translational expression of this transporter (Zheng and Monnot, 2012). It is reported that Cu and Fe can competitively inhibit each other’s transport by DMT1 (Garza-Lombó et al., 2018); yet, DMT1 seems to play a compensatory role for Cu uptake under certain conditions, such as the absence of CTR1 or under low Fe conditions (Garza-Lombó et al., 2018).

Metallothioneins (MT) play as well, a relevant role in the regulation of Cu levels. The main function of this molecule is to transport, store and regulate several cellular ETM, which include Fe, Zn and Mn (Juárez-Rebollar et al., 2017; Xu et al., 2019). There are four types MT in mammalian cells: MT-1 and MT-2, both widely expressed in the body, and in brain mainly expressed in glia cells and astrocytes in the cerebellar cortex; MT-3, which is present primarily in the brain; and MT-4, expressed in some epithelial cells (Okita et al., 2017; Xu et al., 2019). MT-3 has an important role regarding brain Cu (as well as Zn) homoeostasis, being particularly abundant in Zn-enriched neurons (Xu et al., 2019). Under physiological conditions, surplus Cu is chelated in the brain inside the cells by MT, thus controlling its reactivity, and then is exported when reaching a certain concentration threshold (Cruces-Sande et al., 2019).

2.3. Zinc distribution, functions, and homeostasis

Zn is the second most abundant biometal in the body, after Fe (Šulinskienė et al., 2019). In brain, is most concentrated in the limbic system, i.e. the hippocampus and amygdala, and in Zn-containing glutaminergic neuron-rich areas. In these neurons, Zn may serve as an endogenous neuromodulator in synaptic neurotransmission, participating in the storage, release, and uptake of glutamate, and in the modulation of glutamate receptors (Kordas and Stoltzfus, 2004). A large portion of Zn also serves the function of Zn metalloproteins in neurons and glial cells, where it is involved as a cofactor in important functions, such as gene expression or enzymatic reactions and DNA synthesis (Colvin et al., 2000; Garza-Lombó et al., 2018; Takeda, 2000; Xu et al., 2019).

The metal is supplied to the brain via both the BBB and blood–cerebrospinal fluid barrier.

There are two families of Zn transporters: the ZnT family, which act to decrease intracellular Zn concentrations by exporting Zn from the cytoplasm to the lumen of organelles or the extracellular space, and the ZIP family, which import the metal from the extracellular space or organellar lumen into the cytoplasm). In neurons, Zn uptake is mediated by, voltage-gated Ca²⁺ channels, ZIP1 and 3 transporters, and the Zn²⁺ permeable Fe channel a-amino-3-hydroxy-5-methyl-4-isoxazolepropionate receptor (AMPAR). ZnT1 is expressed in astrocytes, microglia, and oligodendrocytes, and their expression levels are directly modulated by Zn (Garza-Lombó et al., 2018; Xu et al., 2019). Although DMT1 act as a common entry pathway for several divalent metals, Zn may not a major substrate for DMT1. Even so, in vitro studies suggest that Zn can regulate DMT1 function and expression in Caco-2 cells (Yamaji et al., 2001). Concerning MT, endogenous Zn equilibrium is also promoted by this molecule (akin to Cu), through Zn chelation (Garza-Lombó et al., 2018). Again, when Zn present is in excess, MT may function in its sequestration, thereby protecting cells from Zn toxicity (Xu et al., 2019).

2.4. Manganese distribution, function and homeostasis

The concentrations of Mn in the human brain are the highest in putamen, caudate nucleus, and globus pallidus (GB), and were found to positively correlate with the age. This ETM is essential for brain physiology, given its role as a cofactor for numerous enzymes; these include arginase, glutamine synthetase, the antioxidant Mn-superoxide dismutase (SOD), pyruvate carboxylase, and protein serine/threonine phosphatase-1 (Horning et al., 2015). There are noteworthy interactions between Mn and Fe in several brain physiological processes, due to their structural and chemical similarities. Both metals are cofactors for several common metalloenzymes, which play critical roles in antioxidant defense and neurochemistry, with one metal compensating for the other, in case of deficiency to correct impaired enzyme function. Such interplay occurs for instance in neurotransmission systems, the DAergic pathway as an example; both metals support the function of tyrosine hydroxylase, which is the rate limiting enzyme for DA synthesis. (Ye et al., 2017).

Mn levels in the brain are strictly regulated, with Mn entering in cells through multiple transporters, including DMT1, the transporters ZIP8 and TfR (common to Fe) and ZIP14 (also proposed for Cu), the citrate transporter, the choline transporter, the DAT transporter, and calcium channels. Among them, DMT1 is the primary transporter for divalent Mn, while TfR is the primary transporter for trivalent Mn. Complex relations exist among brain Mn, Fe, DMT1 and Trf, and some still demand further clarification. Pertaining to DMT1, some studies reveal that its transport affinities for ETM are as follows: Mn>Fe>Zn. DMT1 preferentially transports Mn into the brain through the BBB, especially under low Fe conditions [due to iron-responsive element (IRE)-mediated regulation], which increases DMT1 expression. DMT1 transport of Mn from endosomes into cytosol occurs in a TfR dependent manner (Chen et al., 2015). Hence, Mn transport is mediated in part by Fe transporters, and since their expression is affected by several conditions, such as Fe deficient anemia and Fe overload hemochromatosis, altered body Fe status also modifies Mn transport (Ye et al., 2017).

3. DYSHOMEOSTASIS OF ESSENTIAL TRANSITION METALS IN THE BRAIN

Non-transition metal ions, such as Na⁺, K⁺, and Ca²⁺, are found at high concentrations in the brain (Garza-Lombó et al., 2018). In the contrary, ETM are present in trace amounts, with the imbalance between healthy and toxic levels having the potential of being quite hazardous (Wojtunik-Kulesza et al., 2019). Abnormal metal sequestration that leads to overload and toxicity due its accumulation in specific brain regions, is a common feature of neuronal pathologies (Bowmen et al., 2001). Mechanistically metals imbalance can contribute to mitochondrial dysfunctions, autophagy dysregulation, endoplasmic reticulum stress, and the activation of apoptosis (Wojtunik-Kulesza et al., 2019), among several other injurious outcomes. Among these mechanisms, OS is an important detrimental event causing excessive production of ROS, and is common in response to surplus EMT, when involved in uncontrolled redox reactions. Effectively, ETM such as Fe, Cu, Zn and Mn, can play an active role in ROS production, may resulting in dangerous oxidative reactions that can lead to cellular damage (Balachandran et al., 2020; Bisaglia and Bubacco, 2020; Garza-Lombó et al., 2018; McCord and Aizenman, 2014; Pokusa and Trančíková, 2017; Wojtunik-Kulesza et al., 2019). Furthermore, the brain is particularly prone to oxidative damage, since neurons do not possess high antioxidant capacity or special antioxidant systems (Pokusa and Trančíková, 2017); all such changes lead to neurotransmission and cell disturbances (Pokusa and Trančíková, 2017; Wojtunik-Kulesza et al., 2019). It is in this manner that aberrant levels of ETM in the brain can constitute a toxic endogenous metal mixture, which can exert toxicity via shared mechanisms, such as the induction of OS, in concurrent areas of the organ (Bisaglia and Bubacco, 2020; Garza-Lombó et al., 2018; Martins Jr. et al., 2019; McCord and Aizenman, 2014).

3.1. Neurotoxicity of iron via oxidative stress

Although Fe is mainly stored in the cytosol, mitochondria are the organelles with the highest Fe demand, as they require Fe-S clusters and heme groups for electron transfer during respiration; mitochondria are also considered the main sources of ROS under physiological conditions (McLeary et al., 2019). However, under pathological circumstances Fe accumulation may occur resulting in excessive ROS production through several reactions. In fact, ETM have oxygen transferring properties for the catalytic power to generate the most powerful oxidant for OS, which is the hydroxyl radical (•OH). When Fe in involved, the radical is generated via the Fenton reaction as follows: Fe²⁺ + H₂O₂ → Fe³⁺ + •OH + OH⁻ (equation 1) (Bisaglia and Bubacco, 2020; Kanti Das et al., 2015; Zhao, 2019). Subsequently other ROS such as superoxide (O₂ •⁻) and more •OH can be produced according to following the equations: •OH +H₂O₂ → O₂• ⁻+ H+ +H₂O (equation 2) and O2•⁻ +H₂O₂ → •OH +OH⁻ +O₂ (equation 3). The part of the reaction represented by equation 3 is known as the “Haber-Weiss Reaction” (Kanti Das et al., 2015). Additionally, Fenton-like reactions may occur leading to the production of: i) •OH in aqueous environments, from reactions of Fe²⁺ complexes with a ligand (L) or other ETM complexes (Mⁿ-L; M: other transition metals, e.g., Cu; n: oxidation state) with H₂O₂, according to: Fe²⁺-L+ H₂O₂ → Fe³⁺ - L + •OH+ OH⁻ (equation 4) and Mⁿ-L + H₂O₂ → Mⁿ⁺¹-L + •OH + OH⁻ (equation 5), respectively; ii) lipid alkoxy radicals in lipid environment, from reactions of Fe²⁺ (as well as other ETM) with lipid peroxide (Zhao, 2019). The importance of Fe dysregulation in neurodegenerative processes is underlined by recent studies. These studies associate Fe accumulation to a novel type of cell death pathway, distinct from pathways such as apoptosis, which is named ferroptosis (Ashraf et al., 2020; McLeary et al., 2019). Ferroptotic cell death is precisely a necrotic-like cell death, characterized by Fe-dependent lipid peroxidation due to either the formation of •OH and H₂O₂ (Garza-Lombó et al., 2018). Additionally, there is a still unrecognized, but potentially harmful process by which MT might mediate Fe²⁺ release from Ft, capable of causing metal toxicity in the cell. In this pathways Zn–MT complexes could act as electron donors to reduce the Fe³⁺ mineral core of Ft and facilitate Fe release. Concurrently, Ft interaction with the Zn-complexes of mammalian MT1, MT2 and MT3 would lead to simultaneous Fe²⁺ and Zn²⁺ release. As emphasized by the authors, this process could be especially perilous in the brain, contributing to neurological disorders (Orihuela et al., 2011).

3.2. Neurotoxicity of copper via oxidative stress

Similar to Fe, Cu’s ability to cycle between its oxidized state (Cu²⁺) and its reduced state (Cu⁺) allows certain Cu-containing proteins to act as electron carriers or redox catalysts; again, this implies its involvement in uncontrolled Fenton chemistry reactions, which are in this case Cu-elicited reactions [Cu⁺ + H₂O₂ → Cu²⁺ + •OH + OH⁻ (equation 4)] and Haber–Weiss reactions (Bisaglia and Bubacco, 2020; Cruces-Sande et al., 2019; Pal and Prasad, 2016). Accordingly, experimental increased brain Cu levels is accompanied by augmented brain OS state (Cruces-Sande et al., 2019).

3.3. Neurotoxicity of zinc via oxidative stress

The vast majority of intracellular Zn is normally rendered immobile through buffering by brain cytosolic metal-binding proteins (80–90%) and sequestration into organelles. However, when neurons are damaged, as occurs during OS, bound intracellular Zn can be released into the cytosol, triggering a number of detrimental signaling processes including those that lead to further ROS production, marking the start of a positive feedback loop involving intracellular Zn release and ROS generation. Befittingly, it has also been described that the release of Zn from MT-3 by oxidants causes a substantial increase in intracellular Zn concentration. In addition, mitochondria which are the primary producers of ROS in neurons, may take up excessive cytoplasmic free Zn which can inhibit the electron transport chain by interfering with the activity of complex III, and further lead to increased ROS generation (Li et al., 2017; McCord and Aizenman, 2014). Apart from these mechanisms, Zn can also be involved in injurious oxidant generation from several extra-mitochondrial sources. One of the better-studied Zn-mediated apoptosis cascades involves exogenous ROS-triggered Zn liberation and subsequent generation of endogenous oxidative intermediates, such as O₂• ⁻. Hence, many of the components involved in injurious mitochondrial and extra-mitochondrial ROS production, as well as the down- stream processes triggered by ROS, seem to share a common association with intracellular Zn release (McCord and Aizenman, 2014).

3.4. Neurotoxicity of manganese via oxidative stress

In the same way as illustrated for all the mentioned ETM, OS is a relevant mechanism for Mn-induced neurotoxicity (Balachandran et al., 2020; Chtourou et al., 2011). A significant mechanism for Mn-induced OS is via the oxidation of DA and other catecholamines and accordingly, Mn accumulates in DA-rich regions, especially in the basal ganglia. The sequestration of excessive Mn in mitochondria interferes with proper respiration via damage to the electron transport chain, thereby leading to disproportionate production of ROS. More precisely, when proteins or quinones that participate in transfer of electrons are damaged, the chain begins to donate electrons directly to molecular oxygen, by this means creating the highly reactive O₂• ⁻ (Erikson et al., 2004). In vitro studies demonstrated that Mn can also induce increased mitochondrial H₂O₂ production in a dose-dependent manner. Furthermore, Mn in combination with other metals, either synergistically or antagonistically, has been demonstrated to induce OS, cellular damage and apoptosis. However, it is the opinion of some authors that since other metals are present and can impact the transport and protein binding of Mn, the sites and mechanisms of Mn toxicity cannot be fully understood without assessment of other metals. The multiple types of interactions that Mn has with other metals and metal-dependent systems associated with cell death signaling, still warrants further study (Smith et al., 2017). Nevertheless, it is suggested that Mn-induced OS in ND might be secondary to excessive Fe accumulation, since Mn can block protein translation of APP, which is responsible for the stabilization of the membrane-bound Fe²⁺-exporter Fpn, and Ft. Under normal conditions these proteins sequester Fe²⁺, which is a major intrinsic generator of ROS, via conversion to redox inactive Fe³⁺. Thus, Mn-induced blockage of the protein translation of APP may increase Fe²⁺ levels, resulting in the decomposition of H₂O₂, and in turn, in the producing of •OH by the Fenton/Haber-Weiss reactions (Martins Jr. et al., 2019).

3.5. Unbalanced essential transition metals mixtures and oxidative stress in neurodegenerative diseases

Currently, it is commonly accepted that OS represents a mutual hallmark of various ND (Pimentel et al., 2012), and that redox-active ETM dyshomeostasis, is associated with their etiology and pathogenesis (Braidy et al., 2017). The disruption of ETM homeostasis may cause two major features associated with ND: dysfunction of metalloproteins and aberrant metal-protein interactions, that can lead to protein aggregation and uncontrolled ROS production (Bowmen et al., 2011; Garza-Lombó et al., 2018; Pokusa and Trančíková, 2017). For example, while in AD and PD cases with no evident dietary metal exposure the overall brain metal burden of Fe and Cu was found unaltered, instead there was a charge-dependent redistribution of specific metals in the affected brain regions (Mitra et al., 2014). Altered Zn levels have been also determined in specific regions of AD brains, and also in the SN in PD (Garza-Lombó et al., 2018); and increased Mn may also occur in the cortex of AD patients and in PD brains (Kwakye et al., 2015; Li et al., 2017).

Foremost and less exploited, the impact of an increase or decrease of an individual ETM, due to dyshomeostasis, is not restricted to that metal alone. Rather, it causes a more dramatic overall homeostatic imbalance of several metals presumably due to loss of their regulation across cell membranes (Mitra et al., 2014). Noteworthy examples are the association between elevated brain Mn levels and Fe homeostasis impairment, and subsequent alterations in the homeostatic conditions of other divalent metals, that share similar transporter systems with Mn (Bowmen et al., 2011). Further, the concurrent accumulation of ETM in the same brain areas and the succeeding increase in ROS production, higher than the one which could be induced by a single metal, is of relevance for the progression of ND.

4. UNBALANCED ESSENTIAL TRANSITION METALS IN ALZHEIMER DISEASE

Today, AD is the most prevalent ND affecting 10% of people aged 65 and 50% of people aged 80. In 2016, it was estimated that there were about 47 million AD patients in the world and this number is expected to increase to more than 130 million by 2050 (Bagheri et al., 2018). AD signs include memory loss, paranoia, loss of reasoning powers and confusion, with progressive dementia (Bagheri et al., 2018). These signs are associated with the degeneration of hippocampal and cortical neurons (Garza-Lombó et al., 2018), with the hippocampus representing the first and most severely injured brain areas in AD; this area is associated with neurogenesis and long-term memory storage, being the cortex associated with functions such as argumentation, feeling, and language. AD is recognized only after the manifestation of cognitive signs, which may be too late for effective treatment. The disease is ultimately lethal, due to the developing damage of neuronal tissues in the brain (Bagheri et al., 2018).

The pathology of this ND involves a wide variety of interrelated neurotoxic pathways such as, abnormal protein aggregation, OS, mitochondria dysfunction, reduced synthesis of neurotransmitters, and inflammation (Kim et al., 2018; Sharma et al., 2018). Thus, various descriptive hypotheses exists regarding the causes of AD, comprising the cholinergic, mitochondrial cascade, calcium homeostasis, neurovascular, inflammatory and lymphatic system hypothesis; AN additional hypothesis includes the metal ion, amyloid, and tau propagation hypothesis, which will be further discussed below. Despite the ultimate etiology of AD remains obscure (Liu et al., 2015), recent researches shed light on the possible effects of environmental factors contributing to AD progression, evincing Fe/Cu/Zn/Mn unbalances as a main factor of AD pathogenesis (Bagheri et al., 2018; Lavado et al., 2019; Martins Jr. et al., 2019). Observations on aberrant distribution of ETM in AD brains are very common in the literature, and include clinical studies describing abnormally elevated concentrations of ETM in concurrent specific regions of post-mortem brain tissues of AD patients (Bonda et al., 2011; Kim et al., 2018; Martins Jr. et al., 2019). The unbalance of ETM is a relevant source for excessive ROS, and it also is known that the disease is marked by OS events, such as elevations in oxidatively damaged RNA, DNA, proteins, and phospholipids. These damages temporally precede the appearance of the primary neuropathological signatures and diagnostic criteria for AD: the accumulation in affected brain regions of amyloid plaques, composed primarily of aggregated β-amyloid peptides (Aβ), and neurofibrillary tangles (NFT), which are composed of microtubule-associated tau proteins, (Bonda et al., 2011; Gouras et al., 2015; Kim et al., 2018).

4.1. Distribution of brain Fe, Cu, Zn and Mn in Alzheimer disease

Progressive Fe deposits occur in the normal aging process of the brain, particularly in the SN, GB, caudate nucleus, and cortex, and that these brain regions are closely related to ND. Compared with healthy people of the same age, the Fe deposition in patients with AD is more serious in these areas (Smith et al., 1997). More recent data show also that changes in Fe levels over time in the temporal lobe are associated with cognitive decline in individuals with AD (Damulina et al., 2020). However, distinct subtypes of AD, can show different patterns of Fe accumulation (Bulk et al.,2018).

Several studies have shown that Cu is highly abundant in AD lesions; yet, the metal is present at lower than the limits of detection, in the whole brain (Braidy et al., 2017). The rational explanation for this heterogeneity is that a significant quantity of Cu precipitates with senile plaques in AD-affected regions, leading to Cu deficiency in other regions (Braidy et al., 2017). Remarkably, while accumulated Cu triggers several detrimental events, Cu deficiency in the brain can also have adverse effects on the development and maintenance of myelin, it may be inducing degeneration of the nervous system (Li et al., 2017). Another interesting report is that the ratio of Cu/Fe or Cu levels, were already proposed as biomarkers to distinguish between progressive and cognitively stable mild cognitive impairment (MCI) patients. Even so, while Cu/Fe dysregulation may represent a useful biomarker for AD, is still inconclusive (Braidy et al., 2017).

Brain Zn accumulation is a prominent feature of advanced AD and is biochemically linked to dementia severity (Religa et al., 2006). Elevated levels of this ETM have been observed in the cortical gray matter of the temporal lobe, senile plaques, and synaptic vesicles of AD patients (Nutall and Oteiza,2014). Previous studies have also shown significantly decreased levels of the transporter ZnT-1 in the brain of subjects with MCI but significantly increased ZnT-1 in late stage of AD (Dong et al., 2008). As already illustrated for Cu, though excess Zn is proven to mediate AD pathology, decreased Zn at either the systemic or cellular level may also contribute to this ND. Indeed, Zn accumulation in Aβ, may lead to a functional Zn deficiency, despite a net increase in the level of Zn in cortical gray matter can be observed. Zn sequestration in senile plaques could reduce the pool of readily releasable synaptic Zn, contributing to synaptic dysfunction. Zn deficiency also facilitates Ca influx leading to activation of NADPH oxidase and nitric oxide synthase. The activation of these enzymes combined with mitochondrial dysfunction leads to OS. Additionally, Zn deficiency can disrupt energy metabolism and contribute to chronic inflammation (Nutall and Oteiza, 2014); loss of Zn bioavailability in synaptic clefts, contributing to changes in synaptic plasticity and AD cognitive decline (Kanti Das et al., 2015; Li et al., 2017).

With respect to Mn, several studies have shown links between brain Mn concentration and the development of AD or AD-like symptoms, although the data are mixed as to the nature of the relationship (Martins Jr. et al., 2019). Altered serum Mn levels between AD or MCI and control populations has been observed in multiple studies, suggesting a potentially contributory role for Mn in disease development. Additionally, increased levels of plasma Aβ, which according to the authors was presumed to indicate greater Aβ burden in brain, were already associated with increased concentrations of Mn in brain (Martins Jr et al., 2019). Other works confirmed these data, with Mn detected at significantly higher levels in the brain of AD patients with dementia, especially in the cortex, thus suggesting that Mn overload may be involved in the pathology of AD (Li et al., 2017).

4.2. Aberrant Fe/Cu/Zn/Mn in amyloid plaques

Amyloid plaques are a distinct neuropathological hallmark of AD, and are mainly composed of Aβ, which is a normal peptide generated throughout life. In fact, Aβ is formed through proteolysis of the amyloid precursor protein (APP), is not inherently toxic and might even have a physiological function (Gouras et al., 2015; Kim et al., 2018; Sharma et al., 2018). Most of APP cleavage occurs via α-secretase leading to the production of soluble APP which possesses neuroprotective features. In contrast, amyloidogenic cleavage occurs when APP is cleaved by β- and ϒ-secretases generating Aβ monomers. These monomers aggregate to generate oligomers and fibrils, which are neurotoxic. Ultimately, continued aggregation of Aβ peptides into β-sheets generates the hallmark senile amyloid plaques composed of a multitude of highly aggregated Aβ fibrils, and represents an abnormal pathological lesion (Gouras et al., 2015; Martin Jr. et al., 2019). Several works have been demonstrating that Fe, Cu, Zn and Mn are simultaneously enriched in amyloid plaques of AD patients’ brain (Garza-Lombó et al., 2018; Pokusa and Trančíková, 2017). In an analogous way, ETM unbalance leads to increased production of ROS and OS; OS is also a key factor in Aβ toxicity in neurons (Li et al., 2017).

Fe is accumulated in amyloid plaques in the AD brain (Garza-Lombó et al., 2018; Li et al, 2017), where is involved in a complex vicious cycle of neuronal damage. The metal has a high affinity for binding with Aβ in vitro, promoting its aggregation and accelerating the formation of oligomers (Garza-Lombó et al., 2018; Li et al, 2017). In turn, Aβ aggregation in the brain leads to inflammation and OS, which results in further Fe deposition (Lavado et al., 2019). Additionally, APP levels and Fe homeostasis are also interconnected. More specifically, Fe modulates APP transcription may elevating the production of Aβ by increasing their expression (Li et al, 2017); APP contains a non-canonical Fe response element (IRE) in the 5’ encoded region of mRNA, which allows the translation of mRNA to be placed directly under Fe regulatory proteins (IRPs), specifically IRP1. Concomitantly, APP plays a role in Fe homeostasis containing a sequence that allows it to interact with Fpn and improve Fe export and its ferroxidase activity (Lavado et al., 2019). Pertaining to OS, Fe chelated by amyloid plaques has been recognized as a primary location of intracellular free radical generation, mainly by the Fenton reaction (Pokusa and Trančíková, 2017) and by catalyzing the formation of H₂O₂ (Garza-Lombó et al., 2018). In this manner, high Fe concentrations cause the antioxidant system to become overburdened, resulting in cellular damage over time. In sequence, enhanced OS results in a higher production of Aβ (Li et al, 2017), with Aβ altering redox-inactive Fe³⁺ (which is generally present as a reserve form of Fe) to redox-active Fe²⁺, which through Fenton reactions, leads to the production more toxic free radicals (Lavado et al., 2019).

Both Cu and Zn can bind to Aβ in vitro with high affinity, and in neuronal cells, these interactions result in Aβ aggregation and generation of OS (Pokusa and Trančíková, 2017). Concerning Cu, their metal ions bounded to Aβ can increase their proportions of β-sheet and β-helix structures, responsible for Aβ aggregation. Concomitantly, Cu-mediated increase of the phosphorylation of APP, facilitates its proteolytic cleavage to generate Aβ (Li et al., 2017), while on the other hand, as APP contains a Cu binding site, APP may function as a Cu transporter (Bonda et al., 2011). Once more, Cu-mediated Aβ oligomer cytotoxicity involves OS (Li et al., 2017). More exactly, Cu bound to the Aβ peptide can produce O2•⁻ and H₂O₂ through reduction of Cu⁺² to Cu⁺ [Cu²⁺ + Aβ → Aβ + Cu⁺ (equation 6); Cu⁺+O₂ → Cu²⁺+ O2•⁻(equation 7) and O2•⁻+ O2•⁻+2H⁺→H₂O₂ (equation 8)]; Subsequently •OH can be produced via Fenton reaction [Cu⁺+ H₂O₂ →Cu²⁺+•OH +OH⁻ (equation 8)], increasing OS and neuronal death. Concurrently, Aβ can be converted to reactive •Aβ [Cu²⁺+ Aβ → • Aβ + Cu⁺ (equation 9)], which can also cause lipid protein oxidation with production of 4-hydroxy- 2-nonenal (HNE) carbonyl [•Aβ + Lipid protein → HNE Carbonyl + Aβ (equation 10)] (Kanti Das et al., 2015). Additionally, Cu can interact with APP in an electron transfer reaction that reduces Cu²⁺ to Cu⁺, enhancing the production of a •OH intermediate (Bonda et al., 2011).

With respect to Zn, in vitro work has shown that Zn binds to Aβ with greater affinity then Fe or Cu, upon a wide range of pH. The binding of Zn occurs in histidine residues in the C-terminus of Aβ, with subsequent promotion of aggregates. Paradoxically, although excessive Zn enhances Aβ toxicity, Zn changes the conformation of Aβ preventing Cu or Fe from interacting with the protein. In this way, low Zn concentrations can prevent Cu/Fe-Aβ induced H₂O₂ and free radicals, might ameliorating the OS burden (Kanti Das et al., 2015; Li et al., 2017). Moreover, it has been reported that Aβ plaques and hot spots of accumulated Cu and Zn metal ions co-localize, an interesting dynamic among Aβ, Cu and Zn is described. It has been suggested that while Zn ions induce amyloid aggregation in vitro, Cu inhibit it through competing with Zn for histidine residues. The strongest inhibitory effect occurs at a Cu: Aβ molar ratio of about four. Above this value, Cu ions themselves induce aggregation. Studies on synthetic Aβ (1–40) and Aβ (1–42) peptides show also that at physiological pH, Aβ binds the same ratio of Cu and Zn ions, whereas in acidic conditions Cu ions replace Zn ions (Bagheri et al., 2018). Interestingly, decreased availability of Zn can also contribute to the accumulation of Aβ, as Zn regulates the degradation of Aβ directly through modulation of protease structure and indirectly by increasing protease expression (Nutall and Oteiza, 2014). Concerning ETM mixtures, it has also been found that interactions among Fe, Cu and Zn may influence Aβ aggregation (Karri et al., 2016).

Mn accumulation in young adults has already been associated with diffuse amyloid plaques (in 51%) suggesting a direct impact of the metal on the development of AD. However, contradicting information exits mentioning that Cu and Zn have higher interaction with Aβ aggregation than Mn, with data suggesting weak and transient interactions in this binding. Moreover, exposure of neuroblastoma cell line SH-SY5Y cells to Mn leaded already to decreased viability and lower expression of APP, which could limit non-amyloidic cleavage. Furthermore, it is known that Mn exposure can change the Fe²⁺: Fe³⁺ ratio resulting in oxidative environment, which may drive APP toward amyloidogenic cleavage in the human brain, although this was not tested directly. Nonetheless, additional studies are required to better understand the effects of Mn on Aβ aggregation and the molecular mechanisms involved (Martins Jr et al., 2019).

4.3. Aberrant Fe/Cu/Zn/Mn in neurofibrillary tangles

The primary constituents of NFT, another one of the primary neuropathological signatures for AD, are paired helical filaments (PHF) composed of hyperphosphorylated tau phosphoproteins (Rao and Adlard, 2018). Tau proteins are normally abundant in the brain, where support neuronal structures and functions (Kim et al., 2018). It is established that tau can bind to microtubules and regulate their dynamics (Huang et al., 2014); Yet, over the years novel roles of tau in both normal physiology and disease have been identified, with reported functions in axonal transport, protein trafficking, cognitive function and quite remarkably, interactions with Aβ and also with α-synuclein (α-Syn). The protein α-Syn is the main component of Lewy Bodies (LBs), which is a neuropathological hallmark of PD (Rao and Adlard, 2018).

Under normal physiological conditions, tau exhibits low levels of phosphorylation being their functions regulated by its degree, thus modulating the binding of tau to microtubules and the axonal transport (Rao and Adlard, 2018). The phosphorylation and dephosphorylation of this protein, is controlled by tau protein kinases (PK) and protein phosphatases (PP), respectively. The functional imbalance between PK and PP contributes to tau hyperphosphorylation and AD pathogenesis, and other disease states (Wei et al., 2020). An abnormally hyperphosphorylated state of the tau protein can cause it to dissociate from the microtubule, and aggregate into PHF and then into NFTs, with loss of tau function. As consequence, abnormal tau tangles accumulate in neurons, causing neuronal toxicity and neurodegeneration. Accordingly, tau is demonstrated to be highly phosphorylated in the later stages of neurodegeneration (Huang et al., 2014; Kim et al., 2018; Rao and Adlard, 2018). It is also speculated that a critical component of Aβ induced neurotoxicity is the activation of PK, which results in hyperphosphorylation of tau. Conversely, tau may also be involved in Aβ -induced neurotoxicity as suggested by experimental works. In this studies, transgenic mice expressing human APP, tau suppression resulted in an inhibition of Aβ production, and associated neuronal dysfunction (Rao and Adlard, 2018). In addition, OS is considered as a primary factor of NFT formation in AD, but the mechanism involved between OS and tau hyperphosphorylation remains unclear. Nevertheless, it has been proposed that tau hyperphosphorylation may be induced by OS stress through direct upregulation of tau PK, particularly glycogen synthase kinase-3beta (GSK-3β) (Liu et al., 2015).

Cerebral ETM dysregulation contributes for the progression of tau-mediated neurodegeneration, and namely elevations in Fe levels in affected brain regions correlate with the formation of NFTs in tauopathies (Rao and Adlard, 2018). Indeed, higher Fe concentrations are found in brain regions which accumulate PHF and NFT (Rao and Adlard, 2018) and it is also known that Fe binds to tau, promoting its aggregation in Fe-enriched regions. It has been posited that tau hyperphosphorylation can be modulated by Fe through the aberrant activation of PKs (Li et al, 2017; Rao and Adlard, 2018). On the other hand, tau dysregulation has been implicated in Fe dyshomeostasis, with studies revealing that tau-KO mice accumulate Fe in brain areas with consequent age-dependent neurodegeneration. Once more, Fe imbalance is hypothesized to be a source of OS in tauopathies, with interaction between NFTs and Fe reported to act as a source for ROS in neurons (Rao and Adlard, 2018).

With respect to Cu, this metal can also bind to tau and causes its aggregation in vitro, with in vivo results showing that Cu exposure induce tau hyper-phosphorylation through PK, generating H₂O₂ (Li et al., 2017). Additionally, while is established that Aβ/Cu interaction induce tau phosphorylation, other studies confirm that Cu alone can modulate this reaction in the absence of pathological Aβ. Indeed, studies in vitro and in transgenic mouse expressing wild type human tau (hTau) indicate a direct and independent effect (Li et al., 2017; Voss et al., 2014). The in vitro treatment of Cu to a tau R3 peptide, which is a tau fragment used to simulate the aggregation behavior of full-length tau protein, resulted also in the induction of conformational changes, in which the peptide adopts a monomeric α-helical structure and β-sheet structure. The formation of these two distinct structures facilitates the self-aggregation and assembly of the tau protein into paired helical filaments (Kim et al., 2018; Li et al., 2019b). Despite these results, the mechanism by which tau phosphorylation is modulated by Cu is not entirely clear, as in vitro studies implicate one PK, cyclin-dependent kinase 5 (CDK5), while the in vivo data implicate GSK-3β (Voss et al., 2014). Concerning the involvement of OS, a fragment of tau protein has been already shown to induce Cu reduction, initiating Cu-mediated generation of H₂O_2. (Liu et al., 2015).

Other experiments revealed that Zn effects in tau can occur by direct binding, which affects tau’s properties and behaviors in a way that contributes to tau toxicity (Huang et al., 2014). In tandem, Zn induce tau hyperphosphorylation, again, by activating kinases, and possibly GSK-3β (in the same way as proposed for the other ETM) (Huang et al., 2014; Kim et al., 2011). Thus, Zn binding and phosphorylation seem to represent two independent events for tau, although Zn effects on tau hyperphosphorylation pathways seems to be less important to tau toxicity, than Zn binding directly to the protein (Huang et al., 2014). Inversely, abnormally decreased Zn levels can also contribute to pathologic events related to AD, such as by induction of OS, which can lead to disruption of microtubule stability and accumulation of phosphorylated tau in NFT (Nutall and Oteiza, 2014).

With respect to Mn, challenging neuronal cell lines with Mn also leads to the hyperphosphorylation of tau, and tau mediated neuronal death (Martins Jr et al., 2019; Pokusa and Trančíková, 2017). This process may involve the activation of extracellular signal-regulated kinase (ERK) mitogen-activated protein kinase (MAPK); ERK MAP is by its turn, involved in the activation of the PK GSK-3β (Cai et al., 2011).

5. UNBALANCED ESSENTIAL TRANSITION METALS IN PARKINSON DISEASE

PD is the second most common progressive ND and the most frequent motor system disorder, strongly associated with ageing, affecting mainly the population over age 65 (Wojda et al, 2008). PD is defined by parkinsonism, a motor syndrome which includes resting tremor, bradykinesia, rigidity, and postural instability. Non-motor symptoms (dementia, depression, sleep disorders) also manifest and can arise even before motor diagnosis (Obergasteiger et al.,2018). Bradykinesia, rigidity, tremor, and other motor signs are attributed to the selective degeneration and loss of DAergic neurons (Surmeier et al., 2017; Trist et al., 2018; Wojda et al, 2008). These DAergic neurons innervate the basal ganglia, a subcortical structure involved in the control of movement and action selection. Although there are palliative treatments, there is no proven strategy for slowing the progression of PD (Surmeier et al., 2017).

In the SN, profound loss of DAergic neurons is accompanied by filamentous protein inclusions termed Lewy bodies (LBs) in the surviving neurons, both representing the main pathological hallmarks of PD. In fact, LBs are considered a hallmark pathology not only in idiopathic and autosomal dominant familial PD, but of a group of neurological disorders called synucleinopathies (Hawkes et al., 2007; Martin Jr et al., 2019; Obergasteiger et al., 2018). LBs are mainly composed of phosphorylated α-Syn, which is a chaperon protein with 140 amino acids, widely expressed in the neural tissue, and predominantly localized in the presynaptic terminals. Under normal conditions, α-Syn plays important roles in the regulation of synaptic plasticity, vesicle transport, and DAergic neurotransmission (Martin Jr et al., 2019). While in healthy brains, α-Syn is mostly present in a monomeric form, in PD brains a conformational change occur triggering the formation of small oligomers, which would eventually evolve into higher order structures. The instance when they become cytotoxic and the exact pathologic conformational shape are still controversial, but a growing body of evidence indicates insoluble α-Syn as the pathogenic molecular species (Obergasteiger et al., 2018). It is noteworthy that although LB are mainly composed of aggregated α-Syn, other proteins such as Aβ and tau are sometimes found at neuropathological examinations of PD brains. Aβ and tau are, as mentioned, pathological signatures of AD, leading some authors to classify PD as a proteinopathy (Martin Jr et al., 2019).

Again, oxidation is associated with α-Syn aggregation and has been proposed as one of the mechanisms responsible for the in vitro formation of cross-linked α-Syn oligomers. Correspondingly, in vitro studies exist demonstrating that mitochondrial respiratory-chain defect is responsible for α-Syn oligomerization due to increased ROS production (Esteves et al., 2009). Again, and as important sources of ROS, ETM accumulate also in the brains of patients with PD, suggesting that these metals are involved in PD pathogenesis via ROS-generating pathways (Lan et al., 2016). Occupational long-term exposure to Cu, Fe or Mn (as well as to toxic heavy metals such as Pb), alone or in combination, has been associated with an enhanced risk of developing PD (Bisaglia and Bubacco, 2020). The transporter DMT1 has been shown to increase in the SN of PD brains (Garza-Lombó et al., 2018), suggesting higher metals load. Further, higher intracellular concentration of ROS relates to the release of other biometals from their binding to metalloenzymes (Pokusa and Trančíková, 2017). Additionally, and as recently reviewed, synergistic effects may derive from the combination of different metals, for instance, the combined exposure to Fe/Cu, Pb/Cu, Pb/Fe, Hg/Mn, etc. when compared with the effects of single metals (Bisaglia and Bubacco, 2020).

5.1. Distribution of brain Fe, Cu, Zn and Mn in Parkinson disease

Significant differences in Fe levels can be found in the blood, serum, or cerebrospinal fluid (CSF) of PD patients when compared with healthy subjects; differently, Fe is increased in the SN of PD brain (Garza-Lombó et al., 2018; McLeary et al., 2019). Both neurons and microglia are known to accumulate Fe, and the severity of PD symptoms correlates with the extent of Fe accumulation (McLeary et al., 2019). Elevated Fe in the PD SN has long been associated with neurotoxicity via various mechanisms such as oxidative, ferroptosis, and deleterious interactions between DA and Fe (Genoud et al., 2017). Nonetheless, diverse Fe distributions within different brain regions have been observed throughout the progression of PD. An explanation of this phenomenon can be found in differential expressions of Fe trafficking and storage molecules such as Fpn, and Ft in the affected brain parts (Pokusa and Trančíková, 2017).

Cu levels are increased in PD blood serum, with high serum Cu correlated positively with the disease severity, although a more recent study found a reduced Cu status index in PD serum; Cu is increased in the CSF of PD cases (Montes et al., 2014). In the brain, the SN DAergic neurons which typically degenerate in PD, are known to be rich in Cu (Garza-Lombó et al., 2018). However, the basal ganglia of PD disease patients show decreased Cu (Montes et al., 2014), in the same way as the expression of the Cu transporter CTR1 (Bisaglia and Bubacco, 2020). In addition, Cu in the soluble fraction of brain tissue homogenates is significantly reduced in PD cases, being almost half the normal level, while is increased in the insoluble fraction (Genoud et al., 2017; Pokusa and Trančíková, 2017; Wojtunik-Kulesza et al., 2019). Overall, the varied levels of Cu detected across different tissues and fluids, may be the result in the metal being unable to leave a particular region and becoming deficient in another. As a matter of fact, a general picture is now emerging in which either excessive levels of Cu or its deficiency appears both detrimental for PD. The cellular pathways associated with both high and low Cu levels lead to increased OS conditions, suggesting that converging intracellular stressful conditions could mediate cell damage in both scenarios (Genoud et al., 2017). Concerning the detrimental effects of decreased Cu levels, it is described that under normal conditions Cu regulates Fe homeostasis in the brain via Cu-dependent ferroxidase Cp activity (Bisaglia and Bubacco, 2020; Pokusa and Trančíková, 2017), which promotes the oxidation of Fe²⁺ to Fe³⁺, so that Fe³⁺ can be removed from the brain (Montes et al., 2014). In PD, decreased Cu and increased oxidative environment in the SN could turn the ferroxidase activity of Cp defective contributing to Fe accumulation (McLeary et al., 2019; Montes et al., 2014). In this way, some indirect toxicity mediated by augmented concentrations of Fe could be a consequence of low levels of Cu (Bisaglia and Bubacco, 2020). The determination of elevated levels of Fe, and decreased levels of Cu and Cp in PD brains, consubstantiate these information’s (Bisaglia and Bubacco, 2020; Pokusa and Trančíková, 2017).

With respect to to Zn, elevated levels have been found mainly in the SN, and in other tissues of PD patients (Pokusa and Trančíková, 2017). However, overall data regarding Zn concentrations in PD are conflicting, and many aspects of their role in this disease are still incipiently comprehended (Genoud et al., 2017).

PD and Mn-induced parkinsonism are considered distinct disorders, sharing however some similarities in their pathophysiological mechanisms and a few motor symptoms. In both, there is a characteristic extrapyramidal syndrome primarily involving DAergic neurodegeneration, with shared molecular mechanisms (Benedetto et al., 2010). Regarding the differences, Mn-induced parkinsonism is mainly distinguishable from PD by the absence of LB and the lack of therapeutic response to levodopa (a drug used to treat early stages of PD) (Kwakye et al., 2015). Additionally, some authors have suggested that Mn intoxication is associated with preservation of the nigrostriatal DAergic pathway, and that chronic Mn intoxication causes parkinsonism-like effects by damaging output pathways downstream of the nigrostriatal DAergic pathway, in areas such as the GB, an area with propensity to accumulate high amounts of Mn (Aschner et al., 2009). Other researchers defend that Mn alters nigrostriatal DAergic levels producing a Parkinsonism-like disorder, while other authors affirm that Mn alterations are related to different aspects to those associated to PD in both etiology and pathology (Avila-Costa et al., 2016). Despite the current polemic, it is considered that prolonged and chronic occupational exposure to Mn (>1 mg/m³) represents a risk factor for PD (Kwakye et al., 2015). Furthermore, DAergic neurodegeneration associated with PD or PD-mimicking drugs and Mn neurotoxicity share multiple common effector mechanisms. They are namely, mitochondrial dysfunction, ATP depletion, aberrant signal transduction, activation of cell death pathways, OS and protein aggregation (Benedetto et al., 2010).

5.2. Aberrant Fe/Cu/Zn/Mn in Lewy bodies

Factors driving LB formation in both familial and idiopathic cases of PD, include several agents promoting α-Syn to misfold and aggregate; they include ROS and elevated concentrations of some metals, which can bind α-Syn at distinct metal-binding sites aggravating the formation of dense α-Syn fibrils, and inducing neuronal toxicity (McLeary et al., 2019; Pokusa and Trančíková, 2017). Aggregated α-Syn provokes metal ions to mediate OS, thus closing the harmful circle between α-Syn aggregation and the generation of ROS (Pokusa and Trančíková, 2017).

Significant increases in Fe concentration and redistribution of Fe from the cytoplasm to the perinuclear region within α-Syn inclusions, are documented (Lan et al., 2016; Pokusa and Trančíková, 2017; Zucca et al., 2017). A series of studies has been performed aimed at explaining Fe and α-Syn dependencies. Most of them are based on a theory assuming that an increase in α-Syn concentration influences Fe accumulation, and that, on the contrary, Fe causes protein aggregation (Wojtunik-Kulesza et al., 2019). Effectively, in vitro studies have shown that Fe induces a conformational change of α-Syn to the β-sheet conformation, forming fibrils that are present in LB (Zucca et al., 2017). Fe interacts with α-Syn, at least at two biological levels: the first involves the translation of the protein via Fe-responsive elements (IREs) that exist in the 50-UTR of the α-Syn mRNA. The second consists of direct binding of Fe to the protein itself, leading to its abnormal folding and formation. Although the binding sites for Fe in α-Syn are not clearly identified, previous studies demonstrated a preferential binding of Fe²⁺ in the C-terminal region of α-Syn and Fe³⁺-mediated aggregation of the protein (Garza-Lombó et al., 2018; Lan et al.,2016; McLeary et al., 2019). In addition, recent studies indicated that microglial Fe accumulation may promote a senescence-associated secretory phenotype capable of inducing neuronal α-Syn aggregation (McLeary et al., 2019). Accordingly, α-Syn overexpression is thought to promote neuronal Fe accumulation (Garza-Lombó et al., 2018; McLeary et al., 2019). Post-translational modifications of α-Syn have been found to regulate Fe transport, with a membrane-bound tetrameric form of α-Syn exhibiting ferrireductase activity, which is reduced in PD brain tissue (McLeary et al., 2019). In vivo studies corroborate these findings, showing significant dependencies between ferrireductase, α-Syn, and PD (Wojtunik-Kulesza et al., 2019). In vitro experiments showed also that Fe-induced OS can contribute to protein aggregation (Esteves et al., 2009). In a positive feed-back process, the formation of H₂O₂ from α-Syn, followed by its conversion to •OH via Fenton’s reaction, can establish the conditions for further aggregation of the protein (Wojtunik-Kulesza et al., 2019; Zucca et al., 2017).

As noted above, Cu levels are decreased in PD brain. Nevertheless, free unbound Cu is also involved in the pathogenesis of PD, which is thought to be mainly associated with OS induction and its ability to form a complex with α-Syn, inducing it conformational change, with promotion of oligomerization and aggregation, among other events (Meamar et al., 2016; Montes et al., 2014; Pokusa and Trančíková, 2017). Nonetheless, has been posited that α-Syn has two sites for Cu binding per monomer, and that the protein has high affinity for Cu, effectively causing its fibrillation. This phenomenon yields further precipitation of similar proteins and is ultimately thought to represent the “seeding” that gives rise to LB; in accordance, Cu is present in LB bodies at relatively high concentrations (Montes et al., 2014). On the other hand, there is experimental evidence showing that the Cu-α-Syn complex induces changes in Cu’s redox properties, and, thus, this complex has been linked to increased H₂O₂ production from ascorbic acid oxidation and, in turn, DA oxidation by H₂O₂ (Montes et al., 2014; Pokusa and Trančíková, 2017).

Additionally, it has been known that α-Syn binds both Cu and Fe, the loading of Cu into the protein produces small changes in the Fe binding kinetics, suggesting different binding sites for both metals (Montes et al., 2014).

While the role of Zn in PD is still unclear, some works suggested that Zn also have influence on the development of PD, again, through α-Syn aggregation. The primary region of Zn binding to α-Syn is the C-terminus (is the same way as mentioned for other ETM) (Martin Jr et al., 2019; Wojtunik-Kulesza et al., 2019). With respect to to Mn, nuclear magnetic resonance studies have shown that α-Syn has a poor affinity for Mn in its C-terminal binding site; even so, the metal can contribute to α-Syn toxicity, triggering misfolding and accumulation of the protein. In truth, emerging evidence indicates that α-Syn oligomerization is a major culprit for Mn-induced neurotoxicity, with α-Syn oligomerization proposed to be the major factor responsible for Mn-induced autophagy dysregulation and neuronal injury. Other mechanisms have also been documented, such as Mn-induced overexpression of α-Syn. (Martin Jr et al., 2019).

In addition, some authors mention that the contribution of more than one metal, when present concurrently, can result in synergistic effects on α-Syn aggregation. In fact, the interaction between α-Syn and metal ions has been proposed to stabilize a partially folded conformation of the protein, by decreasing the electrostatic repulsion between their negative charges, which are mostly present in its C-terminal region (Bisaglia and Bubacco, 2020).

5.3. Aberrant Fe/Cu/Zn/Mn and dopamine oxidation

The loss of DA is considered one of the most distinctive PD hallmarks. This neurotransmitter is a stable molecule when stored in synaptic vesicles with an acidic pH inside; however, when released into the extracellular milieu or in the cytosol, is easily oxidized undergoing spontaneous autoxidation process, which leads to the formation of both ROS and DA quinones (DAQs) (Bisaglia and Bubacco, 2020; Yang et al., 2016). These reactions are catalyzed by oxygen or enzymes such as tyrosinase, and by metals (Zucca et al., 2017). Among PD patients and using animal models, oxidized DA as DAQ have been known to be toxic to DAergic neurons (Yang et al., 2016). DAQ forms adducts with several proteins involved in different cellular pathways, which includes protein degradation and processing and neurotransmitter synthesis. Experimental studies showed that DAQ can also forms adducts with several rat mitochondrial proteins including those of complexes I, III and V of the electron transport chain and oxidative phosphorylation, inducing mitochondria dysfunction, may resulting in OS (Zucca et al., 2017). Moreover, at physiological pH DAQ undergoes cyclization to leukoaminochrome, which is subsequently oxidized to aminochrome, reducing oxygen to O₂ •⁻ (Zucca et al., 2017). Aminochrome form also adducts with proteins such as α-Syn and neuromelanin, which is a containing Fe molecule and a brain pigment that may contribute to neurodegeneration by triggering neuroinflammation (Dias et al.,2013).

The contribution ETM to DA oxidation have been reported by several studies (Bisaglia and Bubacco, 2020; McLeary et al., 2019). Some of them suggest that neurodegenerative processes observed in PD may be initiated by aberrant reactions between DA metabolites and redox-active Fe²⁺: Fe³⁺, as a result of defective regulatory and/or antioxidant pathways. Hence, neuronal Fe may overwhelm the mechanisms for Fe storage, thereby releasing the metal to the cytoplasm, where it could interact with DA oxidation products not contained within secretory vesicles to form toxic metabolites (McLeary et al., 2019). The formation of Fe-DA complex, in the form of Fe³⁺-DA, has also been proposed as a selective mechanism of neurotoxicity in DAergic neurons or even cells expressing other monoaminergic transporters such as norepinephrine, and serotonin transporters. The selectivity of Fe-DA complex neurotoxicity depends on the ability of cells to take up such neurotoxic complex. Various factors occurring in PD have been suggested to explain the high accumulation of Fe in the SN of PD patients: i) increased permeability or dysfunction of BBB during the disease; ii) increased pro-inflammatory state, since neuroinflammation can perturb Fe homeostasis on different brain cells; iii) increased expression of lactoferrin receptors (involved in Fe uptake through lactoferrin) in neurons and microvessels of the mesencephalon (Zucca et al., 2017).

Analogously, Cu has been demonstrated to increase the oxidation process of DA leading to a variety of potentially toxic species, which again includes DAQ as well as O₂• ⁻, H₂O₂ and •OH. The presence of labile pools of Cu ions in the brain, is probably the major source of DA oxidation. In addition to the direct reactivity of Cu towards DA, the oxidation process can also be mediated by Cu ions bound to coordinating ligands or peptides and proteins involved in neurodegenerative processes; actually, a major role in promoting Cu-induced DA oxidation is played by the protein α-Syn (Bisaglia and Bubacco, 2020). As previously mentioned, ROS may on the other hand, enhance Cu accumulation in the brain. For instance, 6-hydroxydopamine (6-OHDA), a DA neuron-specific ROS generator, induces an increase of Cu in the brain regions related to DAergic pathways in a rat model of PD. It is yet to be determined which one, ROS or Cu, is the first to initiate DAergic neurodegeneration (Lan et al., 2016).

Zn accumulation is also involved in the DAergic neuron death, as documented by studies with rodent PD models (Yang et al., 2016). Previous work has revealed that cytosolic accumulation of labile Zn²⁺ is a component of the pathogenic events leading to DA neuron death, with others evincing that extracellular Zn²⁺ influx into nigral DAergic neurons is implicated in neurodegeneration induced by DAergic neurotoxins (Sikora et al., 2020). Additionally, it has been reported that Zn and DA can synergistically increase cell apoptosis and deplete striatal DA content in vitro (Yang et al., 2016).

Other studies have reported contradictory results on the DAergic effects of Mn, including decrease, increase, both, or no modification in either SN and/or striatum. It appears that lower dosages of Mn augment DA and its metabolite concentrations, whereas the inverse is detected with more significant Mn exposures. Accordingly, it has been proposed that higher Mn dosages can drastically accelerate DA and other catecholamine oxidation, which concomitantly intensifies ROS formation (Avila-Costa et al., 2016). Mn also elevates intracellular H₂O₂ and related peroxides in DAergic neurons. Additionally, it is proposed that endogenous extracellular, but not intracellular DA potentiates Mn toxicity and that Mn-induced neurodegeneration requires the DAergic neuron-specific DA re-uptake transporter, DAT-1. Upon Mn exposure, DAT-1 facilitates the intracellular transport of toxic species into DAergic neurons. The current literature provides also a conceptual framework for addressing the synergistic mechanism of Mn and DA in DAergic neurodegeneration (in the same way as mentioned for Zn). In this way, Mn may enter DAergic neurons via DAT as a complex with DA-derivatives, such as dopaminochrome. This hypothesis provides explanations both for the specificity of Mn toxicity towards DAergic neurons as well as for the synergistic toxicity of extracellular DA and Mn (Benedetto et al., 2010).

6. ENVIRONMENTAL EXPOSURE TO TOXIC HEAVY METALS AND NEURODEGENERATIVE DISEASES

As a brief illustration of the association of two of the most challenging toxic heavy metals with ND, supportive literature exists positing that early life exposure to Pb is linked to ND, such as AD and PD (Saint-Hilaire et al., 2012). Mechanisms of Pb toxicity include the ability of Pb to bind SH groups of proteins Cys and to mimic or compete with Ca²⁺, Fe²⁺, and Zn²⁺, disrupting corresponding biometal-dependent mechanisms. The generation of oxidative damage by Pb in vitro and in vivo suggests that ROS also participate in Pb toxicity. In addition, Pb can interfere with heme synthesis leading to the accumulation of heme precursors, such as delta-aminolaevulinic acid (ALA) through Pb-induced aminolaevulinic acid dehydratase (ALAD) inhibition. Excess levels of ALA can lead to it subsequently oxidation generating O₂ •⁻, H₂O₂ and •OH. Additionally, Pb per se has been reported to stimulate Fe²⁺-initiated lipid peroxidation (Garza-Lombó et al., 2018). Specifically, and with respect to AD, SH-SY5Y cells treated with Pb displayed a significant reduction in the mRNA and protein levels of DNA methylation enzymes, which resulted in increased APP expression and reduced Aβ degradation. Another study showed that neonatal rats exposed to Pb exhibited an elevation in APP expression and its amyloidogenic Aβ product. These data provide evidence that Pb poisoning during brain development can increases the risk of AD (Li et al., 2017). Other studies linking Pb to PD, hve shown that the metal cxa disrupt the DAergic function concomitant with induction of OS. Some authors have suggested that cumulative exposure, might be more relevant than recent exposure for the development of PD. Even so, acute Pb exposure increases spontaneous DA release, inhibits depolarization-evoked DA release, decreases DA neuron spontaneous activity in vivo, and can alter DA-dependent behaviors. Several studies have indicated that DA synthesis, turnover, and uptake in the midbrain and basal ganglia are decreased after Pb exposure, in the same way as is DA D₁/D₂ receptor expression (Weisskopf et al., 2010).

With respect to Cd, this metal can cross the BBB and long-term exposure causes its accumulation in the brain, which can activate various signaling cascades to stimulate inflammation, OS, and lead to neuronal death, eventually influencing attention, and cognitive function. A large number of clinical and experimental investigations suggested already that Cd neurotoxicity is linked to the pathogenesis of AD. Cd exposure and its association with Aβ production have been well-described, in the same way as a pulse-chase study showed also that Cd chloride increased APP (Li et al., 2017). Concerning PD, studies in a DAergic neuronal model of PD showed that α-Syn exhibits neurotoxic properties upon acute Cd exposure, causing cell death after OS, increased Cd uptake, among other detrimental events (Chong et al., 2017). Cd has also been reported to cause PD and motor dysfunctions; yet, the mechanism how Cd causes the motor dysfunction remains elusive, despite DAergic alterations and down regulation of DA D2 receptors, are known to occur (Gupta et al., 2019).

Regarding mixtures, in vivo studies showed that rats exposed to the ternary mixture of arsenic (As), Cd, and Pb bear the essential features of AD-like pathology, and increased processing of hippocampal and cortical APP gene, causing cognitive dysfunction. In another study, it has been found that independently or in mixture, Fe, Pb and Mn may raise the risk of PD. Limited numbers of studies have found sufficient indication that the quaternary metal mixture of Pb, Cd, As and Hg may have high capability to cause cognitive dysfunction, however the evidence of mixture and disease relation is not clear (Karri et al., 2016).

Overall, Fe, Cu, Zn or Mn share important roles in fundamental brain functions, such as cellular respiration, neurotransmitter synthesis, myelination, and synaptic transmission. The preservation of their homeostasis involves mutual transporters, which includes DMT1, TfR, CTR1 and members of the ZIP family, as well as MT. Moreover, each ETM homoeostasis is dependent of other ETM control. The involvement of transporters such as the Cu dependent Hp and Cp ferroxidades, for Fe uptake, the Cu uptake mediated by Zrt (Zn-regulated transporter) or Irt (Fe-regulated transporter)-like proteins 4 (ZIP4), and the interplay between Mn and Fe, with one metal substituting the other warranting some enzymatic activities, constitute examples. In this way, global EMT homeostasis is a vital and a delicate process, as each ETM balance is dependent on other EMT equilibriums, due to their complex interplay and shared biological activities. Indeed, aberrant ETM homeostasis is a central feature in ND, leading to disturbed distributions of some of these metals in the brain, as significant amounts accumulate in specific areas, and lead to metal deficiency in others. The ensuing disruption of processes involving complex interplays among ETM, may result in loss of metal balance. Additionally, excessive ROS production and OS can be consequences shared by decreased/increased ETM, steaming from aberrant homeostasis. The mechanisms involved may include Fenton/Haber-Weiss reactions (with the involvement of both Fe or Cu) and Fenton-like or lipid peroxidation reactions, with commitment of several ETM. Furthermore, Fe/Zn-MT or Fe/Mn interactions may lead to further Fe-mediated production of ROS. Global imbalances in brain ETM plus OS, are contributors to distinctive hallmarks of AD or PD. In AD, both ETM excess and deficiency, may contribute to their pathogenesis of, being Cu or Zn deficiency an example. Concomitantly, excessive ETM such as Fe, Cu, Zn and Mn, can bind, co-accumulate and trigger detrimental modifications in amyloid plaques or NTF, in AD patients’ brain. Positive feed-back processes contributing to cellular damage, via shared mechanisms, are generally involved. Among them, OS/increased ETM accumulation/ETM-mediated OS positive loops, are key conditions for amyloid plaques toxicity, among other shared pathways.

As α-Syn aggregation with loss of DAergic neurons represent main pathological hallmarks of PD, OS/ETM disruption constitutes a common and relevant mechanism for these events. Fe, Cu, Zn and Mn, all contribute to α-Syn aggregation, with ensuing Fe/Cu-mediated OS, and synergistic effects on α-Syn aggregation due to the concurrent presence of more than one metal, is also described. DA oxidation mediated by the mentioned ETM have been also reported in several studies.

In conclusion, ETM such as Fe, Cu, Zn and Mn can act as a dangerous endogenous neurotoxic mixture, involved in complex positive loops pathways that may lead to neuronal damages, contributing to relevant AD or PD hallmarks. Their presence as a mixture implies intricacies, that should be kept in mind when developing therapies for complex disorders of metal dyshomeostasis, commonly occurring in ND.