Functional analysis of SLC39A8 mutations and their implications for manganese deficiency and mitochondrial disorders

SLC39A8 encodes ZIP8, a divalent metal ion transporter. Mutations in the SLC39A8 gene are associated with congenital disorder of glycosylation type II and Leigh syndrome. Notably, affected patients with both disorders exhibited severe manganese (Mn) deficiency. The cellular function of human SLC39A8 (hSLC39A8) and the mechanisms by which mutations in this protein lead to human diseases are unclear. Herein, we show that hSLC39A8 mediates 54Mn uptake by the cells, and its expression is regulated by Mn. While expression of wild-type hSLC39A8 increased 54Mn uptake activity, disease-associated mutations abrogated the ability of the transporter to mediate Mn uptake into the cells, thereby providing a causal link to severe Mn deficiency. All mutants failed to localize on the cell surface and were retained within the endoplasmic reticulum. Interestingly, expression of hSLC39A8 mutants of both CDG type II and Leigh syndrome reduced mitochondrial 54Mn levels and activity of Mn-dependent mitochondrial superoxide dismutase MnSOD, and in turn increased oxidative stress. The expression of wild-type hSLC39A8, but not the disease-associated mutants, promoted mitochondrial functions. Moreover, loss of function analyses further corroborate hSLC39A8’s critical role in mediating Mn uptake and mitochondrial function. Our results provide a potential pathogenic mechanism of diseases that are associated with hSLC39A8 mutations.


Results
hSLC39A8 is a Mn-specific transporter. To directly test whether hSLC39A8 mediates Mn uptake, we measured 54 Mn uptake into the cells. As shown in Fig. 1A, hSLC39A8-transfected cells strongly stimulated the uptake of 54 Mn compared to empty vector (control)-transfected cells. hSLC39A8-dependent 54 Mn uptake was concentration-dependent and saturable, with an apparent Km of ~1.44 ± 0.39 μM (Fig. 1A). To assess the substrate specificity of hSLC39A8, competition assays were performed with cells expressing hSLC39A8 in the presence of excess non-radioactive metal cations. Figure 1B shows the degree to which different metal ions compete with Mn uptake. The application of a 10-or 50-fold excess of non-radioactive Mn and Cd caused a marked decrease in 54 Mn uptake by the cells, with the order of the inhibitory effect expressed as Mn = Cd > Zn > Fe = Cu. These results indicate that hSLC39A8 mediates Mn uptake into the cells with a very high affinity, and cadmium has the strongest inhibitory effect on Mn uptake. These observations are consistent with previous reports on SLC39A8 5,7 as well as other mammalian SLC39 transporters 13,14 .
We then investigated whether Mn regulates the expression of hSLC39A8. The HeLa cells were treated with a Mn chelator, para-aminosalicylic acid (PAS), with a range of Mn concentrations. The mRNA levels of endogenous hSLC39A8 were determined by quantitative PCR (qPCR). As shown in Fig. 1C, hSLC39A8 expression was elevated ~6-fold by Mn-limiting conditions compared to that in the basal medium (Fig. 1C). Moreover, various Mn concentrations were tested until the hSLC39A8 mRNA level reached the plateau. Figure 1D shows that hSLC39A8 mRNA levels were significantly upregulated by ~1.7-fold (P = 0.02), ~4.6-fold (P < 0.001), and ~6.8-fold (P < 0.0001) with Mn treatments of 100, 200, and 300 µM, respectively (Fig. 1D). The increased expression of endogenous hSLC39A8 by Mn implies the existence of a homeostatic mechanism, by which hSLC39A8 maintains optimal Mn levels within cells.
Mn sensor GPP130 validates the Mn uptake activity of hSLC39A8. To further test whether hSLC39A8 functions as a Mn uptake transporter, we used the Golgi protein GPP130, which is known to be a specific Golgi Mn sensor 15 . Previous studies have shown that a specific increase in intra-Golgi Mn led to GPP130 trafficking from the Golgi to late endosomes and lysosomes, where GPP130 was degraded 15 . Thus, we hypothesized that if hSLC39A8 is a Mn-specific uptake protein, it will uptake Mn from the cell exterior into the cytoplasm and provide the metal to the Golgi, inducing GPP130 degradation. The stability of GPP130 was assessed with and without MnCl 2 treatment by immunoblot in cells expressing hSLC39A8 (Fig. 1E). Consistent with previous studies 16,17 , the level of GPP130 was significantly reduced when control cells were treated with Mn (Fig. 1E,F). Importantly, this Mn-induced degradation was significantly reduced in cells expressing hSLC39A8 compared to control cells treated with Mn. Quantification indicated that 79% of GPP130 was lost in cells expressing hSL-C39A8-WT after 4 h Mn treatment, while only a 46% decrease was seen in control cells (Fig. 1F). To test if GPP130 is indeed degraded in the cells expressing hSLC39A8, we pretreated cells with the lysosomal inhibitor NH 4 Cl. The pretreatment blocked the Mn-induced reduction of GPP130 levels, indicating that GPP130 is degraded by lysosome upon SLC39A8 expression ( Supplementary Fig. 1). Taken together, these data validate that hSLC39A8 is a key Mn transporter in mammalian cells.

Disease-associated mutations impair Mn uptake activity of hSLC39A8. Mutations in hSLC39A8
were identified to cause severe Mn deficiency in patients with CDG type II and Leigh syndrome 8,9,11 . To determine the impact of these mutations on Mn transport, we constructed an expression vector encoding hSLC39A8 with a C-terminal HA epitope tag and introduced four disease-associated mutations into hSLC39A8. Four mutations included two missense and two compound missense mutations of SLC39A8 that are found in CDG type II or Leigh syndrome patients. The mutations and their localization are summarized in Fig. 2A. The first mutation associated with CDG type II patients (hSLC39A8-M1, G38R) was a homozygous c.112 G > C substitution. The second mutation associated with CDG type II patients (hSLC39A8-M2, G38R; I340N) was a compound heterozygous for c.112 G > C and c.1019 T > A. The third mutation associated with CDG type II patients (hSLC39A8-M3, V33M; G204C; S335T) was a compound heterozygous for c.97 G > A, c.610 G > T, and c.1004 G > C. The fourth Scientific REpoRtS | (2018) 8:3163 | DOI:10.1038/s41598-018-21464-0 mutation associated with Leigh syndrome (hSLC39A8-M4, C113S) was a homozygous c.338 G > C substitution. All four mutants are missense changes that affect highly conserved amino acid residues (Fig. 2B).
To probe the impact of these mutations on Mn transport activity, HeLa cells were transfected with the control, hSLC39A8-WT or hSLC39A8-mutants (hSLC39A8-M1, -M2, -M3, or -M4) and then assayed for 54 Mn uptake activity. The expression of hSLC39A8-WT resulted in a ~2.2-fold increase (P = 0.008) in 54 Mn uptake activity compared to the control cells (Fig. 3A). In contrast, all four mutants completely abrogated the increase in 54 Mn uptake observed in cells expressing hSLC39A8-WT (P < 0.05). Moreover, the expression of hSLC39A8-WT Data represent means ± SEM (n = 3 samples/group). # P < 0.05 vs. control: Student's t-test. (E) Representative immunoblot of GPP130 in total cell lysates isolated from empty vector-(control), or hSLC39A8-WT expressing cells incubated with 500 µM MnCl 2 for 4 h or 100 µM for 16 h. Equal loading was verified by immunoblotting with actin antibody. Full-length blots are presented in Supplementary Figure 3. (F) Quantification of GPP130 relative protein after normalization with actin. Results are means ± SEM (n = 3 samples/group) of three independent experiments. # P < 0.05 vs. control at the same concentration, *P < 0.05 vs. control in the different groups; Student's t-test. Mutations in hSLC39A8 are associated with severe Mn deficiency, congenital disorder of glycosylation type II, and/or Leigh syndrome. (A) A model of predicted topology of human SLC39A8 is shown with eight transmembrane domains, extracellular NH 2 or COOH terminus, and the COOH terminal HA tag used in these experiments. Four mutations identified in patients with Mn deficiency were examined in this study and the corresponding locations of these mutations are indicated. (B) Evolutionary alignment of SLC39A8 amino acid sequence showing strict conservation of Mn deficiency-associated hSLC39A8 mutations. Protein sequences for SLC39A8 were aligned using ClustalW2. Residues identical to the human SLC39A8 sequence are marked with an asterisk (*). Conservation between amino acids of strongly and weakly similar properties is indicated by a colon (:) and a period (.), respectively. Putative protein domains are predicted using MEMSATSVM and include a signaling peptide in turquoise (position 1-22), a histidine-rich region in green (position 226-231), and the metalloprotease motif in green (position 343-351). Disease-associated mutations are highlighted in yellow (V33M, G38R, C113S, G204C, S335T, I340N). The protein sequences used to generate this alignment include NP_001128618.1 (human), JAA33667.1 (chimpanzee), NP_001192559.1 (cow), NP_001128622.1 (mouse), AAH89844.1 (rat), and XP_009305480.1 (zebrafish).
Scientific REpoRtS | (2018) 8:3163 | DOI:10.1038/s41598-018-21464-0 significantly increased intracellular Mn levels (P < 0.05), whereas the expression of hSLC39A8-mutants did not (Fig. 3B). Despite the known ability of SLC39A8 to transport zinc, iron, and cadmium, no differences were observed in intracellular zinc, iron, and copper levels in cells expressing hSLC39A8-WT or its mutants ( Supplementary Fig. 2). Note that the intracellular cadmium levels were below the detection limit in cells expressing hSLC39A8-WT or mutants.
To further test the effect of hSLC39A8 mutations on Mn transport activity, we again used GPP130. We hypothesized that if hSLC39A8 mutations abrogate Mn uptake activity, they will inhibit GPP130 degradation. Consistent with our data in Fig. 1, the level of GPP130 was significantly reduced in cells expressing hSLC39A8 treated with Mn (Fig. 3C,D). Importantly, expression of the disease-associated hSLC39A8 mutants blocked the Mn-induced loss of GPP130 (Fig. 3C,D). Taken together, these data clearly indicate that the disease-associated mutations inhibit the Mn transporting activity of hSLC39A8. No difference was detected in protein levels between hSLC39A8-WT and mutations at the predicted molecular mass (~50 kDa) (Fig. 3E,F). No difference was detected in the endogenous expression of SLC39A8 in cells expressing hSLC39A8 and its mutants (Fig. 3G), confirming that Mn deficiency by hSLC39A8 mutations was not affected by the endogenous expression of hSLC39A8 in HeLa cells. These results, combined with the severe Mn deficiency observed in both CDG type II and Leigh syndrome 8,9,11 , demonstrate loss-of-function mechanisms underlying CDG type II and Leigh syndrome that are associated with hSLC39A8 mutations.
Disease-associated hSLC39A8 mutants are mislocalized. The impaired Mn uptake activity observed in the pathogenic mutants may be caused by defects in localization and/or protein folding. To assess the effects of mutations on protein localization, we first examined the localization of HA-tagged hSLC39A8-WT. HeLa cells expressing hSLC39A8-WT were analyzed by confocal microscopy with or without the detergent Triton X-100, which permeabilizes the plasma membrane. In non-permeabilized cells, surface labeling with anti-HA antibody was strong for wild-type hSLC39A8-HA (Fig. 4A,B). Surface staining for all mutants was substantially weaker or undetectable. In permeabilized cells, hSLC39A8-WT was predominantly detected at the plasma membrane and throughout the cytosol with intracellular vesicle staining (Fig. 4C,D). In contrast, all mutant proteins showed clear staining of the nuclear envelope and a reticulated pattern of fluorescence extending from the nucleus into the cytoplasm, suggesting retention of the mutants within the ER. Our measurement of Pearson's correlation coefficient 18,19 showed statistically significant increases in HA−calnexin colocalization signals within mutant-expressing cells compared to WT-expressing cells (P < 0.0001, Fig. 4D). Taken together, our data indicate that hSLC39A8-WT is primarily localized on the cell surface, whereas the disease-associated mutants are retained in the ER. The failure of localization at the plasma membrane likely explains the inability of hSLC39A8-mutants to transport Mn into cells.

The expression of hSLC39A8, but not its disease-associated mutants, led to increased mitochondrial Mn levels and MnSOD activity. Considering the reduction in intracellular Mn levels, together
with the association of SLC39A8 mutations with the mitochondrial disorder Leigh syndrome, we hypothesized that these mutations may lead to reduced activity of MnSOD, the Mn-requiring antioxidant enzyme in mitochondria. We first aimed to determine whether these mutants affect Mn levels in the mitochondria. Mitochondria were isolated in cells expressing hSLC39A8-WT or its mutants, and the Mn levels in the mitochondria were directly measured in vitro with 54 Mn. The expression of hSLC39A8-WT resulted in a ~2.3-fold increase (P < 0.001) in 54 Mn levels in the mitochondria compared to control cells (Fig. 5A). In contrast, all four mutants displayed significantly lower 54 Mn levels in the mitochondria compared to WT (P < 0.001). These data suggest that the expression of hSLC39A8-WT, but not its mutants, provides Mn to the mitochondria.
We next aimed to determine whether the disease-associated mutations affect MnSOD activity. Mitochondria were isolated in cells expressing hSLC39A8-WT or its mutants, and MnSOD activity in the mitochondria was measured. Compared to hSLC39A8-WT cells, MnSOD activity was significantly reduced in cells expressing hSL-C39A8-M1 (~19.2%, P = 0.008), M2 (~31.4%, P = 0.002), M3 (~36.3%, P = 0.001), and M4 (~63.3%, P < 0.001) (Fig. 5B). Taken together, these results clearly suggest that disease-associated hSLC39A8 mutations in both CDG type II and Leigh syndrome result in low levels of mitochondrial Mn, which are required for key mitochondrial enzymes, such as MnSOD.
Disease-associated mutations of hSLC39A8 negatively influence the expression of genes involved in oxidative phosphorylation. While Leigh syndrome is characterized by deficits in mitochondrial functions, mitochondrial abnormality has not been reported in CDG type II. The mitochondrial Mn levels reduced by the hSLC39A8 mutants associated with both Leigh syndrome (M4) and CDG type II (M1-M3) led us to postulate that these two previously unconnected disorders may share mitochondrial dysfunction as a common feature. We first examined the effect of the mutant expression on mitochondrial function by exploring both mitochondrial DNA (mtDNA)-and nuclear DNA (nDNA)-encoded electron transport genes. For mtDNA genes, we chose ND1-ND6, COXI-COXIII, 12SRNA, and 16SRNA, which are encoded by either the heavy strand or the light strand of mtDNA and known to play important roles in oxidative phosphorylation. Data in Fig. 6A show that the expression of hSLC39A8-WT, but not the CDG type II mutants (M1-M3) or the Leigh syndrome mutant (M4), led to increased expression levels of these mtDNA-encoded genes (P < 0.05). We further analyzed the mRNA transcript levels of the nDNA-encoded subunits, including succinate dehydrogenase subunits SDHA and SDHB. The gene expression of nDNA-encoded subunits was also significantly reduced in cells expressing each mutant compared to hSLC39A8-WT (P < 0.05), yet the reduction was generally milder than mtDNA genes (Fig. 6B). We also tested the effect of hSLC39A8 mutations on the mtDNA copy number. Interestingly, hSL-C39A8-WT expression, but not its mutants, led to a marked increase in the mtDNA copy number (P < 0.05, Fig. 6C), suggesting that hSLC39A8 can promote the production of mtDNA, but disease-associated mutations interfere with such function. No differences were observed in mitochondrial mass using MitoTracker green in cells expressing hSLC39A8-WT or its mutants (Fig. 6D), suggesting that mitochondria in the mutant-expressing cells contain less mtDNA. The reduced mtDNA might be responsible for the reduced transcript levels from mtDNA, at least in part.
Because oxidative metabolism is an important function of mitochondria, we measured the mitochondrial respiration using a Seahorse extracellular flux analyzer. We found a significant decrease in basal respiration, ATP-linked respiration, and maximal respiration in cells expressing hSLC39A8-M2, -M3, and -M4, respectively, compared to hSLC39A8-WT (Fig. 7D,E). Taken together, these results indicate that compared to the WT, hSLC39A8 mutants are less capable of supporting mitochondrial functions, including mitochondrial membrane potential, ATP production, mitochondrial redox activity, and mitochondrial respiration.
The expression of hSLC39A8 mutants enhances ROS generation. The mitochondrial electron transport chain is the major intracellular source of reactive oxygen species (ROS), including superoxide anion, hydroxyl radical, and various peroxides and hydroperoxides 20 . Scavenging ROS is an important role of mitochondria 21 . To determine whether the observed mitochondrial dysfunction also involves impaired ROS scavenging, we measured ROS levels using fluorescent indicators specifically for superoxide (O 2− ) (MitoSox Red) and H 2 O 2 production (H 2 DCFDA). MitoSOX Red selectively targeted to the mitochondria exhibits red fluorescence following oxidation by superoxide anion, thereby serving as an indicator of mitochondrial superoxide levels. Compared to cells expressing hSLC39A8-WT, MitoSOX Red fluorescence intensity was significantly increased in cells expressing hSLC39A8-M4 by ~30.4% (P < 0.001), while hSLC39A8-M1, M2, and M3 were not observed to have an effect (Fig. 7F). We then measured global oxidative stress level by H 2 DCFDA, which is converted to highly fluorescent 2′,7′-dichlorofluorescein (DCF) following the removal of the acetate groups by intracellular esterases and ROS-induced oxidation. The DCF fluorescence was significantly increased in cells expressing SLC39A8 M1 (~16.4%, P < 0.05), M2 (~35.0%, P < 0.001), M3 (~16.4%, P = 0.006), and M4 (~79.8%, P < 0.001) compared to cells expressing hSLC39A8-WT (Fig. 7G). Furthermore, we measured the oxidative damage levels using 8-isoprostane levels. Our data indicated that cells expressing each of the four hSLC39A8 mutants significantly increased isoprostane levels compared to hSLC39A8-WT cells (P < 0.001). These results indicate that disease mutations interfere with hSLC39A8 function to suppress excess ROS levels with a stronger impact of Leigh syndrome associated mutation (M4) within the mitochondria.
Knockdown of hSLC39A8 reduces Mn uptake activity and impairs mitochondrial function. In addition to the effects of hSLC39A8 overexpression, we tested the effect of hSLC39A8 depletion on Mn uptake and mitochondrial function. We used siRNA to knockdown hSLC39A8 in HeLa cells. RT-PCR analysis showed that hSLC39A8 is effectively reduced by ~90% (P < 0.001) compared with cells transfected with a control siRNA that did not target any human gene (Fig. 8A). Our data indicated that the knockdown of hSLC39A8 reduced 54 Mn uptake activity by ~45% (P < 0.001) compared to control siRNA-transfected cells (Fig. 8B). Furthermore, 54 Mn levels in the mitochondria (~35%, P < 0.001) and MnSOD activity (~47%, P < 0.001) in hSLC39A8-depleted cells were significantly reduced (Fig. 8C,D). We also found a significant reduction in the mtDNA copy number in hSLC39A8-depleted cells (~24%, P = 0.002) (Fig. 8E), with no significant change in mitochondrial mass (Fig. 8F). Importantly, hSLC39A8 RNAi significantly reduced mitochondrial membrane potential (~20%, P < 0.001), ATP production (~37%, P < 0.001), and mitochondrial redox activity (~20%, P < 0.001). We also measured the mitochondrial respiration using the Seahorse extracellular flux analyzer, and found a significant decrease in basal respiration, ATP-linked respiration, maximal respiration, and spare respiratory capacity in hSLC39A8-depleted cells compared with control siRNA cells (Fig. 8J,K). Concordantly, the knockdown of hSLC39A8 significantly increased superoxide production by MitoSox Red (~10%, P < 0.001) and H 2 O 2 production by H 2 DCFDA (~16%, P < 0.001). Taken together, the knockdown results complement our overexpression studies and indicate that hSLC39A8 plays an essential role in mediating Mn uptake and modulating mitochondrial function.

Discussion
In this report, we describe the first in vitro functional analysis of human SLC39A8 mutations associated with severe Mn deficiency 8,9,11 . Our studies have led to several insights regarding the function and regulation of human SLC39A8 as a Mn transporter, and shed novel light on the mechanisms by which mutations in hSLC39A8 cause Mn deficiency-associated developmental conditions. First, we provided several lines of evidence for the specific requirement of hSLC39A8 for Mn uptake among other metals, such as zinc. Our radiolabeled 54 Mn uptake studies have shown that hSLC39A8 has a high affinity for Mn. The Km of ~1.44 ± 0.39 µM for Mn 2+ is consistent with previous reports on SLC39A8 5,7 and other mammalian SLC39 transporters 13,14 and within a similar range observed in other cell lines or tissues 5,7,13,14 . Our metal competition studies further confirm that hSLC39A8 has a higher affinity for Mn and cadmium compared to zinc. Previous in vitro studies have shown that SLC39A8 mediates the uptake of zinc, iron, and cadmium 5,6 . In the competition assay, we also found that zinc could compete with Mn, yet to a lesser extent compared to Mn and Cd. The ability of hSLC39A8 to transport zinc or the competition by zinc in these assays may not necessarily mean that this protein physiologically transports zinc. This idea is supported by the fact that unlike in vitro studies 5,6 , human patients carrying hSLC39A8 mutations 8,9,11 or SLC39A8-inducible global and liver-specific knockout mice 7 have shown no changes in zinc and iron levels. Furthermore, in our experiments, the difference in metal transport between hSLC39A8-WT and the four disease mutants was only observed for Mn, not for zinc, iron, or copper ( Supplementary Fig. 2). Based on these observations, it is likely that hSLC39A8 acts primarily as a Mn transporter under physiological conditions. However, we cannot rule out the possibility that hSLC39A8 may be necessary for transporting zinc under certain circumstances, such as when cells lack other efficient zinc transporters.
The mislocalization of pathogenic mutations in SLC39A8 to the ER (Fig. 4) suggests that these mutations may alter protein targeting motifs and/or lead to loss-of-function of its intrinsic metal transport activity. As shown in Fig. 2A,B, topology prediction of hSLC39A8 reveals that hSLC39A8 adopts the classic SLC39 structure of eight transmembrane (TM) domains with their N-and C-termini facing the extracytoplasmic space 4 . Disease-associated mutations reside in either the N-terminal cytoplasmic region, TM III, or TM IV. Importantly, these mutations change non-polar residues (Glycine, Isoleucine, Valine, and Cysteine) to polar (Asparagine, Threonine) or charged (Arginine) amino acids that are likely to disrupt protein folding. All four mutations found in the N-terminal cytoplasmic regions might affect the sequence motifs that control their subcellular localization, trafficking, and function. Further studies will be required to test whether mutations in these distinct protein segments influence Mn transport activity.
Mn is an essential nutrient that acts as a cofactor for key biological enzymes, including carboxylases and phosphatases in the cytosol, sugar transferase in the Golgi, and superoxide dismutase in mitochondria [22][23][24][25] . Despite its essentiality, excess Mn accumulates in the mitochondria 26,27 , producing Mn toxicity associated with mitochondrial dysfunction, oxidative stress, and cell apoptosis 28,29 . While a number of studies have shown the effect of excess Mn on mitochondrial function, our study demonstrates that Mn deficiency caused by hSLC39A8 deficiency may impair mitochondrial function and enhance oxidative stress. The expression of both CDG type II-associated hSLC39A8 mutants (hSLC39A8-M1, -M2, -M3) and the mitochondrial disorder-associated hSLC39A8 mutant (hSLC39A8-M4) resulted in mitochondrial dysfunction and oxidative stress. Glycosylation defects in CDG type II are thought to be explained by the requirement of Mn for the activity of β-1,4-glycosyltransferase 9 . It is currently unknown whether these patients have mitochondrial abnormalities. Testing if CDG type II also involves abnormalities of mitochondria, including their morphology, dynamics, and function, will likely provide important insights into the pathogenesis of the developmental and neurological symptoms observed in this condition.
The novel role of hSLC39A8 in mitochondrial functions may involve multiple mechanisms. Mn is required as a cofactor for MnSOD, a reactive oxygen species scavenger found in mitochondria 30 . Mn deficiency in yeast leads to the reduced activity of MnSOD and elevated levels of superoxide 31 . ROS can damage enzymes containing Fe-S clusters, including complex I, II, and III of the respiratory chain, and damage mtDNA, which encodes subunits of complex I, III, IV, and V 30 . Our data clearly demonstrate that the expression of hSLC39A8 mutants reduced mitochondrial Mn levels and MnSOD activity concurrent with increased ROS levels. Thus, we speculate that reduced MnSOD activity by hSLC39A8 mutants generates ROS, which may negatively influence oxidative phosphorylation machineries and impair mitochondrial function. Additional mechanisms of hSLC39A8-mediated enhancement of mitochondrial functions could be promoting the expression of oxidative phosphorylation-related genes and maintaining the appropriate mtDNA copy number (Fig. 6). Again, altered mitochondrial ROS might underlie the change in the mtDNA copy number. Future studies are required to examine the mechanistic relationship between intracellular Mn levels and mitochondrial function.
The hSLC39A8's role in providing Mn to MnSOD to suppress excess ROS may provide a hint regarding the pathogenesis of neurological symptoms observed in human patients. All patients with hSLC39A8 mutations show intellectual disability and cerebellar atrophy 8,9,11 . Recent epidemiological studies have shown that both low and high Mn exposures are associated with adverse neurodevelopmental outcomes 32,33 . MnSOD knockout mice showed neuronal degeneration in the basal ganglia and brainstem that was characterized by extensive mitochondrial damage 34 . Similar clinical pathologies, such as progressive neurodegeneration with brainstem and basal ganglia dysfunction, are observed in the SLC39A8-deficienct patients with Leigh-like mitochondria diseases 35 . However, we note the limitation of our study as well. SLC39A8 is widely expressed in many tissue cell types 36 . Studies are currently underway to validate the roles of SLC39A8 in other cell types, especially the ones related to neurological symptoms, and should be able to define the causal roles of Mn, ROS, and mitochondria in neurological symptoms that are linked to SLC39A8.
In summary, our current study provides direct and functional evidence that hSLC39A8 is a Mn-specific transporter and genetic mutations in hSLC39A8 interfere with Mn uptake and mitochondrial function (Fig. 9). Concomitantly, the expression of these mutants and depletion of hSLC39A8 by RNAi reduces the mitochondrial Mn levels, MnSOD activity, and gene expression of oxidative phosphorylation enzymes, which are accompanied by mitochondrial dysfunction and increased oxidative stress. Our studies provide an important link between Mn and mitochondrial function in the pathogenesis of diseases that are associated with hSLC39A8. The findings extend our current knowledge of the pathogenesis of inherited Mn deficiencies, which may facilitate the development of therapeutic targets to treat these disorders. Further studies are warranted to evaluate the incidence of these mutations in the human population and their relevance to human health and disease.

Methods
Vector Construction and Site-directed Mutagenesis of hSLC39A8. hSLC39A8 cDNA was obtained from a cDNA clone (GenBank TM accession number BC012125.1). The entire coding region of the wild-type forms of hSLC39A8 were amplified and tagged at the C terminus with two tandem hemagglutinin (HA) epitopes by PCR using primers (FWD 5′-CCACCATGGCCCCGGGTCGCGCGGTG-3′ and Rev 5 ′-GGCGTAGTCGGGG ACGTCGTAGGGGTAGGCGTAGTCGGGGACGTCGTAGGGGTACTCCAATTCGATTTCTCCTGC-3′ (25 cycles: 94 °C for 30 s, 58 °C for 30 s, and 72 °C for 2.5 min) and inserted into pcDNA3.1/V5-His TOPO vector (Invitrogen). Point mutations were introduced into this construct using the QuikChange mutagenesis kit (Agilent Technologies). The site-directed mutation, orientation and fidelity of the insert, and incorporation of the epitope tag were confirmed by directed sequencing (University of Michigan DNA Sequencing Core). The sequencing data and the plasmids are available upon request.
Cell culture and the expression of hSLC39A8 and its mutants in HeLa cells. All culture media and supplements were purchased from Invitrogen (Carlsbad, CA, USA). Heat-inactivated fetal bovine serum was purchased from Sigma-Aldrich (St. Louis, MO, USA). HeLa cells were grown in DMEM containing 10% fetal bovine serum, penicillin (100 IU/ml), and streptomycin (100 mg/ml) at 37 °C in a humidified, 5% CO 2 incubator. DNA transfections were performed with Lipofectamine 3000 (Invitrogen) according to the manufacturer's specifications. Cultures were generally transfected 24 h after plating and used 48 h after transfection. For small interfering RNA-mediated gene suppression of SLC39A8, HeLa cells were plated in six-well plates overnight and then transfected with 50 nM of control siRNA (Sigma, SIC001-10NMOL) or hSLC39A8 siRNA (Sigma, SASI_ Hs02_00355573) using Lipofectamine 3000 (Invitrogen) as described above.
Immunofluorescence and microscopy. For confocal studies, transfected cells were plated onto glass coverslips and were fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) for 10 min, and immunofluorescence staining was performed as described previously 18,37 . To permeabilize the cells, cells were incubated with 0.2% Triton X-100 in PBS for 5 min. Nonspecific binding was blocked with 4% bovine serum albumin (BSA) in PBS for 30 min, and transfected HA-tagged hSLC39A8-WT or its mutants were detected following incubation with Alexa 488-conjugated anti-mouse HA (1 μg/ml, Biolegend, Cat no. 90159) for 1 h. Rabbit anti-calnexin (1 μg/ml, Abcam, Cat no. ab22595) or pan-cadherin (1 μg/ml, Abcam, Cat no. ab6529) was used as ER or cell surface markers, respectively. Detection of calnexin or pan-cadherin was performed using an anti-rabbit IgG  Table 1 and were all purchased from Integrated Genomics Technologies.
MtDNA copy number. Total DNA was extracted from cell samples via TRIzol (Invitrogen) extraction.
Following complete removal of the RNA-containing aqueous phase, DNA extraction buffer [Tris base (1 M), sodium citrate dibasic trihydrate (50 mM), and guanidine thiocyanate (4 M)] was added to the tubes containing the remaining Trizol-separated interphase and infranatant. The tubes were shaken vigorously and centrifuged at 12,000 × g at room temperature for 30 min. The aqueous phase was collected, and the genomic and mitochondrial DNA were precipitated in isopropanol. Samples were respun at 12,000 × g at 4 °C to pellet the DNA. The DNA pellet was then washed in 70% ethanol, respun, and, after careful ethanol removal, resuspended in TE buffer. To quantify the mtDNA copy number, qPCR was performed as described above against external standards for mtDNA and β-globin using primers listed in Table 1.
MitoTracker assay. The Mitochondrial mass was measured by a MitoTracker Green FM dye (Invitrogen), a dye that stains mitonchondria independent of its membrane potential. Cells were stained with 100 nM MitoTracker probes at 37 °C for 30 min and washed two times with PBS. The fluorescence of MitoTracker (excitation 495 nm, emission 520 nm) was measured at 25 °C using a BioTek Synergy microplate reader (BioTek Instruments, Winooski, VT). Immunoblot analysis. The total lysates were prepared in a hypotonic buffer using 1% NP-40 plus protease inhibitors (Roche, Cat. No. 11836153001). Protein concentration was determined by Bradford assay. Samples (30-50 μg) were separated by electrophoresis and transferred to a nitrocellulose membrane (Bio-Rad, Cat. No. 1620115). The membrane was immunoblotted with anti-mouse HA antibody (Biolegend, Cat. No. 901501), anti-rabbit GPP130 antibody (Biolegend, Cat. No. 923801), anti-rodent total OXPHOS antibody (Abcam, Cat. No. ab110413), anti-mouse GFP antibody (Santa Cruz, Cat. No. 101536), and anti-mouse actin (Proteintech, Cat. No. 60008-1-Ig). The blots were visualized with infrared anti-mouse or anti-rabbit secondary antibodies, using a LI-COR Odyssey fluorescent Western blotting system (LI-COR Biosciences). Protein expression was quantified using densitometry (Image Studio Lite; LI-COR). Mn superoxide dismutase assay. Mitochondrial MnSOD activity was determined using a Superoxide Dismutase Assay Kit (Cayman Chemical, Ann Arbor, MI, USA) as described previously 38 . Briefly, the cells were scraped into ice-cold MB buffer (10 mM HEPES, pH 7.5, 210 mM mannitol, 70 mM sucrose, 1 mM EDTA) and then dounce homogenized with a glass homogenizer. The cell extract was centrifuged at 1,000 × g for 5 min at 4 °C, and the mitochondrial fraction was pelleted by centrifugation at 12,000 × g for 12 min at 4 °C. MnSOD activity was determined in the presence of 2 mM potassium cyanide to inhibit Cu/Zn-SOD.
54 Mn uptake assay. 54 Mn uptake assay was determined as described previously 39,40 . Briefly, cells were incubated at 37 °C for 30 min in serum-free media containing 1 µM 54 Mn. Cells were then chilled on ice and washed three times with PBS containing 1 mM EDTA to remove any unbound 54 Mn. Cell-associated radioactivity was determined with a gamma counter and was normalized to the cell protein measured in lysates using the Bradford assay. For mitochondrial 54 Mn levels, mitochondria were isolated through differential centrifugation, and then mitochondria-associated 54 Mn levels were measured with a gamma counter.
Trace element analysis. HeLa cells transfected with hSLC39A8-WT or other mutants were analyzed for metals by inductively coupled plasma mass spectrometry (ICP-MS) (Lumigen Instrument Center, Department of Chemistry, Wayne State University, MI, USA), as described previously 38,41 . Briefly, the cell samples were digested with 2 mL/g total wet weight nitric acid (BDH ARISTAR ® ULTRA) for 24 h, and then digested with 1 mL/g total wet weight hydrogen peroxide (BDH Aristar ® ULTRA) for 24 h at room temperature. The samples were preserved at 4 °C until quantification of metals. Ultrapure water was used for final sample dilution. MTT mitochondrial redox activity assay. Mitochondrial metabolic function was assessed using an MTT assay, as described previously 38,39 . This assay is based on the ability of the mitochondrial enzyme succinate dehydrogenase to metabolize MTT into formazan. The cells were incubated for 3 h with MTT (0.5 mg/ml), and then MTT solvent (isopropyl alcohol containing 0.1 N HCl) was added to dissolve the formazan formed. The resulting formazan was quantified spectrophotometrically at 570 nm, and the background was subtracted at 690 nm using a plate reader.

Monitoring of mitochondrial membrane potential (ΔΨm
Assessment of ATP production rate. The ATP production rate was measured by a CellTiter-Glo ® Luminescence Cell Viability Assay kit (Promega) according to the manufacturer's instruction. The amount of ATP was directly proportional to the number of living cells present in the culture. Briefly, cells expressing hSL-C39A8-WT or its mutants were cultured in 96-well white-walled plates. Then, 100 µl of CellTiter-Glo reagent was added to lyse the cells. The contents were mixed for 2 min on an orbital shaker to induce cellular lysis followed by incubation at room temperature for 10 min to stabilize the signal; the luminescence was then recorded immediately. The luminescence intensity was measured at 25 °C using a plate reader.
Analysis of mitochondrial respiration. Extracellular flux analyses were performed using a Seahorse XFe 96 analyzer (Agilent Technologies, Santa Clara, CA, USA), according to the manufacturer's instructions. Twenty-four hours before assay, HeLa cells transfected with hSLC39A8-WT or its mutants, hSLC39A8-siRNA, or scrambled siRNA, were cultured on Seahorse XF96 plates at a density of 8 × 10 3 cells per well. Cells were washed and incubated with XF assay Medium (Seahorse Bioscience), and supplemented with 25 mM glucose, 1 mM sodium pyruvate, and 2 mM L-glutamine at 37 °C and 0% CO 2 for 1 hour. Baseline oxygen consumption rate (OCR) were measured at 37 °C four times before sequentially injecting the following: Oligomycin (1 uM) to measure the ATP-linked OCR, oxidative phosphorylation uncoupled FCCP (0.5 uM) to determine maximal respiration, and rotenone (1 uM) and antimycin A (1uM) to determine the non-mitochondrial respiration. OCR were automatically calculated by the Seahorse XFe-96 software. sites available, whereas AChE is held constant. The assays were performed in accordance with the manufacturer's instructions.