Metals in Alzheimer’s Disease

The role of metals in the pathogenesis of Alzheimer’s disease (AD) is still debated. Although previous research has linked changes in essential metal homeostasis and exposure to environmental heavy metals to the pathogenesis of AD, more research is needed to determine the relationship between metals and AD. In this review, we included human studies that (1) compared the metal concentrations between AD patients and healthy controls, (2) correlated concentrations of AD cerebrospinal fluid (CSF) biomarkers with metal concentrations, and (3) used Mendelian randomization (MR) to assess the potential metal contributions to AD risk. Although many studies have examined various metals in dementia patients, understanding the dynamics of metals in these patients remains difficult due to considerable inconsistencies among the results of individual studies. The most consistent findings were for Zn and Cu, with most studies observing a decrease in Zn levels and an increase in Cu levels in AD patients. However, several studies found no such relation. Because few studies have compared metal levels with biomarker levels in the CSF of AD patients, more research of this type is required. Given that MR is revolutionizing epidemiologic research, additional MR studies that include participants from diverse ethnic backgrounds to assess the causal relationship between metals and AD risk are critical.


Alzheimer's Disease
Alzheimer's disease (AD) is the most common cause of dementia worldwide (60-70% of cases), affecting over 55 million people. It is predicted that 74.7 million people will have dementia by 2030 and approximately 131.5 million by 2050. By 2030, the cost of treating and caring for dementia patients will rise to USD 2 trillion (http://www.worldalzreport2 015.org/ accessed on 22 February 2023).
The accumulation of two proteins, amyloid beta (Aβ) and tau, in amyloid plaques and neurofibrillary tangles (NFTs), respectively, is thought to contribute to the development and progression of AD. Although the precise relationship between these two pathologies is unknown, there is evidence that they may interact in a bidirectional manner to promote each other's aggregation and toxicity [1]. Overall, the relationship between Aβ and tau in AD is complex, with multiple feedback loops and interactions with other pathological processes [2].
Aβ is a small peptide derived from the amyloid precursor protein (APP). APP is a transmembrane protein with an unknown function in the brain, but it is thought to be involved in cell adhesion, signaling, and synapse formation [3]. Different enzymes can cleave APP, resulting in the production of a number of peptides of variable lengths. β-Secretase cleaves APP at the N-terminus of the Aβ domain in the amyloidogenic pathway, followed by γ-secretase cleavage within the transmembrane domain, resulting in the production of Aβ peptides with varying lengths [4]. Excessive Aβ formation, aggregation, induced neuroinflammation dependent on oxidative stress [45]. In addition, some essential metals such as Fe [46], Cu [47], Zn, and calcium (Ca) [39] can induce oxidative stress. Fe participates in Fenton reactions and can therefore contribute to the formation of reactive oxygen species [46]. Both the observed disruption of the BBB [48,49] and the apoptosis and necrosis of neurons [37,38] upon exposure to heavy metals may be preceded by oxidative stress, according to experimental evidence. Neurons are extremely sensitive to oxidative stress. Wang et al. [39] proposed that metal ion imbalance could induce oxidative stress, with the following downstream effects: (1) imbalance of protein kinases and phosphatases, increasing tau protein phosphorylation, and (2) imbalance of secretases, resulting in an increase in Aβ production (reviewed in [39]). On the other hand, essential metals also serve as cofactors in enzymes that combat oxidative stress. Cu, Zn, and manganese (Mn) are enzyme components of superoxide dismutase enzymes, while selenium is an enzyme component of glutathione peroxidase [50].
Although there is a substantial body of evidence linking metals to AD-related pathological processes, it is unclear whether disrupted metal homeostasis is involved in the pathogenesis of AD, results from AD pathological processes, or both. Given that AD is a complex disease driven by both genetic and environmental factors, it is unlikely that AD pathogenesis will be explained by a single factor, but rather by the interaction of many.

Heavy Metals in Alzheimer's Disease
Heavy metals including arsenic (As) [51], cadmium (Cd) [49], lead (Pb) [52], and mercury (Hg) [52] can cross the BBB and accumulate in the brain, or they can bypass the BBB and enter the brain directly through the olfactory pathway [53]. Some researchers have hypothesized that early exposure to heavy metals is associated with the later development of AD. Based on their observations of experimental animals, they concluded that early-life exposure to As [54], Pb [55], and Cd [56] may contribute to the development of neurodegeneration later in life, which is consistent with the developmental hypothesis of AD [57][58][59]. Although there is a substantial body of evidence linking metals to AD-related pathological processes, it is unclear whether disrupted metal homeostasis is involved in the pathogenesis of AD, results from AD pathological processes, or both. Given that AD is a complex disease driven by both genetic and environmental factors, it is unlikely that AD pathogenesis will be explained by a single factor, but rather by the interaction of many.

Heavy Metals in Alzheimer's Disease
Heavy metals including arsenic (As) [51], cadmium (Cd) [49], lead (Pb) [52], and mercury (Hg) [52] can cross the BBB and accumulate in the brain, or they can bypass the BBB and enter the brain directly through the olfactory pathway [53]. Some researchers have hypothesized that early exposure to heavy metals is associated with the later development of AD. Based on their observations of experimental animals, they concluded that early-life exposure to As [54], Pb [55], and Cd [56] may contribute to the development of neurodegeneration later in life, which is consistent with the developmental hypothesis of AD [57][58][59].

Arsenic
As is a metalloid that can be ingested through contaminated water, soil, and air, but primarily through drinking contaminated water. More than 220 million people are estimated to consume water that exceeds the permissible level of 10 µg/L [60]. Epidemiological studies suggest that As contributes to cognitive impairment [61] and an increased risk of AD [62], and that elevated As levels in soil are associated with an increase in AD-related mortality [63]. As exposure has also been associated with memory impairments in animal studies [64][65][66]. As exposure also increases Aβ levels [67], promotes tau hyperphosphorylation [32,68,69], tau aggregation [32], oxidative stress caused mainly by mitochondrial dysfunction [70], vascular damage [71], neuroinflammation [34], and apoptosis and the necrosis of neurons [37,38] (Figure 1). In the majority of human studies, there were no significant differences in As levels between AD patients and the controls, although some studies observed a significant increase in As levels in AD patients [72,73] and a positive association with CSF AD biomarkers [74].

Cadmium
Humans are exposed to Cd through food, air, and water [75]. Smokers have Cd levels that are two to four times higher than nonsmokers [76]. Cd may also play a role in the development of AD pathological changes. Cd has been linked in human studies to increased mortality due to AD [77,78] and cognitive decline [79][80][81]. Ruczaj and Brzoska proposed that Cd primarily exerts its effects by inducing oxidative stress [82]. Nevertheless, it also interacts with Aβ [83] and increases Aβ aggregation [30,40], promotes tau hyperphosphorylation [33] and aggregation [42], impairs the BBB [48,49], impairs cholinergic transmission and causes the death of cholinergic neurons in the basal forebrain [84], and disrupts intracellular cation homeostasis by being an anti-metabolite of Zn and replacing it in Zn enzymes [85] (Figure 1). In human studies, there is either an increase [86] or no difference [87] in Cd levels between AD patients and healthy controls (Table 1).

Mercury
Exposure to Hg occurs through food, air, and water, with seafood consumption being the primary source of mercury poisoning [152]. Three-to 5-fold increases in Hg levels in the air and water have been documented as a result of industrialization [153]. A systematic review [154] and meta-analysis [155] demonstrated an association between Hg exposure and cognitive decline and progression of AD, but a subsequent report [156] did not confirm these findings. In addition, a neuropathological study of 286 brains by Morris et al. revealed no correlation between higher brain Hg levels and neuropathological alterations [152]. However, there are multiple molecular mechanisms through which Hg may contribute to the pathogenesis of AD. It promotes Aβ production [157] and aggregation [30], tau hyperphosphorylation [158,159] and aggregation [160], induces oxidative stress [35], and alters calcium homeostasis [161] (Figure 1). Human body fluid Hg measurements yielded contradictory results. Both an increase and a decrease were observed in Hg levels between the AD and control subjects, or there was no change (Table 1). In addition, the CSF Hg level was positively correlated with several CSF AD biomarkers [74], whereas the blood Hg level was positively correlated with the CSF Aβ 1-42 level [145].

Lead
In addition to food, air, and water, humans are also exposed to lead [29] through ingestion. Epidemiological studies have demonstrated that lead exposure contributes to cognitive impairment [162,163]. Moreover, experimental studies have reported an association between Pb and AD pathological changes. Pb interacts with Aβ [31] and increases Aβ production [45,164] and aggregation [31], increases tau hyperphosphorylation [165], compromises the BBB [36], induces epigenetic modifications by altering the expression of AD-related genes [166,167], disrupts intracellular cation homeostasis by interfering with Ca homeostasis and replacing Zn ions in Zn enzymes [168], and induces oxidative stress [169]. In human studies, there was a decrease or no difference in the Pb levels between the AD patients and control subjects (Table 1), whereas a recent MR study found that higher blood Pb levels were a risk factor for AD [170].

Aluminum
Aluminum (Al), the most abundant metal in the Earth's crust [171], is not an essential element for life; however, in its free, solvated, and trivalent forms, Al 3+ is biologically reactive [172], accumulating in the central nervous system [173,174]. In AD-affected brain regions including the entorhinal cortex, hippocampal region, and amygdala, the concentration of Al is higher [175,176]. Al was co-deposited with fibrillar Aβ in amyloid plaques in a study of brain tissue samples from donors with familial AD (fAD) and the PSEN1-E280A (Glu280Ala) mutation [172,177]. Cortical Aβ levels are elevated in donors with this mutation, and this mutation is associated with an aggressive etiology of AD [178]. Aluminum's unique association with Aβ and the high levels of Al found in these brain tissues suggest that Al plays a role in the neuropathology of fAD [177].
When Al binds to various proteins, oligomerization can occur, resulting in conformational changes that prevent proteases from degrading the proteins. In addition, Al 3+ binds strongly to phosphorylated amino acids, causing highly phosphorylated cytoskeleton proteins to aggregate and accumulate [179]. As a result, Al induces the apoptotic death of neurons and glial cells. Al-Aβ co-deposition in fAD has been hypothesized, but its association with intraneuronal NFTs has not been confirmed [177,180], as demonstrated by Mold et al. [181]. While Al binding to Aβ in amyloid plaques is anticipated in the early stages of disease progression [177,178,182], an association with tau may occur in later disease stages [177,178,182]. Numerous studies have investigated the association between oral exposure to Al in drinking water and AD [183]. According to Martyn et al. [184], AD is more prevalent in regions with high levels of Al in their drinking water. In conclusion, even though Al has been proposed as a potential risk factor for AD, there is insufficient evidence to support a causal relationship ( Table 2). Many studies have investigated the association between oral exposure to Al in drinking water and AD; however, more research is required to better understand how genetic, environmental, and lifestyle factors influence the onset and progression of AD.    (6), clinically non-demented individuals (82) 131 [194] CSF Cu intake Aβ 1-42 , t-tau, p-tau 181 Cu intake did not affect the t-tau and p-tau 181 199 , tau protein phosphorylated at serine 199; S100B, S100 calcium-binding protein B; t-tau, total tau; VaD, vascular dementia; VILIP-1, visinin-like protein 1; YKL-40, chitinase-3-like protein 1.

Essential Metals in Alzheimer's Disease
The homeostasis of essential metals is altered in AD patients [26][27][28]. This term refers to metals that are naturally present in the body and play a role in the function of numerous proteins and enzymes or act as second messengers. Sodium (Na), Ca, and magnesium (Mg) are the most abundant essential metals in the human body, while Fe, Cu, Zn, molybdenum (Mo), cobalt (Co), Mn, and chromium (Cr) are present in trace amounts. Many previous studies have also demonstrated the association between essential metals (primarily Fe, Cu, and Zn) and AD pathological changes.

Iron
Many biological processes in the body including the brain are regulated by Fe ions. Fe is essential for protein synthesis [197], cell growth and differentiation [198,199], the regulation of Fe-dependent enzymes [200], oxygen transport [201], and the electron transfer chain in oxidation-reduction reactions [201]. Fe is also crucial for the processes of myelination [202], development [203], and the function of numerous neurotransmitter systems [204]. Both amyloid plaques and NFTs have been found to have elevated Fe concentrations [205]. Fe is also involved in oxidative stress and the formation of reactive oxygen species in the brains of AD patients via the Fenton reaction [46]. Fe also promotes in vitro Aβ aggregation [206], tau protein phosphorylation [207][208][209], and tau aggregation [210] (Figure 1). It is interesting to note that APP is necessary for the persistence of ferroprotein (iron exporter) on the cell surface, and thus promotes Fe release [211].
In meta-analyses, a significant decrease in Fe levels was observed in the plasma [87] and serum [129] of AD patients, but no significant change was observed in the CSF [129] ( Table 1). In contrast, a number of studies observed a correlation between the Fe levels in CSF and various CSF AD biomarkers [74,188,193] (Table 2). Nonetheless, in many observational studies, there was no difference in the Fe levels between the AD patients and controls (Table 1).

Zinc
The brain has a higher Zn concentration than other organs [212]. Zn is essential for neurotransmission because, as an antagonist of glutamate NMDA (N-methyl-D-aspartate) receptors, it protects neurons from glutamate-induced excitotoxic damage [213]. Zn accumulates in amyloid plaques [214], binds to Aβ, and promotes Aβ aggregation and plaque formation [214]. Zn also promotes tau protein aggregation [215], phosphorylation [216,217], and translation [217] (Figure 1). In meta-analyses, however, a significant decrease in Zn levels was observed in the serum and plasma [123] as well as in the hair of AD patients [87], whereas there was no significant change in the CSF [123] and brain [115] levels (Table 1). To date, MR studies have not identified Zn as a risk factor for AD [218][219][220]. An in vivo study demonstrated positive effects of Zn supplementation in mouse models of AD [221], and a small double-blind clinical trial observed the stabilization of cognitive abilities in AD patients after six months [222]. Thus, adding Zn to the diet has been suggested to improve the cognitive abilities of AD patients [223], whereas Loef et al. found no significant benefit of Zn supplementation in AD [224]. In addition, in vivo studies have shown that Zn supplementation promotes the formation of NFTs [225] and Aβ deposition [226].

Copper
Normal brain function requires optimal Cu levels, as indicated by the disruption of its metabolism. Patients with Menkes syndrome, for example, suffer from intellectual deficits and neurodegeneration. This disorder is caused by a sex-linked mutation of the ATP7A gene on the X chromosome (which encodes a protein involved in the transmembrane transfer of Cu ions) and is characterized by the decreased absorption of Cu in the intestine, and consequently, a decreased concentration of Cu in the cytosol of all body cells except in the intestines and kidneys [227]. In Wilson's disease, excessive Cu accumulation in the body is associated with psychosis, parkinsonism, and dementia [228,229]. Cu homeostasis is also impaired in AD [28]. Cu promotes the formation and accumulation of Aβ-oligomers by binding to Aβ [41]. Cu chelation can prevent the cytotoxic effect of the Cu-Aβ complex [230]. Cu accumulates in plaques [231,232], and the interaction between Cu and APP has been demonstrated [232]. Cu can induce both the phosphorylation and aggregation of tau [42,43] ( Figure 1) and its interaction with apolipoprotein E (ApoE) contributes to the pathogenesis of AD. ApoE2 has the highest binding affinity for divalent Cu, Zn, and Fe ions, while ApoE4 has the lowest [233,234]. In meta-analyses, Cu levels in the serum of AD patients increased significantly [87,92,122,133], whereas Cu levels in the brains of AD patients decreased [92] ( Table 1). Recent MR studies [218,220] have surprisingly found that higher Cu levels are protective against AD risk.

Calcium
Ca is an indispensable second messenger that regulates hundreds of signaling pathways crucial for the normal functioning of memory and cognition-related cells and networks [235]. Many neurodegenerative diseases including AD [236] are characterized by a disruption of cellular Ca signaling. The excessive entry of Ca ions through ionotropic glutamate receptors is a known mechanism of excitotoxic neuronal death [237,238]. Ca homeostasis disruption promotes Aβ and tau pathology [239]. However, human studies have produced contradictory results, with both decreased [240,241] and increased Ca [186,242] being risk factors. In recent MR studies, higher Ca levels were shown to reduce the risk of AD [241,243], or no association between Ca levels and AD risk has been observed [218,220] (Table 3). Table 3. Mendelian randomization studies that investigated the role of metals in AD.

MR Analysis Measured Metals
Observed Association [218] Two-sample MR Mg, Ca, Fe, Cu, Zn, Se, P Higher Cu levels as a protective factor for AD risk [170] Two-sample MR Pb (in the blood) Higher Pb levels as a risk factor for AD [241] Two-sample MR Ca (in serum) Higher Ca levels as a protective factor for AD risk [243] MR Ca (in serum) Higher Ca levels as a protective factor for AD risk [219] Two-sample MR Cu, Zn, Fe No association [220] Two-sample MR Ca, Mg, Fe, Cu, Zn (in the blood) Higher Cu levels as a protective factor for AD risk AD, Alzheimer's disease; MR, Mendelian randomization.

Manganese
Mn is a crucial element for protein synthesis, lipid and glucose metabolism, and oxidative stress protection [244]. However, Mn is also an environmental toxin, and elevated Mn levels have been linked to diminished cognitive performance [187,245,246]. A rise in Mn levels has also been observed in patients with AD [109]. Nonetheless, a meta-analysis by Du et al. [89] revealed a significant decrease in Mn levels between AD and the controls.

Magnesium
Human studies have demonstrated that Mg deficiency impairs memory [247] and that Mg supplementation can improve memory in dementia patients [248][249][250]. In addition, a decrease in Mg concentration has been observed in the tissues of AD patients [251,252]. However, no change in Mg concentration was observed in the brains of AD patients in some studies (reviewed in [253]). Mg influences the processing and transport of APP, with low Mg levels favoring the β-secretase pathway and high Mg levels favoring the α-secretase pathway [254], whereas the treatment of experimental animals with Mg sulfate reduces tau phosphorylation and influences the maintenance of cognitive functions and synaptic plasticity [255]. According to the meta-analysis by Du et al. [89], the serum and plasma Mg concentrations were lower in the AD patients than in the controls, whereas the CSF Mg concentrations did not differ between groups. Thomassen et al. [95] did not find an association between the plasma Mg levels and the risk of AD in a study involving more than 100,000 participants. Kieboom et al. demonstrated that both low and high Mg concentrations were associated with an increased risk of dementia. They concluded that the relationship between Mg and the risk of dementia was U-shaped rather than linear [108].

Other Essential Metals
AD also perturbs the homeostasis of Na, K, and Co. Previous studies have associated elevated Na levels with AD [27,96,256,257]. Both increased [102] and decreased [195] K levels have been associated with AD, whereas in some studies, no change in the K levels was observed in AD. Co is an essential component of vitamin B 12 and is an environmental toxin. Zheng et al. showed that mice exposed to Co develop age-related neurodegeneration [258].

Treatment of Alzheimer's Disease Based on the Metal Hypothesis
So far, therapeutic interventions based on the metal hypothesis of AD have progressed in two directions. Taking metal supplements is one approach. As previously stated, there have been several clinical studies on the effect of Zn and Cu supplementation on cognitive performance [259]. Although it was previously thought that metal supplementation could delay the onset of dementia, the majority of studies have found no significant effect on cognitive function improvement [259]. Cu supplementation had no positive effects in a pilot phase 2 clinical trial in AD patients [260]. In contrast to this viewpoint, there is a theory about the therapeutic effect of chelating excess Zn, Cu, or Fe metals based on their ability to stimulate Aβ aggregation [261]. Chelators are substances that bind metals so that they cannot interact with the Aβ further. Based on promising results from in vivo studies on animal models [262], one of the compounds with chelating activity, clioquinol (CQ), has entered the second phase of clinical trials [263]. CQ can effectively bind Zn 2+ and Cu 2+ ions, and in vitro studies have shown that its effect can be achieved by stimulating neuroprotective signaling pathways by increasing the cellular uptake of Zn and Cu ions [264]. Furthermore, CQ can reduce the number of Aβ aggregates within the cells [264]. A clinical study found that the CQ had a promising effect on cognitive improvement, but only in patients with the worst starting point [263]. PBT-2, another similar substance, had a similar effect on AD patients. In a 2008 clinical trial, PBT-2 demonstrated an impact on Aβ metabolism by lowering the Aβ CSF concentrations, and a group of patients who received a higher dose of PBT-2 had better results in two of the executive functions tests, while other cognitive tests revealed no significant differences [265]. A recent study investigated the different capacities for chelating Cu from Cu(II)Aβ  complexes, and the results showed that CQ and B2Q were more efficient than PBT-2 [266], but it must be noted that when using such substances, they must be not too effective as chelators to disrupt the normal metal functions in the brain. Another recent in vivo study examining the effect of Cu ions on tau protein pathological changes found that lowering the Cu concentrations in the brain could help alleviate spatial memory deficits, but, neither lowering nor increasing the amounts of Cu affected the tau protein pathology [267]. A recent review study called the chelating theory into question on multiple levels [268]. As a result, there is no single metal-target approach for treating AD, and more research is needed to better understand this complex aspect of the disease.

Conclusions
In this review, we discussed the studies examining the role of essential metals and heavy metals in AD. Relevant studies involving human subjects were included. It is still challenging to paint a complete picture of how metals interact in AD pathogenesis because of the significant degree of variability in the results between studies. Cu and Zn showed the most consistent results, with most studies revealing that the AD patients' Cu levels rose while their Zn levels fell (Table 1). However, several studies also failed to find such a link (Table 1). Comparing the metal levels with biomarkers from the AD subjects' CSF has rarely been undertaken, and the results were sometimes contradictory ( Table 2). The use of various methodologies to determine the metal levels and examine various body fluids may be the cause of the studies' contradictory findings. Atomic absorption spectrophotometry (AAS) was the second most often employed method, while inductively coupled plasma mass spectrometry (ICP-MS) was the method of choice, being the method utilized in the majority of investigations into the metal measurements. Several MR studies have investigated the relationship between metals and the risk of AD (Table 3). Since MR is revolutionizing epidemiologic research [269], and given the importance of elucidating the role of metals in AD pathogenesis, additional MR studies examining the causal association between metals and AD risk and including people from various ethnic backgrounds are crucial. Funding: This work was funded by The Croatian Science Foundation grants IP-2019-04-3584 ("Role of the blood-brain barrier, innate immunity, and tau protein oligomerization in the pathogenesis of Alzheimer's disease") and IP-2014-09-9730 ("Tau protein hyperphosphorylation, aggregation, and trans-synaptic transfer in Alzheimer's disease: cerebrospinal fluid analysis and assessment of potential neuroprotective compounds") to GŠ, and by the Scientific Center of Excellence for Basic, Clinical, and Translational Neuroscience CoRE-NEURO ("Experimental and clinical research of hypoxic-ischemic damage in perinatal and adult brain"; GA KK01.1.1.01.0007 funded by the European Union through the European Regional Development Fund), and in part by the NIH grant P30 AG066514 to PRH.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.

Conflicts of Interest:
The authors declare no conflict of interest.