Manganese Neurotoxicity and Familial Disorders of Manganese Transport

ABSTRACT

Manganese is the 12th most common element in the Earth's crust and is an essential industrial component. Biologically, this metal plays an important role as a constituent of numerous enzymes. While manganese is required for normal biochemical and physiological processes, manganese excess can lead to significant toxicity, particularly to the central nervous system. Over the past 25 years, three inherited disorders of manganese transport have been described, leading to a model of how this essential metal is absorbed, distributed to tissues, and eliminated. The first part of this review describes the clinical features and cellular mechanisms of manganese toxicity due to occupational and environmental exposure. The remainder of the review discusses the familial disorder associated with hypomagnesemia, SLC39A8 deficiency, and the two genetic diseases resulting in hypermanganesemia, SLC30A10 deficiency and SLC39A14 deficiency. These latter two disorders are treatable, and the early recognition and institution of chelation therapy and iron supplementation can improve neurological disability. Ongoing research regarding the tissue‐specific role of each transporter and their regulatory interplay, the varied mechanisms of manganese toxicity, and its treatment are also described.

1. Introduction

Manganese (Mn) is a transition metal that plays many roles in biological systems. Mn‐dependent enzymes include numerous kinases, arginase, pyruvate carboxylase, glutamine synthase, β‐1,4‐galactosyltransferase, and Mn superoxide dismutase (MnSOD). As such, Mn is essential to numerous biochemical processes in intermediary metabolism, neurotransmitter systems, and protein glycosylation and as an antioxidant [1]. Despite being critical to these fundamental reactions, Mn is also highly toxic, particularly to the nervous system. Therefore, a tightly regulated system of Mn absorption and elimination has evolved to prevent overload of this trace metal. Our understanding of the function of the transporters that are responsible for Mn regulation has been enhanced over the past two decades, primarily with the discovery of three rare genetic disorders, two of which result in Mn toxicity [2, 3, 4, 5, 6] and one that causes a systemic Mn deficiency [7, 8]. Following a preliminary section covering the dietary and environmental sources of Mn and the fundamental features of Mn toxicity (manganism), this review will focus on the physiology of Mn uptake and elimination, and the three genetic disorders of Mn transport.

2. Manganese as an Essential Nutrient and a Toxicant

Mn is abundant in the Earth's crust, primarily as inorganic compounds. Due to its atomic structure, it can exist in 11 different oxidation states, but Mn²⁺ and Mn³⁺ are primarily found in organic compounds in plant and animal systems [9]. Mn is quite plentiful in plant‐based foods (nuts, whole grains, legumes) and shellfish; for example, whole grain bread contains 2.5 mg/100 g (around 2 ½ slices) [10]. The recommended dietary allowance for Mn in adults is between 2.5 and 3 mg/day depending on the sex and pregnancy/lactation status [11]; consequently, a dietary Mn deficiency is quite rare. However, out of concern for avoiding any toxicity from dietary Mn, a recent opinion from the European Food Safety Authority has established a safe level of Mn intake of 8 mg/day [12].

For individuals with normal metabolism, the likelihood of developing Mn toxicity through dietary exposure is remote. This is not the case for workers in certain occupations such as battery production, welding, mining, and refining, where there is a risk of inhalational exposure to Mn [9]. After inhalation, Mn may be absorbed via the mucosa in the nasal cavities directly into the bloodstream, and it may also be directly transported from the olfactory bulb to the cerebral cortex. In both circumstances, Mn thereby avoids the regulatory transporters involved in biliary excretion (see below). From the blood, Mn may both diffuse and be transported into the brain, and, given its atomic similarity to iron (Fe), it then accumulates in Fe‐rich brain regions, including the caudate, putamen, globus pallidus, substantia nigra pars reticulata, and subthalamic nucleus [13].

Symptoms of manganism were first described in 1837 in Scottish workers involved in the production of powdered bleach, one step of which required the crushing of Mn ore [14]. The affected workers exhibited various symptoms of parkinsonism, including progressive lower extremity weakness, masked facies, and whispering speech. Various other reports of neurological illness secondary to chronic industrial exposure to Mn appeared during the 19th and 20th centuries, notably the 1955 report describing Mn miners exhibiting progressive extrapyramidal symptoms (including bradykinesia and rigidity), together with aggression, mental excitement, and incoherent speech [15]. It is now well accepted that while the syndrome of manganism shares similarities with Parkinson disease, there are certain significant differences. Specifically, patients with manganism have no tremor but do exhibit dystonia, a “cock‐gait” walk, and, importantly, minimal or no response to levodopa [16]. A cohort of workers in a ferromanganese alloy plant who developed manganism in 1986 have been longitudinally followed for 18 years. Despite a cessation of exposure, their symptoms, which did not have a sustained response to levodopa, were progressive over the first decade and then plateaued [17, 18]. Unfortunately, there are no agreed‐on standards for biomonitoring of industrial Mn exposure. Mn measurement in whole blood and in toenails appear to be the most useful [16, 19].

Fortunately, during current times, Mn toxicity from industrial exposure is quite uncommon, but a similar syndrome secondary to the intravenous abuse of the amphetamine‐like substance methcathinone (ephedrone) has been described in several individuals in Eastern Europe. This designer drug is produced by oxidizing ephedrine or pseudoephedrine with potassium permanganate, and the end product contains significant residual Mn [20]. Similar but generally less clinically overt neurological impairment due to Mn toxicity has been described in patients with chronic liver disease associated with decreased biliary excretion (acquired hepatocerebral degeneration) [21], and due to iatrogenic Mn exposure from long‐term use of total parenteral nutrition [22]. In all of the above forms of clinical Mn neurotoxicity, evidence of Mn brain deposition, primarily in the globus pallidus, can be demonstrated as T1 hyperintensities on magnetic resonance imaging (MRI) [23, 24]. Lastly, as Mn can be both an airborne and drinking water contaminant, chronic exposure by these two routes may be associated with an increased risk of neurodevelopmental impairment such as autism spectrum disorder and attention‐deficit/hyperactivity disorder [25, 26]. For example, several studies from Bangladesh have shown dose‐dependent effects of Mn exposure from well water on the intellectual performance of children [27, 28]. Environmental Mn exposure has also been suggested as a potential pathogenic factor for Alzheimer disease [29].

A variety of mechanisms have been demonstrated to be responsible for the neurotoxic effects of Mn. It is paradoxical that while Mn is necessary for the normal function of the antioxidant MnSOD, Mn itself is a powerful oxidant. Mn‐induced oxidative stress secondary to disturbed mitochondrial respiration, as well as the production of free radicals from the oxidation of dopamine, are considered to be primary mechanisms of Mn toxicity [30, 31]. Mn exposure has been shown to result in epigenetic modification of the expression of several antioxidant genes, including MnSOD [32]. Another toxic mechanism includes the uptake of Mn by astroglia resulting in neuroinflammatory processes [33].

Given the profound extrapyramidal features of manganism, it is not surprising that Mn‐induced disruption of several neurotransmitter systems has been demonstrated. The effects of Mn on dopaminergic neurotransmission have been studied extensively via positron emission tomography (PET) in patients with manganism and via PET and dopaminergic‐specific immunolabeling techniques in various animal models. While some inconsistencies exist between results from clinical studies and different animal models, it is generally agreed that Mn exposure results in decreased dopamine release in the striatum without a loss of dopaminergic neurons in the substantia nigra [34]. These findings may help to explain the reduced or absent clinical response to levodopa. A variety of effects of Mn on the function and regulation of central nervous system cholinergic, GABAergic, and glutaminergic systems have also been demonstrated [13].

3. Regulation of Mn Homeostasis

Several proteins are responsible for transporting Mn across cell membranes. While the organ‐ and cell‐specific rolls of each transporter in regulating Mn levels are not fully elucidated, an overall model of Mn homeostasis with respect to the genetic disorders of Mn transport has come together. Many of these proteins are not specific for transporting Mn and in certain circumstances will also transport other metal ions, including Fe, zinc (Zn), and cadmium. A fundamental aspect of systemic Mn homeostasis is the close relationship between body Fe and Mn stores [35]. When Fe stores (measured as ferritin) are reduced, Mn is accumulated, whereas when Fe levels are high, Mn efflux is increased [36].

After inhalational or intestinal absorption via a variety of transport mechanisms, systemic Mn levels are regulated through the interplay of several transporters, including three solute carrier (SLC) proteins present in bipolar cells of the liver, intestine, and kidney, as well as in the brain and other tissues (Figure 1). Specifically, SLC30A10, also known as Zn transporter 10 (ZnT‐10), is located on the apical surface of hepatocytes and enterocytes. Originally characterized as a Zn efflux transporter [37], it is now recognized that SLC30A10 specifically functions to transport Mn from hepatocytes into the bile and from enterocytes into the intestinal lumen for excretion [13]. SLC39A8, also known as Zrt‐ and Irt‐like protein 8 (ZIP8), is located on the apical membrane of polarized hepatocytes, intestinal epithelial cells, and renal tubule cells and mediates Mn influx from the bile, intestinal lumen, and proximal renal tubule, respectively [13]. Conversely, SLC39A14 (ZIP14) is a Mn influx transporter located on the basal‐lateral membrane of hepatocytes and enterocytes and transports Mn from the blood into those cells for subsequent excretion by SLC30A10 [13, 38]. Lastly, a fourth SLC protein, the Fe transporter ferroportin (Fpn), or SLC40A1, the product of the FPN1 gene, is expressed in hepatocytes, as well as other cell types, and also plays a role in Mn removal into the extracellular space [38, 39]. While not the primary topic of this review, it should be stated that pathogenic variants in FPN1 result in autosomal‐dominant hemochromatosis type 4 (ferroportin disease) [40].

Figure 1.

Proposed model for the manganese (Mn) homeostasis by transporters SLC30A10, SLC39A8, and SLC39A14. SLC30A10 is a Mn efflux transporter localized at the cell membrane. This transporter removes excess Mn from neurons and facilitates excretion from hepatocytes to the bile and from enterocytes to the intestinal lumen. SLC39A8 localizes to the apical domain of enterocytes and hepatocytes, and also the luminal side of proximal tubule cells. SLC39A8 mediates intestinal uptake of Mn into enterocytes and reuptake of Mn from the bile into hepatocytes and from urine into proximal tubule cells. SLC39A14 localizes to the basolateral aspect of hepatocytes and enterocytes and mediates influx of Mn from the blood into hepatocytes and enterocytes, followed by excretion through SLC30A10. The locations and functions of other transporters such as ferroportin and the divalent metal transporter 1 are not shown. Figure created with BioRender.com.

Mn is carried in the bloodstream as free Mn²⁺ and Mn³⁺ ions or bound to transferrin or citrate [39]. Upon reaching the brain, several mechanisms allow Mn to cross the blood–brain barrier, including the citrate transporter, transferrin receptor, calcium channels, Fpn, and SLC39A8 [38, 41, 42]. Fpn localizes to the abluminal side of brain capillary cells and likely serves to export both Fe and Mn from blood into brain parenchym [43], while SLC39A8 has also been demonstrated in brain microvasculature tissue [44]. The role of the choroid plexus in brain Mn transport and homeostasis is complex and likely involves a combination of diffusion and transport mechanisms. Via in vitro studies, the Mn importers SLC39A8 and SLC30A14 have been localized to the apical membrane and basolateral membrane, respectively, of choroid plexus papilloma cells [44, 45]. The dual regulatory roles of these two transporters within the blood cerebrospinal fluid (CSF) barrier have not been determined. Once Mn enters the brain parenchyma, its entry into neurons and glia is facilitated by the transferrin receptor and another SLC, the divalent metal transporter 1 (DMT1), also known as SLC11A2. In particular, DMT1 is highly expressed in several basal ganglia nuclei, and this localization accounts, in part, to preferential accumulation of Mn to these brain regions in patients affected by manganism [38]. The removal of excess intracellular Mn from brain tissue is accomplished by SLC30A10 and Fpn. The Mn exporter SLC30A10 is present in a variety of CNS cells, in particular neurons in the globus pallidus and spinal cord, and in brain microvascular tissue [5, 44, 46]. Importantly, SLC30A10 has been shown to be neuroprotective in animal models of Mn neurotoxicity [47, 48]. Fpn has been demonstrated in a wide variety of cell types in the brain, including neurons, blood‐brain barrier endothelial cells, and cells within the choroid plexus and ependyma. While patients with ferroportin disease do not demonstrate any neurological symptoms, cell culture studies have demonstrated that pathogenic FPN1 variants result in altered intracellular Mn levels and evidence of Mn cytotoxicity. Therefore, the exact role of Fpn in regulating Mn, as well as Fe, within the brain is not established [40].

4. Neurogenetic Disorders of Mn Transport

Between 2000 and 2016, three disorders of Mn transport with profound neurological impairment were clinically described and shown to be due to biallelic pathogenic variants of the three Mn transporters SLC39A8, SLC30A10, and SLC39A14 [2, 3, 4, 5, 6, 7, 8]. As described above, being a Mn importer in the gut, kidney, and liver, SLC39A8 is, in part, responsible for increasing systemic Mn levels; its dysfunction results in hypomanganesemia [7, 8]. On the other hand, SLC30A10 is a Mn exporter, and together with SLC39A14, which imports Mn into hepatocytes for biliary excretion by SLC30A10, these two proteins function to decrease systemic Mn levels; dysfunction of either of these transporters results in hypermanganesemia [4, 5, 6]. All three of these disorders are uncommon, but early recognition of the two disorders resulting in hypermanganesemia (SLC30A10 deficiency and SLC39A14 deficiency) is critical, as treatments for these disorders can result in significant improvement of neurological disability, which includes childhood or adolescent onset of dystonia and early onset of parkinsonism. Consensus guidelines for the diagnosis and treatment of patients with SLC30A10 deficiency and SLC39A14 deficiency have recently been published [49].

5. SLC39A8 Deficiency; Congenital Disorder of Glycosylation Type IIn (OMIM 616721)

In 2015, two groups of investigators identified a multisystem disorder that resulted in dysmorphic features, neurodevelopmental disability, and hypomagnesemia [7, 8]. This condition was first described in seven German, Egyptian, and Hutterite families and was due to biallelic pathogenic variants in SLC38A8. A total of 16 patients from 13 families with this disorder have been reported [7, 8, 50, 51]. These patients have low to undetectable blood levels of Mn, and some also demonstrate elevated Mn in the urine consistent with renal wasting of the trace metal. Clinically, these children are hypotonic and have growth retardation, intellectual disability, strabismus, inability to ambulate, and increased susceptibility to infections. Some affected patients have had dysmorphic craniofacial features and seizures, including infantile spasms. Brain MRI has demonstrated cortical and cerebellar vermal atrophy, with some patients showing increased T2 signal in the basal ganglia that at first was believed to be consistent with Leigh syndrome (Figure 2). While reduced Mn levels likely result in dysfunction of several enzymes, these patients have abnormal patterns of transferrin isoelectric focusing, consistent with a congenital disorder of glycosylation type II, likely due to reduced activity of the Mn‐dependent enzyme β‐1,4‐galactosyltransferese. Dysfunction of mitochondrial MnSOD may underlie the Leigh syndrome–like features in some patients [51].

Figure 2.

Brain magnetic resonance imaging from a 9‐year‐old boy with biallelic pathogenic variants of SLC39A8. (A) Sagittal and (B) coronal images demonstrate a small cerebellum with wide interfoliate sulci and major fissures demonstrating diffuse cerebellar vermian and hemispheric atrophy. The posterior fossa is normal in size. Reproduced from Boycott et al. [7] with permission from Elsevier.

A few animal studies evaluating loss of SLC39A8 activity have been conducted. A study of both Slc39a8 global‐ and liver‐specific knockout mice demonstrated blood and organ tissue reductions in Mn together with decreased protein N‐glycosylation and reduced activity of Mn‐dependent arginase [52]. A second study of Slc39a8 liver‐specific knockout mice that were also fed a Mn‐deficient diet to increase the effects of systemic Mn depletion also reported abnormal transferrin glycosylation profiles. This study also demonstrated an abnormal balance between CSF glutamate and glutamine levels, consistent with reduced Mn‐dependent glutamine synthase activity, together with abnormal open field motor activity [53].

Targeted therapy for this disorder has been reported in only a small number of patients. Treatment with galactose (and in some patients with uridine) has resulted in improvement of transferrin glycosylation patterns; Mn supplementation has also been added to the regimen in a few patients [8, 51, 54]. This combined approach led to some improvement in motor performance, vision, and hearing together with an improvement in the electroencephalogram and control of seizures.

6. SLC30A10 Deficiency; Hypermanganesemia With Dystonia 1 (OMIM 613280)

The first patients with a hereditary form of hypermanganesemia were described in 2000 [2] and 2008 [3], and the genetic etiology of this multisystemic disorder was discovered by two research groups in 2012 [4, 5]. Originally identified as the “syndrome of hepatic cirrhosis, dystonia, polycythemia, and hypermanganesemia,” this condition is now termed "hypermanganesemia with dystonia 1" (HMNDYT1) and is due to biallelic pathogenic variants in SLC30A10. A total of 60 patients from 37 families have been reported [49]. While some phenotypic variability has been documented, most of these patients present with childhood onset of progressive dystonia of the limbs and tongue and exhibit a distinctive “cock‐walk” gait, decreased manual dexterity, and dysarthria. Together with these neurological symptoms, they have polycythemia that may require treatment with phlebotomy, and progressive liver disease. In many instances, the polycythemia is associated with elevated erythropoietin and low levels of ferritin, indicative of reduced Fe stores [2, 4, 46, 55]. The liver disease is variable in presentation and progression and is associated with elevated transaminase levels, hepatomegaly, and signs of cirrhosis, and in some instances it has been fatal. Blood Mn levels are extraordinarily elevated (up to 20 times normal), and Mn deposition in the globus pallidus, striatum, pituitary, brainstem, and dentate nucleus of the cerebellum is visualized in MRI (Figure 3). An important contrast with the neuropsychiatric features of manganism is that behavioral and cognitive symptoms are uncommon in patients with HMNDYT1 [2, 3, 4, 5, 55, 56, 57].

Figure 3.

Brain magnetic resonance imaging from a 19‐year‐old man with biallelic pathogenic variants in SLC30A10. (A) Axial T1‐weighted image shows increased signal in the lenticulate nuclei (arrowheads), greatest in the globi pallidi. Tiny low signal areas within the nuclei represent perivascular spaces of the lenticulostriate arteries. (B) Axial T1‐weighted image shows high signal in the pituitary gland (white arrow) and in the brainstem (black arrows). Brainstem atrophy is present. (C) Sagittal T1‐weighted image demonstrates high signal in the pituitary gland (white arrow), anterior commissure region, brainstem, and cerebellum (white arrows heads). (A and B) reproduced from Gospe et al. [2] with permission from BMJ Publishing Group Ltd., (C) reproduced from Caruso et al. [24] with permission from Elsevier.

A small number of patients presented in adulthood with levodopa‐resistant parkinsonian features, including rigidity, bradykinesia, hypomimia, and hypophonia [5]. Another patient presented with adolescent‐onset spastic paraparesis but no evidence of dystonia or other extrapyramidal symptoms [2, 46]. Epilepsy was described in a 5‐year‐old child [58], and sensory motor neuropathy was reported in two patients with adult‐onset parkinsonism [5]. Lastly, two children presented with isolated polycythemia, one of whom had subtle dystonia after diagnostic testing revealed hypermanganesemia and brain MRI abnormalities [59, 60].

Absence of functional SLC30A10 underlies the pathophysiology of HMNDYT1. As this efflux transporter is responsible for trafficking Mn both from hepatocytes into the bile and from enterocytes into the intestinal lumen for excretion, deficient SLC30A10 results in hepatic Mn excess and ineffective systemic Mn clearance. This leads to Mn hepatotoxicity, hypermanganesemia, and Mn deposition in the brain with concomitant neurotoxicity. Absence of SLC30A10 from neuronal tissue likely also contributes to the pathophysiology. Cell‐based studies have shown that pathogenic variants in SLC30A10 result in a block in the trafficking of the transporter to the cell membrane and the accumulation of Mn‐containing vesicles within the Golgi apparatus [61]. A study of Slc30A10 knockout mice where the transporter was selectively deficient in either hepatocytes, enterocytes, or whole body demonstrated that mice with global knockout had the greatest evidence of Mn excess. Mice with Slc30a10 knocked out in either hepatocytes or enterocytes had minimal evidence of Mn excess, while those with the gene knocked out in both tissues had a midrange of excess Mn measured. These results suggest that while Slc30A10 function in hepatocytes and enterocytes are essential for Mn homeostasis, there are additional sites where Slc30a10 is expressed that are important for overall Mn elimination [62]. Another study in Slc30a10‐deficient mice showed that the associated polycythemia and Fe depletion are due to stimulation by excess hepatic Mn of hypoxia inducible factor (HIF) 2, a transcription factor that results in increased production of hepatic erythropoietin [63].

A few pathology studies have been performed in patients with HMNDYT1. Liver biopsy has demonstrated micronodular cirrhosis and mononuclear cell infiltration with patchy rhodamine‐positive hepatocytes consistent with Mn deposition [2, 3, 5, 64]. An autopsy was performed on a 38‐year‐old man who died of pneumonia [46]. The liver demonstrated similar findings together with a ninefold increase in Mn concentration and decreased SLC30A10 by Western blot analysis compared with control tissue. The autopsy also revealed mottling of the basal ganglia, primarily the globus pallidus, with severe neuronal loss together with rhodamine‐positive astrocytes, microglia, and residual neurons. Reduced SLC30A10 immunoreactivity was present in the residual neurons of the globus pallidus, and Mn content was increased 16‐fold. Several other brain regions were also involved, including the corticospinal tracts, which may explain the spastic paraparesis that was his primary motor symptom [2].

Treatment of HMNDYT1 with a combination of chelation with intravenous disodium calcium edetate (Na₂CaEDTA), to mobilize systemic Mn for renal excretion, and Fe supplementation, to competitively inhibit Mn influx transport, has proven to be effective in reducing dystonia, reducing brain deposition of Mn (Figure 4), and improving liver function, particularly if started early in the course of the disease [3, 4, 49, 55, 65]. However, chelation therapy is quite burdensome, and the necessary supplies for this long‐term treatment may not be readily available in some locales.

Figure 4.

Serial magnetic resonance imaging (MRI) brain over a 10‐year follow‐up of a 21‐year‐old woman with SLC30A10 deficiency treated with Na₂CaEDTA. (A) T1‐sequences at the age of 12 years, before treatment. (B) T1‐sequences at the age of 15 years, 3 years after treatment, slightly improved, compared with (A). (C) T1‐sequences at the age of 16 years, 4 years under treatment, 1 year after increasing the frequency of infusions, and 1 month after adding oral ferrous iron; hyperintensities are less pronounced than before. (D) T1‐sequences at the age of 20 years, 1 year after reducing chelation therapy because of a lack of supplies. There is reaccumulation of manganese compared with MRI 4 years earlier (C). (E) T1‐sequences at the age of 21 years, 1 year after increasing the frequency of infusions, overall 9 years after onset of treatment; hyperintensities have clearly diminished. Reproduced from Stamelou et al. [65] with permission from Wiley.

7. SLC39A14 Deficiency; Hypermanganesemia With Dystonia 2 (OMIM 617013)

Four years after the publication of the studies demonstrating SLC30A10 deficiency as the cause of HMNDYT1, additional patients with a somewhat similar neurological presentation and pattern of Mn deposition on brain MRI scan were reported to have biallelic pathogenic variants of SLC39A14 [6]. A total of 33 patients from 26 families with hypermanganesemia with dystonia 2 (HMNDYT2) have been reported [49]. In most patients, the neurological symptoms presented between infancy and 3 years of age as either a delay or loss of motor milestones, together with long tract and extrapyramidal signs, bulbar dysfunction, and axial hypotonia. Over the first decade of life, levodopa‐resistant parkinsonian features developed, including dystonia, tremor, bradykinesia, and hypomimia. The motor symptoms in these patients were profoundly progressive, with dystonic posturing leading to contractures, scoliosis, and loss of ambulation. Importantly, HMNDYT2 is purely a neurological disorder; there is no evidence of either hepatic involvement or polycythemia. As with patients with HMNDYT1, these individuals also have striking elevations of blood Mn and MRI evidence of brain Mn deposition [6, 55, 66]. In addition, Mn concentration in the CSF is threefold higher than the elevated levels in the blood [66]. A substantially milder phenotype appearing in late adolescence with minimal progression through 65 years of age has been described in an individual [67].

The pathophysiology of HMNDYT2 is explained by the absence of functional SLC39A14 protein. This results in decreased transport of Mn into hepatocytes and enterocytes for excretion by SLC30A10 into the bile and intestinal lumen, respectively. Studies of Slc39A14 knockout mice demonstrated increased Mn levels in the blood and brain but not in the liver, confirming this elimination model and consistent with the clinical features of HMNDYT2 [68, 69]. Subsequent studies of Slc39a14 liver‐specific and intestine‐specific knockout mice demonstrated that enterocyte import of Mn may play a more substantial role in Mn elimination than transport into the hepatocyte followed by biliary excretion [70, 71].

A pathology study of a 4‐year‐old patient with HMNDYT2 who died of septic shock associated with pneumonia demonstrated marked loss of neurons in the globus pallidus and dentate nucleus of the cerebellum, sparing other basal ganglia regions and the cerebral cortex [6]. Immunohistochemical studies of SLC39A14 protein in the brain and liver were not conducted.

As with HMNDYT1, treatment of HMNDYT2 with a combination of chelation with intravenous Na₂CaEDTA and Fe supplementation is recommended [6, 49, 55]. If started early in the course of the disease, Na₂CaEDTA has proven to be effective. For example, a 5‐year‐old girl had a cessation of upper extremity extrapyramidal movements and regained her ability to walk after 6 months of Na₂CaEDTA treatment, while a 17‐year‐old girl had no improvement in her tremor or rigidity, despite a clear demonstration of a marked increase in urine Mn excretion [6]. A chelation and Fe supplementation protocol, identical to the regimen recommended for HMNDYT1, has been described in the consensus guidelines [49].

8. Ongoing and Future Research Questions

While significant progress has been made in our understanding of Mn homeostasis as well as the clinical recognition and treatment of the three disorders of Mn transport, a number of important questions regarding the biology and toxicology of this trace metal remain. In particular, a persistent conundrum concerns the mechanism underlying the extrapyramidal symptoms of manganism, HMNDYT1 and HMNDYT2, and the unresponsiveness of these disorders to levodopa. How does Mn induce a reduction in striatal dopamine release in the absence of degeneration of dopaminergic terminals? Ongoing research has focused on the toxic effects of Mn leading to a disruption of calcium‐sensitive mechanisms of neurotransmitter release [72].

Another area for investigation concerns the varied Mn homeostatic roles of SLC30A10, SLC39A14, and SLC39A8, as well as DMT1, Fpn, and other transporters, in hepatocytes, enterocytes, renal tubular cells, and numerous brain cell types. While somewhat simple models of the pathophysiology of the three disorders of Mn transport can now be stated, clearly many nuances of function and interplay of these proteins remain to be discovered. For example, a series of epidemiologic studies conducted in diverse geographic regions have shown that certain single‐nucleotide polymorphisms (SNPs) in SLC30A10 are associated with subtle but statistically significant alterations in blood Mn levels and with altered measurements of motor function. It was concluded that the SNPs affected neurological function that was independent of peripheral Mn levels and therefore possibly due to altered brain regulation of Mn levels by SLC30A10 [73]. A subsequent study of certain SLC30A10 and SLC39A8 polymorphisms in a cohort of Italian children showed that individuals with polymorphisms associated with high blood Mn levels had increased body sway velocity and higher assessment scores for attention‐deficit/hyperactivity disorder–related behavioral problems [74]. It is not at all clear how these polymorphisms may alter brain Mn levels.

Research focusing on the neuroprotection from or reduction of Mn neurotoxicity has been an active area. The roles of various drugs and natural substances with antioxidant and anti‐inflammatory activity have been of interest. For example, punicalagin (PUN), a natural tannin substance from pomegranate, has both properties. In a rat model of MnCl₂‐induced Parkinson disease, pretreatment with PUN provided protection from Mn neurotoxicity when ambulation parameters were measured. This effect was magnified when PUN was administered with other substances, such as vitamin E [75]. In a rat brain slice model of Mn neurotoxicity, treatment with the lipid‐lowering drug probucol prevented MnCl₂‐induced cell death and levels of oxidative injury markers in both the cerebral cortex and striatum [76]. Cell‐based studies have shown that probucol prevents Mn‐induced inhibition of mitochondrial respiratory chain complexes I and II and therefore reduces the production of reactive oxygen species [77]. Future research on the protective and ameliorative properties of new and repurposed substances for the treatment of Mn toxicity are clearly warranted.

9. Summary

Our understanding of the neurotoxic effects of Mn has evolved over nearly 200 years, initially through the recognition of occupational manganism, and recently with the characterization of the three forms of genetic Mn transporter deficiency. By characterizing these disorders and studying the roles of the three transporters in different tissues, a model of Mn uptake, distribution, and elimination has emerged. Ongoing research will certainly refine this model and improve our understanding of the cellular mechanisms that underlie Mn toxicity. Importantly, consensus guidelines for the clinical recognition, genetic confirmation, and treatment of the two inherited disorders of hypermanganesemia have been published. Hopefully, this will lead to an earlier diagnosis and initiation of treatment with improved clinical outcomes.

Author Contributions

Sidney M. Gospe, Jr.: conceptualization, writing – original draft, writing – review and editing.

Conflicts of Interest

The author declares no conflicts of interest.