Aluminium and lead: molecular mechanisms of brain toxicity
The fact that aluminium (Al) and lead (Pb) are both toxic metals to living organisms, including human beings, was discovered a long time ago. Even when Al and Pb can reach and accumulate in almost every organ in the human body, the central nervous system is a particular target of the deleterious effects of both metals. Select human population can be at risk of Al neurotoxicity, and Al is proposed to be involved in the etiology of neurodegenerative diseases. Pb is a widespread environmental hazard, and the neurotoxic effects of Pb are a major public health concern. In spite of the numerous efforts and the accumulating evidence in this area of research, the mechanisms of Al and Pb neurotoxicity are still not completely elucidated. This review will particularly address the involvement of oxidative stress, membrane biophysics alterations, deregulation of cell signaling, and the impairment of neurotransmission as key aspects involved Al and Pb neurotoxicity.
KeywordsAluminium Lead Neurotransmission Oxidative stress Neurotoxicity Cell signaling Calcium Toxicology
Since the last decades of the nineteenth century and until today, the study of the consequences of human exposure to heavy metals has risen as a central research area in the toxicological field. Among the group of metals with proven human toxicity aluminium (Al) and lead (Pb) are known to be highly neurotoxic. The systemic absorption of Al is low, but select human subpopulations (infants, individuals with altered renal function) can be more susceptible to Al neurotoxicity. Although still controversial, Al is proposed to be involved in the pathophysiology of neurodegenerative disorders (Parkinsonism dementia, Alzheimer’s disease). On the other hand, Pb is a major worldwide public health concern, given the high levels of environmental contamination and the severe and long term neurotoxic effects of Pb.
Several mechanisms have been proposed to underlie the neurotoxicity of Al and Pb. This review will focus on the involvement of oxidative stress, membrane biophysics alterations, deregulation of cell signaling, and impairment of neurotransmission, as key aspects of Al and Pb neurotoxic effects.
Al is one of the most abundant metals of the Earth’s crust, representing approximately 8% of total mineral components. In nature, Al is present as a trivalent cation (Al3+), most of it being associated with silicate and forming water-insoluble complexes. Through the formation of these complexes, Al bioavailability is highly reduced. From a chemical point of view, Al belongs to the group III of the elements. Its relatively small size (atomic radii 0.51 Å), together with its trivalent positive charge, confers Al a high-polarizing effect on its neighbor atoms. This cation has a propensity to form hydroxyl complexes in water solution which, due to their amphoteric character, evolve from free Al3+ towards Al(OH)4− within the 3–8 pH range. At physiological pH, Al forms the scarcely soluble Al(OH)3, that can be easily dissolved by slight changes in the acidity of media. Al can also bind to oxygen- and nitrogen-containing compounds, particularly to inorganic and organic phosphates. Through these kinds of interactions, Al binds to many biological macromolecules (Ganrot 1986).
Until today, no biological function has been assigned to this metal. On the contrary, Al accumulation in tissues and organs results in their dysfunction and toxicity, effects that usually correlate with the local concentration of the metal.
Human exposure to Al is mainly caused by environmental factors, such as soil contamination. Al can be mobilized as a consequence of soil acidification due to the use of certain fertilizers or acid rain. Human Al exposure could also be a result of its use in some industrial processes such as metallurgy, food preservation (Yokel et al. 2008), water purification (Krewski et al. 2007), and the use of a number of pharmacological and cosmetic products (Ganrot 1986). Main sites of Al absorption are the gastrointestinal tract (Ittel 1993), the skin (Exley 1998, 2004a), and the olfactory and oral epithelia (Roberts 1986). Interestingly, since the intranasal epithelium is a portion of the nervous system in close contact with the external milieu, Al could be rapidly absorbed and distributed into the brain through axonal transport (Perl and Good 1987; Zatta et al. 1993).
Once absorbed, Al reaches the blood and circulates mainly bound to transferrin and citrate (Yokel et al. 2002). Tissue and organ Al uptake can occur via the binding of the Al-transferrin complex to the transferrin receptor, following the same internalization pathway than iron. Al crosses the blood-brain barrier (Yokel 2002), and accumulates into glial and neuronal cells (Golub et al. 1999; Levesque et al. 2000; Aremu and Meshitsuka 2005). In order to avoid Al deposition in the brain, the blood-brain barrier has an active efflux of this cation through a monocarboxylate transporter (Yokel et al. 2001). However, this system can be overcome by an increase in blood Al concentration. Al brain concentrations should be lower than 2 µg g−1 (Andrasi et al. 2005). A tenfold increase in Al concentrations was reported in patients intoxicated with Al through the use of hemodialysis solutions containing high levels of Al (Alfrey et al. 1976). Although not fully proven, Al accumulation in the brain is proposed to be associated with neurodegenerative diseases, including Alzheimer’s dementia, Parkinson’s disease, amyotrophic lateral sclerosis, and dialysis encephalopathy (for a review see Gonçalves and Silva (2007), and references therein).
Given the variety of biomolecules able to bind Al, and the capacity of Al to displace other biological cations (such as calcium and magnesium) from their binding sites, almost every metabolic pathway is a potential target for the adverse effects of Al. Therefore, Al neurotoxicity is not caused by a single alteration, but it is probably a result of adverse effects at multiple cellular levels.
Aluminium and oxidative stress
Al has no redox capacity in biological systems. However, extensive experimental evidence demonstrates both, in vitro and in vivo, that high Al concentrations cause oxidative stress. This condition was defined by Sies and Jones as “an imbalance between oxidants and antioxidants, in favor of the oxidants, leading to a disruption of redox signaling and control and/or molecular damage” (Sies and Jones 2007). The nervous system is particularly sensitive to oxidant-mediated damage because of: (a) its high oxygen consumption rate (approximately 20% of total oxygen consumed), (b) brain membranes are enriched in highly oxidizable polyunsaturated fatty acids, (c) brain antioxidant enzymes (catalase, superoxide dismutase, and gluthatione peroxidase) activities are comparatively lower than those found in other tissues, and (d) the brain content of iron is high (Youdim 1988). The latter is particularly important because iron is a redox-active metal, that can interact with molecular oxygen to generate superoxide anion (O2·−), which in turn, generates hydroxyl radical (·OH), a highly reactive oxygen specie (ROS). As a consequence of Fe2+ reaction with O2, Fe3+ is generated, that can trigger lipid oxidation through its reaction with lipid hydroperoxides normally present in biological systems. Finally, many neurotransmitters are auto-oxidizable molecules. For example, dopamine and noradrenaline can react with molecular oxygen generating ROS (O2·− and H2O2), and active quinones.
Evidence of an oxidative stress status has been found in association with most neurodegenerative disorders in which Al is present in relative high amounts. These findings led to an extensive investigation on the possible link between Al and the promotion of oxidative stress.
In vitro, Al stimulates iron-initiated lipid oxidation (Gutteridge et al. 1985). This ability of Al is enhanced by the presence of polar head groups with either negative charge (Oteiza 1994) or polihydroxylation (Verstraeten et al. 1998), which allows the binding of Al through electrostatic interactions. As it will be discussed in a following section, Al binding to negatively charged lipids results in their lateral rearrangement with creation of discrete domains with lower fluidity than the original membrane (Verstraeten and Oteiza 1995; Verstraeten et al. 1997b, 1998). Key negatively charged lipids found in biological membranes are phospholipids. Among them phosphatidylserine, phosphatidic acid, and poliphosphoinositides bear an overall negative charge due to the presence of either carboxylate or phosphate moieties, making these phospholipids preferential Al-binding sites in the bilayer. These phospholipids are particularly susceptible to ROS attack, because they contain poly-unsaturated fatty acids. Thus, negatively charged phospholipids clustering, along with a higher packing of their acyl chains, create a favorable scenario for iron-mediated lipid oxidation. This hypothesis was corroborated in vivo, using an animal model of Al exposure during gestation and until postnatal day (PND) 20. In myelin fractions isolated from Al-intoxicated mice, not only Al content was elevated, but membranes were more rigid and contained higher amounts of lipid oxidation products than controls (Verstraeten et al. 1997a). Al can also interact with zwitterionic phospholipids, such as phosphatidylcholine (PC). However, due to a lack of an overall negative charge, free Al binds to a lesser extent to PC than to negatively charged liposomes, only showing an important pro-oxidant capacity when forming part of a lipophilic complex such as Al-acetylacetonate (Ohyashiki et al. 1998).
Neuromelanin is a complex, redox-active, brain pigment mainly found in substantia nigra and locus coeruleus. The deposition of neuromelanin in these brain areas has been strongly associated with neurodegeneration, constituting a hallmark of Parkinson’s disease (Fahn 2003). In addition to melanin deposits, both Al and iron are found in high amounts in brains from patients affected with this pathology. In vitro, Al can facilitate both iron- and copper-supported oxidation of dopamine to melanin (Di and Bi, 2003a; Di and Bi, 2004), and mediate per se the conversion of melanochrome to melanine (Di and Bi, 2003b). Once formed, melanin can bind transition and other heavy metals, such as iron, Al, zinc, chromium, selenium, strontium and cobalt. Al binding to melanin leads to an increased melanin-induced lipid oxidation, in a process mediated by the formation of an Al–O2·− complex (Meglio and Oteiza 1999). In this complex, the negative charge of O2·− is neutralized by the positive charge of Al, increasing the oxidant capacity of O2·− (Exley 2004b). Moreover, in the presence of ascorbic acid, Al binding to 6-hydroxidopamine results in a sustained increase in ·OH due to auto oxidation, an effect that was partially prevented by glutathione and other sulphydryl-containing molecules with metal chelating ability (Mendez-Alvarez et al. 2002).
The above evidence show that although Al is a non redox metal, it can cause oxidative stress through multiple mechanisms.
Aluminium and membrane biophysics
Al, as a trivalent cation, can bind to membrane components and modify membrane physical properties, ultimately affecting membrane-associated processes.
To understand the possible consequences of Al interaction with lipids, different membrane models have been used. Liposomes are useful tools for biophysical studies because they constitute a well-defined and controlled system, in which lipid composition can be defined by the user. Working with phosphatidylcholine and phosphatidylserine liposomes we found that Al affects both inter-vesicle interactions (trans-interactions) and the intrinsic properties of the bilayer (cis-interactions) (Oteiza 1994; Verstraeten 2000). Although liposomes are stable structures that do not spontaneously interact among them (especially those formed by negatively charged phospholipids with high electrostatic repulsion), Al promotes their aggregation and fusion (Verstraeten and Oteiza 1995). Similar to calcium, Al neutralizes superficial charges, promoting the intercrossing of phosphatidylserine molecules located in neighbor vesicles and causes their partial dehydration, an effect that favors membrane fusion (Verstraeten and Oteiza 2000). Al binding to phospholipids within the same vesicle also results in lipid clustering with formation of phosphatidylserine-enriched domains where membrane fluidity is significantly reduced (Verstraeten and Oteiza 2000). These effects of Al were also observed in membranes containing galactocerebrosides or polyphosphoinositides, which favor Al binding through electrostatic interactions. In these membranes, the effects of Al promoting lipid clustering, dehydration and rigidification were significantly higher than in bilayers of PC, a zwitterionic phospholipid (Verstraeten et al. 1998, 2003; Verstraeten and Oteiza 2002). Supporting the relevance of Al ionic interaction with lipids in these processes, Kaneko et al. (2007) recently demonstrated that Al complexation with maltolate have little effect on membrane fluidity. Al promotion of membrane dehydration was reported to be even higher in the presence of Fe2+ (Dousset et al. 1997). However, given that Al enhances Fe2+-mediated lipid oxidation, the observed effects could be a resultant of Al interaction with the bilayer, but also to Fe-induced membrane oxidation with the subsequent loss of poly-unsaturated fatty acids which would make membranes more fluid.
Biological membranes are complex, containing a variety of molecules that could be possible targets of, and/or modulators of Al effects. In non-differentiated human neuroblastoma (IMR-32) cells, Al significantly decreased plasma membrane fluidity, an effect that was enhanced upon cell differentiation to a cortical neuron phenotype (Verstraeten et al. 2002). Metal complexation strongly determines the extent of Al effects on membranes. In this regard, the exposure of SH-SY5Y human neuroblastoma cells to Al–Aβ1–42 (β-amyloid peptide), but not to Aβ1–42, caused a marked increase in membrane fluidity (Drago et al. 2008). In this case, membrane fluidification could be a consequence of Al-supported formation of large aggregates of β-amyloid which could interact with membranes and alter their rheology rather than a direct effect of Al on cell membranes (Drago et al. 2008).
Myelin is also susceptible to the deleterious effects of Al (Verstraeten et al. 1997a; Pandya et al. 2004). Myelin is a peculiar biological membrane, given that around 80% of its mass is composed of lipids, creating an environment favorable for Al effects on membrane physical properties. The fluidity of myelin membranes decreases upon the in vitro incubation with micromolar concentrations of Al (Verstraeten et al. 1997a). In a model of chronic exposure to Al during gestation and until PND 20, we found that brain myelin not only had higher contents of Al in the Al-exposed than in control mice, but also that myelin membranes were more rigid and had major alterations in lipid composition (Verstraeten et al. 1998). Particularly, myelin membranes from Al-intoxicated mice were rich in galactocerebrosides that favor Al binding, and Al-induced lipid clustering and decreased membrane fluidity. On the contrary, in a model of chronic exposure to Al imposed to adult rats, while myelin composition was mildly affected, no significant alterations in membrane fluidity were observed (Pandya et al. 2004). The above findings demonstrate that the deleterious effects of Al depend not only on the dose, but also on the time and the developmental stage at the onset of Al exposure.
Synaptosomal membranes are also affected by Al. Similarly to that found in liposomes, Al caused in vitro a decrease in synaptosomal membrane fluidity (Ohba et al. 1994; Silva et al. 2002). In vivo, the effects of Al on synaptosomal fluidity are contradictory. While Silva et al. (2002) reported that synaptosomal membranes isolated from the brain of rats chronically exposed to Al were more fluid than those isolated from control animals, previous evidence showed a significant decrease in the fluidity of rat brain synaptosomal membranes (Julka and Gill 1996). The disparity between both in vivo studies could be due to differences in Al dose and administration protocols that may differentially affect membrane lipid composition. In fact, Silva et al. (2002) observed a lower ratio of cholesterol to phospholipid contents in Al-exposed than in control animals. In summary, current evidence indicates that Al alters the biophysical properties of myelin and synaptic membranes, effects that could have a negative impact on neurotransmission and on the release and/or uptake of neurotransmitters.
Aluminium and cell signaling
The effects of Al on cell signaling are well characterized in plants. However, there is limited information on how cell signaling could be affected by this metal in the brain.
Several biological molecules, such as neurotransmitters, neuromodulators and hormones exert their physiological actions through an intracellular second messenger system, in which the receptor-ligand complex stimulates polyphosphoinositides turnover. This pathway is inhibited by Al both in vivo and in vitro (McDonald and Mamrack 1988, 1995; Nostrandt et al. 1996). This effect of Al could be attributed to its interaction with one or more of the following components of this signaling cascade: membrane receptors, receptor-associated G protein, the enzyme phosphatidylinositol-specific phospholipase C (PI-PLC), or its substrates. The first two possibilities were dismissed by early works of Shafer and Mundy (1995). As described above, Al preferentially binds to negative charges of polyphosphoinositides, leading to the clustering of these particular lipids. Both, the increase in the local concentration of polyphosphoinositides, as well as the partial neutralization of their negative charge upon Al binding, resulted in an impairment of PI-PLC binding to its substrates, and consequently to a decreased enzyme activity (Verstraeten and Oteiza 2002; Verstraeten et al. 2003). On the contrary, PLC does not seem to be affected by Al probably because its binding to phosphatidylcholine is weak and does not cause major changes in membrane physical properties capable to impair the access of the enzyme to its substrates (Verstraeten et al. 2003). On the other hand, in the absence of transferrin, oligodendrocytes do not accumulate Al and show a high PIP2 metabolism upon exposure to low Al concentrations. This altered PIP2 metabolism could be mediated by the stimulation by Al of G-protein linked cation-sensitive receptor in the oligodendrocyte membrane (Golub et al. 2002). Together, the above results suggest that Al can adversely affect signaling cascades that either involve the binding of regulatory proteins to polyphosphoinositides in the membranes or that involve phosphatidyl inositol-derived second messengers.
Al forms with maltolate a lipophilic complex that is stable at physiological pH, and that has been proposed to be formed in vivo in the gastrointestinal tract. This Al complex is strongly neurotoxic and causes cytoskeletal alterations that lead to tangle formation (Savory et al. 1998), and cell apoptosis (Johnson et al. 2005). Al-maltolate induces apoptosis in N2a neuroblastoma cells through the intrinsic pathway, with an upregulation of transcription factor p53, an increased expression of the proapoptotic protein Bax, and decreased expression of the antiapoptotic protein Bcl-2 (Johnson et al. 2005). Accordingly, a Bax inhibitor significantly prolonged the survival of yeasts exposed to AlCl3 (Zheng et al. 2007). In rabbits, the intracisternal administration of Al induces apoptosis in the hippocampus, which was suppressed by the co-administration of glial cell neuronal-derived factor (Ghribi et al. 2001a, b).
Neurodegenerative diseases are usually accompanied by tissue inflammation. In human glioblastoma cells, Al causes NF-κB activation and increased TNF-α expression (Campbell et al. (2002). TNF-α activates transcription factor NF-κB which promotes the expression of genes involved in inflammation, such as cytokines, iNOS, and complement factors. Therefore, this cycle of cytokine generation and NF-κB activation could lead to cell death, and the proliferation of reactive glial cells increasing tissue damage (Campbell et al. 2002). The activation of mitogen-activated protein kinases (MAPKs) can also play an important role in Al toxicity (Fu et al. 2003). In cultured cortical neurons, Al activates the stress-responsive c-jun N-terminal kinase (SAPK/JNK), a well-known activator of transcription factor AP-1 that participates in the regulation of cell proliferation and apoptosis. Al also up-regulates the expression of hypoxia induced factor (HIF-1) in human neuronal cells (Lukiw et al. 2005), an effect that was also observed in Al-exposed hepatocytes (Mailloux and Appanna 2007). Given that NF-κB, MAPK/AP-1 and HIF-1 are activated by oxidative stress, the observed effects could be caused by either a direct action of Al on gene expression or indirectly, through an increase in cellular oxidants. Furthermore, the stimulation of pro-inflammatory cell signals, together with a decrease of anti-inflammatory molecules, such as the neurotrophins nerve growth factor and brain derived neurotrophic factor (Johnson and Sharma 2003), could contribute to the inflammatory process associated with Al deposition in the brain.
Aluminium and neurotransmission
Al negatively impacts neurotransmission, either by directly inhibiting the enzymes responsible for the synthesis and/or utilization of neurotransmitters, or by affecting the physical properties of synaptic membranes, that could affect the release and/or uptake of these molecules. The impact of Al on neurotransmission has been reviewed extensively by Gonçalves and Silva (2007).
Parkinson’s disease is a neurodegenerative disorder characterized by a dopamine deficiency in the nigro-striatal region of the brain. The finding of high contents of Al in post-mortem brains of Parkinson’s patients triggered a number of studies which focused on characterizing the participation of Al in this pathology. In animal models of Al intoxication, a 40% decrease in striatum dopamine content (Ravi et al. 2000), and an imbalance of dopamine metabolites were observed (Tsunoda and Sharma 1999), which suggested an altered metabolism of this neurotransmitter. In addition, Al inhibits the enzyme dopamine-β-hydroxylase, responsible of dopamine conversion into norepinephrine (Milanese et al. 2001), and promotes α-synuclein aggregation and fibrillation which enhances striatum neurodegeneration (Uversky et al. 2001). Furthermore, a decrease in dopamine D1 and D2 receptors in brain cortex and striatum from Al-intoxicated mice has been recently reported (Kim et al. 2007). Therefore, even though there is no clear evidence supporting Al participation in Parkinson’s disease ethiopathology, this metal could interfere with certain key events of striatum neurotransmission and contribute to neurodegeneration.
Cholinergic and GABAergic metabolisms are also affected by Al. In a model of 4 week administration of Al to rats, a severe impairment of cognitive functions was associated with a decrease in acetylcholine synthesis, degradation, and decreased muscarinic acetylcholine receptors (Julka et al. 1995). Al exposure in rats caused a significant increase in the excitatory aminoacids glutamine and glutamate, accompanied by a decrease in GABA content, suggesting a role for this metal in glutamate-mediated excitotoxic neuronal injury (El-Rahman 2003; Nayak and Chatterjee 2003). Serotonine is a key neurotransmiter involved in eating, sleeping, behavior and neuroendocrine functions. Low serotonine levels have been associated with cholinergic hypofunction. Al intoxication causes a decrease in serotonine levels and its deaminated derivative, 5-hydroxyindole acetic acid in brain cortex and hippocampus (Kumar 2002). These alterations could be due to the loss of cholinergic input which normally inhibits serotonine release (Hortnagl et al. 1987), and/or to the activation of the enzyme monoaminooxidase, responsible for 5-hydroxyindole acetic acid generation, which is activated by Al (Zatta et al. 1998, 1999).
Limited information is available on the effects of Al on neurotransmitter release. Short-term exposure to Al depressed acetylcholine release from differentiated SN56 cells, while long-term incubations doubled the release of this neurotransmitter (Jankowska et al. 2000). In synaptosomes, Al inhibited pyruvate-supported calcium-evoked acetylcholine release and partially reversed the effects of verapamil on the inhibition of acetylcholine release (Bielarczyk et al. 1998). GABA synaptosomal release and uptake is also inhibited by Al as a result of the inhibition of calcium/calmodulin-dependent calcineurin activity (Cordeiro et al. 2003). Given that calcium plays a central role in neurotransmitter release, alterations in calcium metabolism and/or utilization could be a central mechanism in Al neurotoxicity.
Pb is a neurotoxin that continues to be considered a major global environmental health hazard. Pb is found in nature as a divalent cation, mainly forming stable complexes with sulfur. From a chemical perspective, Pb belongs to the group IVa of the elements, it has a relatively large ionic radius (1.2 Å) and high electronegativity (2.33 in the Pauling scale), which favors its interactions with the coordinating groups of proteins (Godwin 2001). Particularly, the ability of Pb to interact in a flexible coordination number with protein oxygen and sulfur atoms, and to form stable complexes with them, increases Pb affinity for proteins with respect to other metals (Garza et al. (2006) and references therein).
Pb is a heavy metal with no known biological function in humans. On the contrary, it can damage various systems of the body including the hematopoietic, renal and skeletal systems with the central nervous system being its primary target (Wilson et al. 2000). The susceptibility to Pb toxicity is influenced by several factors such as environmental exposure, age and nutritional status. Human exposure to Pb occurs via food, water, air and soil. Food and water Pb sources include the use of Pb-containing ceramic dishware, metal plumbing, and food cans that contain Pb solder (White et al. 2007). Another major source of Pb exposure is deteriorated Pb painting in older housing (White et al. 2007). Young children can be easily intoxicated from chronic ingestion of paint chips, house dust or soil containing Pb particles. People can also be exposed to Pb contamination from industrial sources such as smelters and Pb manufacturing industries (Goyer 1996; White et al. 2007). The risk of Pb contamination from motor vehicle exhaust of leaded gasoline has decreased the last couple of decades as a consequence of a gradual reduction of Pb addition to petrol (Strömberg et al. 2008). However, because Pb is a cumulative metal, the latter still remains a major hazard for human health.
Pb toxic effects depend on both, the duration of exposure and the magnitude of the dose. Though, the half life of Pb in blood is only 35 days, in the brain it is about 2 years and in bone it persists for decades. Bone Pb stores may pose a threat to women at reproductive age long after their exposure to Pb has ended. During pregnancy and the postpartum period, Pb is released from bone stores to the blood stream thus increasing blood Pb levels (Gulson et al. 1999). High Pb exposures in pregnant women can cause a low infant birth weight, prematurity, miscarriage, or stillbirth. In addition, when Pb level in maternal blood increases, Pb levels in breast milk also increase, posing an additional risk to the neonate (Li et al. 2000).
Children are particularly sensitive to the deleterious effects of Pb. They can absorb 30–75% of the Pb ingested (Lidsky and Schneider 2003), while adults absorb about 11% of ingested Pb. In addition, pre and peri-natal exposure to Pb result in higher brain metal accumulation than later postnatal exposure due to an under-developed blood–brain barrier in early life. The Center for Disease Control (CDC) established 10 µg/dl blood Pb concentration as the concern limit for childhood Pb intoxication (CDC, 2000). However, recent data suggest that harmful effects on children’s cognition may occur even at blood concentrations below 10 μg/dl (Lanphear et al. 2000; Canfield et al. 2003). Epidemiological studies in children demonstrated that for each increase of 1 μg/ml in blood Pb levels, within the concentration range of 5–35 μg/ml, their IQ is reduced between 2 and 4 points (Klaassen 2001). Consequences of high blood Pb levels in children may be irreversible, including learning disabilities (abstract reasoning, cognitive flexibility, verbal memory, verbal fluency) (White et al. 1993), behavioral problems, and mental retardation. At very high blood concentrations, Pb can cause convulsions, coma and even death.
Nutritional status is another significant risk factor for Pb intoxication and its effects. Iron, zinc and calcium deficiencies increase the retention of ingested Pb, which can also increase Pb gastrointestinal absorption (Goyer 1996; Ruff et al. 1996), and affect the susceptibility to Pb neurotoxicity (Aimo and Oteiza 2006).
The mechanisms underlying Pb neurotoxicity are still a matter of research. So far, the effects of Pb on calcium fluxes and calcium-regulated events have been suggested as major mechanisms involved in Pb toxicity (Bressler et al. 1999; Marchetti 2003; Toscano and Guilarte 2005). Other potential mechanisms for Pb toxicity include the capacity of Pb to affect cell membrane biophysics, cause oxidative stress, and trigger oxidant-sensitive transcription factors.
Lead and oxidative stress
Oxidative stress, oxidative damage to cellular components, and the activation of oxidant-sensitive transcription factor could in part underlie some of the toxic effects of Pb. The deleterious effects of Pb can involve both, ROS and reactive nitrogen species.
Oxidative stress has been associated with Pb exposure in humans and in experimental animal models. In humans occupationally exposed to Pb, biomarkers of oxidative stress such as malondialdehyde, GSH status, glutathione peroxidase and catalase, exceeded the mean value of the control population (Costa et al. 1997; Garcon et al. 2004; Devi et al. 2007). In rats chronically exposed to Pb, higher levels of brain 2-thiobarbituric acid-reactive substances, an indicator of lipid oxidation, and higher activities of the antioxidant enzymes glutathione reductase and glutathione peroxidase were found compared to controls (Adonaylo and Oteiza 1999a). In rats exposed to Pb during gestation and until PND 45, metal accumulation was associated with high levels of lipid oxidation products in different brain regions, such as the parietal cortex, striatum, hippocampus, thalamus, and cerebellum (Villeda-Hernandez et al. 2001). Another gestational study in rats showed not only higher lipid oxidation products, but also higher catalase and superoxide dismutase activities in brain cortex, hippocampus and cerebellum from Pb-exposed rats compared to their corresponding controls at PND 10 and 20 (Bokara et al. 2008).
Different mechanisms could be involved in Pb prooxidant actions (Ahamed and Siddiqui 2007). Pb may induce oxidative stress by inhibiting the enzyme delta-aminolevulinic acid dehydratase, which results in the accumulation of delta-aminolevulinic acid, that can be rapidly oxidized to generate O2·−, ·OH, and H2O2 (Bechara 1996). Pb can stimulate iron-initiated membrane lipid oxidation, by inducing changes in membrane physical properties (Aruoma et al. 1989; Adonaylo and Oteiza 1999b). As discussed in the following section, Pb interacts with negatively charged phospholipids in membranes and, inducing changes in membrane physical properties, it facilitates the propagation of lipid oxidation (Adonaylo and Oteiza 1999b). Pb can also increase the oxidizing potential of oxidant species (Adonaylo and Oteiza 1999a), it can form a Pb2+–O2·− complex with higher oxidizing capacity than O2·− per se (Adonaylo and Oteiza 1999b). Finally, Pb can decrease glutathione levels (the most important non enzymatic antioxidant). Pb binds exclusively to thiol groups, thus, decreasing glutathione’s reductive potency levels and interfering with its antioxidant activity (Gurer and Ercal (2000) and references therein).
The well described interactions of Pb with glutamate could be also involved in Pb-induced oxidative stress. Glutamate increases oxidants production, and the co-exposure of SH-SY5Y neuroblastoma cells to Pb and glutamate markedly increases cell oxidants and decreases GSH concentrations compared to cells incubated only with glutamate (Naarala et al. 1995). It was recently shown that although oxidants production by glutamate does not depend on protein kinase c (PKC), Pb potentiates glutamate-mediated oxidants production in hypothalamic GT1-7 neurons through a mechanism involving PKC activation (Loikkanen et al. 2003). Pb, especially in conjunction with glutamate, can activate PKC through an increase in intracellular calcium or by mimicking calcium (Loikkanen et al. 2003). PKC is a critical enzyme involved in many cell-signaling pathways. PKC activation can trigger the activation of the reduced nicotinamide adenine dinucleotide phosphate (NADPH) oxidase complex (Abramov et al. 2005). It has also been proposed that Pb-induced increase in neuronal calcium levels may cause an excessive calcium influx into mitochondria, resulting in the production of free radicals and in the opening of the membrane transition pores (Sidhu and Nehru 2003).
The interference with nitric oxide production might represent another mechanism accounting for Pb neurotoxicity. Pb affects neuronal nitric oxide synthase (nNOS) in the developing rat brain (Chetty et al. 2001). Rats perinatally exposed to Pb had decreased nNOS protein levels and activity in cerebellum and hippocampus at PND 21 and 35 as compared to their age-matched controls (Chetty et al. 2001). Nitric oxide is primarily produced by nNOS in response to the activation of N-methyl-d-aspartate receptor (NMDAR), a class of glutamatergic receptors involved in brain development and synaptic plasticity. Several studies show that calcium influx through the NMDAR triggers nNOS activation (Garthwaite et al. 1988). Because Pb can compete with calcium at the NMDAR level (Guilarte and McGlothan 2003), it could interfere with nNOS activation and consequent nitric oxide production.
In summary, the exposure to high Pb concentrations can, through multiple mechanisms, affect ROS and reactive nitrogen species production, leading to oxidative stress and/or to the altered modulation of ROS- and reactive nitrogen species-sensitive cellular events.
Lead and membrane biophysics
Even when Pb is cation that can bind to membrane components, the information about Pb effects on membrane rheology is relatively scarce.
Working with liposomes as a membrane model, studies from our laboratory demonstrated that micromolar concentrations of Pb caused similar membrane effects to those described above for Al. Pb promotes lipid rearrangement in the lateral phase of the bilayer (Adonaylo and Oteiza 1999b), effect that relies on the number of negative charges on the membrane surface (Adonaylo and Oteiza 1999b). Similarly to Al, the ability of Pb to promote lipid clustering significantly correlated with its capacity to stimulate iron-initiated lipid oxidation (Adonaylo and Oteiza 1999b). The above evidence supports a general mechanism that explains the capacity of these two non-redox metals, Al and Pb, to induce a lipid environment where iron-initiated lipid oxidation can occur at a higher rate than in native membranes. In this regard, Al, Pb and other trivalent cations (lanthanum, gallium yttrium, scandium, and indium) interact with the phospholipid polar headgroups and, given their capacity to bind more than one phospholipid at a time, promote lipid rearrangement and rigidification (Verstraeten et al. 1997b; Adonaylo and Oteiza 1999b). In this newly generated environment, the propagation of lipid oxidation occurs at higher rates, and the deleterious consequences of this oxidative damage expands farther than the actual metal binding site (Adonaylo and Oteiza 1999b; Verstraeten 2000).
In the absence of oxidants, Pb affects other membrane physical properties. Working with liposomes composed of lipids extracted from myelin membrane and in the absence of proteins, Diaz and Monreal (1995) demonstrated that at micromolar concentrations, Pb can permeate the membrane and disrupt the liposome transmembrane pH gradient, resulting in the alkalinization of the intraliposomal aqueous space. Being a divalent cation, free Pb is unlikely the permeant specie, but the monohydroxylated specie Pb(OH)+, can permeate and neutralize protons in the inner aqueous space, causing liposome alkalinization (Diaz and Monreal 1995). On the other hand, Pb complexation with organic compounds, regardless the stability of Pb–ligand complex, allows metal permeation through membranes (Diaz and Monreal 1995; Zhang et al. 2006; Zhang et al. 2007). These findings introduce an alternative to explain how Pb can access the cell interior, in a non-channel dependent way which will be discussed in the following section.
Pb also affects the physical properties of biological membranes. An early work from Valentino et al. (1982) demonstrated that erythrocytes obtained from twenty workers occupationally exposed to Pb for up to 26 years were significantly more rigid than those obtained from age-matched control individuals. Changes in membrane fluidity were strongly associated with blood Pb concentration (Valentino et al. 1982). In HepG2 cells exposed for 24 h to Pb, plasma membrane fluidity significantly decreased in a concentration-dependent manner (Chen et al. 2002). The opposite effect was observed in rats chronically subjected to low levels of Pb. In this model, Pb accumulated in brain myelin and myelin membrane fluidity was higher than in controls (Dabrowska-Bouta et al. 1999). These conflictive results could be related to the fact that myelin from Pb-intoxicated animals showed gross morphological alterations, and altered composition (higher phosphatidylethanolamine content and altered glycoproteins) when compared to controls (Dabrowska-Bouta et al. 1999, 2008), that can per se affect myelin membrane physical properties.
In summary, even when the amount of information about the impact of Pb on membrane rheology is not abundant, current evidence suggest that by altering membrane fluidity and/or by promoting alterations in membrane permeability, Pb could have a negative impact on membrane structure and functionality, which could be partially responsible for Pb neurotoxicity.
Lead and cell signaling
Pb can alter the modulation of cell signals through different mechanisms including; (a) through its capacity to interact with protein components of the signaling cascade, (b) through changes in the cell redox status or (c) affecting second messengers and/or cell signals (e.g., calcium) that subsequently modulate transcription factors expression and/or activity.
The ability of Pb to interact with oxygen and sulfur, both of critical relevance as part of protein metal binding sites, allows it to substitute for diverse divalent cations, such as calcium and zinc (Godwin 2001). Because Pb bears larger ionic radius and greater electronegativity compared to zinc and calcium, Pb establishes very favorable interactions with the coordinating groups of proteins. In contrast to these metals, the distribution of the electrical charge around Pb is irregular because of the presence of an inert electron pair in its electronic cloud (Garza et al. 2006), that affects the protein structure and/or functionality. It has been observed that Pb has an inhibitory effect on certain zinc binding proteins (Hanas et al. 1999; Huang et al. 2004; Magyar et al. 2005). On the contrary, Pb can cause an abnormal activation of several calcium-binding proteins (Markovac and Goldstein 1988; Kern et al. 2000).
Several studies have shown both, in vitro and in vivo, that Pb exposure stimulates calmodulin and cAMP phosphodiesterase (Habermann et al. 1983; Sandhir and Gill 1994; Gill et al. 2003; Kursula and Majava 2007). Pb also enhances calmodulin-mediated protein phosphorylation in synaptic vesicles (Sandhir and Gill 1994; Gill et al. 2003). It has been suggested that Pb exerts its neurotoxic effects by interfering with calcium/calmodulin-mediated neurotransmitter release that is eventually responsible for Pb-induced behavioral alterations (Gill et al. 2003). Furthermore, calmodulin also participates in the modulation of several ionic channels and cyclic nucleotide-gated channels, such as the cGMP-activated channel (Saimi and Kung 2002). Thus, Pb not only Pb affects cell signaling by replacing calcium in protein binding sites, but also alters calcium cellular concentration by modulating the activity of ion channels.
Similar to calmodulin, PKC is also activated by Pb at even lower concentrations than those required to activate calmodulin (Markovac and Goldstein 1988; Long et al. 1994). By affecting PKC, Pb modifies several signaling cascades, including zif268 and AP-1 (Kim et al. 2000, 2002). These transcription factors modulate the expression of genes such as synapsin I and II, tyrosine hydoxylase, and neural cell adhesion molecule, which are crucial for neuronal development and function. Pb exposure activates transcription factor AP-1 in neuronal cells and in rat hippocampus and cortex. In PC-12 and rat cerebellum granule cells, Pb activates AP-1 and induces transactivation of AP-1-dependent genes (Chakraborti et al. 1999; Ramesh et al. 1999; Kim et al. 2000). The activation of AP-1 in PC-12 cells is mediated by PKC, particularly the alpha and beta isoforms, as evidenced by the inhibition of AP-1 activation by PKC inhibitors (Chakraborti et al. 1999; Kim et al. 2000). High AP-1-DNA binding was found in the hippocampus and cortex of PND 3 pups from dams exposed to Pb from gestation day 13 (Pennypacker et al., 1997). However, AP-1 activation was not observed at PND 9 and 15. AP-1 was activated in frontal cortex, brain stem, striatum, and hippocampus from weanling rats exposed for 90 days to Pb (Ramesh et al. 2001). In IMR-32 neuroblastoma cells, Pb at concentrations as high as 50 μM did not activate AP-1 nor the upstream MAPKs JNK and p-38 after 24 h of incubation (Aimo and Oteiza 2006), although oxidant levels increased in the Pb-treated cells. When IMR-32 cells were incubated in zinc-deficient media, Pb caused a higher increase in oxidant levels and the activation of the MAPK/AP-1 signaling pathway (Aimo and Oteiza 2006).
Less evidence exists on the effect of Pb on the modulation of NF-κB, another important oxidant-responsive transcription factor. The activation of NF-κB was described in PC-12 cells incubated with Pb and in different brain regions from Pb intoxicated rats (Ramesh et al. 1999, 2001). Given the relevance of AP-1 and NF-κB in brain development and function, and the limited and conflicting evidence on their modulation by Pb, further research is necessary to elucidate the role of AP-1 and NF-κB transcription in the pathology of Pb neurotoxicity.
Lead and neurotransmission
Pb is recognized as a risk factor for neurologic and psychiatric disorders. In fact, Pb-induced brain damage occurs preferentially in the prefrontal cerebral cortex, cerebellum and hippocampus. Cognitive functions are localized in cerebral cortex, whereas the cerebellum regulates the execution of motor skills, and the hippocampus, which is the memory storage center, has also been related to behavior. Therefore, these anatomical sites are crucial in modulating emotional response, memory, behavior, learning, and neuromuscular function. Pb causes several morphological alterations in the brain, including cellular damage, encephalopathy and reduced axonal and dendritic development (Alfano et al. 1983). Recent studies have demonstrated that Pb exposure adversely affects a variety of neurotransmitter systems in the developing brain (Guilarte et al. 2000; Leret et al. 2002; Devi et al. 2005; Verina et al. 2007).
It has been extensively demonstrated that rat perinatal exposure to low levels of Pb induces neurochemical changes in the central monoaminergic and aminoacidergic systems of rats at PND 21 (Kala and Jadhav 1995; Leret et al. 2002). Maternal Pb exposure causes a decrease in glutamate content in the cerebral cortex, hippocampus and cerebellum, while GABA levels only decreases in the cerebral cortex (Leret et al. 2002). Pb is a potent non-competitive antagonist of the NMDAR (Engle and Volpe 1990) and it alters NMDAR subunit mRNA levels (Guilarte et al. 2000; Nihei et al. 2000). Since glutamate and its receptors play a critical role in the regulation of cognitive processes, learning and memory, the disruption of the glutamatergic system could explain in part, some of Pb-related neurobehavioral dysfunctions.
Pb-induced effects on monoamine levels in different areas of the brain rely on various aspects, such as the exposure level, the duration of exposure, the animal species used, and animal developmental stage at the onset of Pb exposure. That explains why the evidence on the effects of Pb exposure on catecholamine levels, tyrosine hydroxylase activity, acetylcholinesterase activity, etc., is still controversial. In general, perinatal rat exposure to low levels of Pb causes an increased sensitivity of brain dopamine D2 and D3 receptors (Cory-Slechta et al. 1992; Gedeon et al. 2001), increased hippocampal activity of tyrosine hydroxylase (Bielarczyk et al. 1996), produces higher levels of dopamine (Leret et al. 2002; Devi et al. 2005), and enhances catecholaminergic neurotransmission in cerebral cortex, hippocampus and cerebellum due to increased turnover of norepinephrine (Devi et al. 2005). On the other hand, perinatal exposure to high Pb concentrations decreases norepinephrine, epinephrine and dopamine levels in the cerebral cortex, hippocampus and cerebellum (Dubas and Hrdina 1978; Sidhu and Nehru 2003; Devi et al. 2005), and decreases the activities of acetylcholinesterase (Sidhu and Nehru 2003), monoamino oxidase (Devi et al. 2005), and tyrosine hydroxylase (McIntosh et al. 1989). The observed effects of high Pb levels on brain monoamino oxidase and catecholamines could be indirect, involving an overall metal-mediated cellular damage rather than a direct interaction of Pb with these enzymes and neurotransmitters.
As described above, Pb competes with calcium for common binding sites and is incorporated into calcium transport systems in the nervous system, where calcium is crucial for neurotransmitter release and regulation. Thus, the alteration of calcium homeostasis and the ability of Pb to mimic this cation may also have an important role in Pb-induced neurotransmission alterations. Besides, Pb can induce catecholamine release from its storage vesicles in a calcium-independent manner (Westerink and Vijverberg 2002). Although this latter mechanism is still not fully understood, it involves exocytosis stimulation probably due to a direct stimulation of calcineurin by Pb (Westerink and Vijverberg 2002), an effect that could be related to Pb-mediated enhancement of excitatory post-synaptic currents in rat hippocampal neurons (Braga et al. 1999).
The above evidence indicates that Pb-induced alterations of the major neurotransmitter systems can be a key mechanism underlying Pb-induced neurological alterations.
From the first discoveries during the last decades of the nineteenth century about Al and Pb neurotoxic effects, a large research effort has been made to unravel the molecular mechanisms involved in Al and Pb toxicity. However, there are currently no clear explanations for the neurotoxicity of these metals, and the field is still an incomplete puzzle.
In spite of this, there are several common denominators in Al and Pb toxicities. Among them, their negative impact on cell signaling, neurotransmission, and cell redox status have been the most investigated, and all studies demonstrate a negative impact of these metals on these events that are critical for the nervous system. The effects of heavy metals on membrane rheology, even when profusely documented, are less integrated in the general concept of neurotoxicity. However, it is important to consider that all these events can be interconnected and contribute, as a whole, to the deleterious effects of Al and Pb on the nervous system.
This work has been supported by grants from the University of Buenos Aires, Fundación Antorchas, and CONICET, Argentina, and by grants from the National Institute of Environmental Health Services Center for Environmental Health Sciences, University of California, Davis, and from the University of California, Davis. P.I.O. is a member (Investigador correspondiente) of CONICET, Argentina. S.V.V. is a career investigator of CONICET, Argentina.
- DC C (2000) Blood lead levels in young children—United States and selected states, 1996–1999. MMWR Morb Mortal Wkly Rep 49:1133–1137Google Scholar
- Klaassen CD (2001) Heavy metals and heavy-metal antagonists. In: Hardman JG, Limbird LE, Goodman Gilman A (eds) The pharmacological basis of therapeutics. McGraw Hill, New York, pp 1851–1875Google Scholar
- Magyar JS, Weng TC, Stern CM, Dye DF, Rous BW, Payne JC, Bridgewater BM, Mijovilovich A, Parkin G, Zaleski JM, Penner-Hahn JE, Godwin HA (2005) Reexamination of lead(II) coordination preferences in sulfur-rich sites: implications for a critical mechanism of lead poisoning. J Am Chem Soc 127:9495–9505PubMedGoogle Scholar
- Mendez-Alvarez E, Soto-Otero R, Hermida-Ameijeiras A, Lopez-Real AM, Labandeira-Garcia JL (2002) Effects of aluminum and zinc on the oxidative stress caused by 6-hydroxydopamine autoxidation: relevance for the pathogenesis of Parkinson’s disease. Biochim Biophys Acta 1586:155–168PubMedGoogle Scholar
- Sies H, Jones D (2007) Oxidative Stress. In: Fink G (ed) Encyclopedia of stress. Elsevier, San Diego, pp 45–48Google Scholar
- Verstraeten SV (2000) Mechanisms involved in the oxidative damage to membranes mediated by aluminum and aluminum-related cations. Effects in vitro and in vivo. In: School of Pharmacy and Biochemistry. University of Buenos Aires, Buenos Aires, ArgentinaGoogle Scholar