Slc30a10 is a manganese transporter in zebrafish

As a first step toward determining whether zebrafish are a suitable model for studying Mn metabolism, we examined whether Mn is metabolized in the same organs in zebrafish as in humans. Wild-type zebrafish embryos were exposed to various concentrations of MnCl₂ for 24 hours at different developmental stages, and their behavior was monitored the following day. We found that embryos are most sensitive to Mn exposure at 5–6 dpf (days post-fertilization); concentrations lower than 0.3 mM had no effect on swimming ability, whereas concentrations higher than 1 mM caused complete immobility in 100% of embryos (S1A Fig). Based on these preliminary data, we consider Mn concentrations lower than 0.3 mM and higher than 1 mM to be relatively low and relatively high, respectively. Importantly, embryos exposed to 1 mM MnCl₂ developed dark-colored brain and liver tissues (S1B and S1C Fig), suggesting that these two organs are sensitive to Mn exposure in zebrafish just like humans [22,23,26].

Next, we measured Mn accumulation in the head, middle trunk, and tail of Mn-exposed embryos. We found higher levels of Mn accumulated in the head and middle trunk, and lower levels in the tail (S1D Fig). After transferring Mn-exposed embryos to fresh Mn-free Holt buffer, the color of both the liver and brain returned to baseline (S1E Fig), and Mn levels were reduced in these embryos (S1F Fig), suggesting that the accumulation of Mn in the brain and liver of wild-type embryos is reversible.

Next, we examined whether zebrafish models are suitable for studying HMDPC, a genetic condition associated with impaired function of the Mn transporter SLC30A10. By searching the zebrafish genomic database, we identified the zebrafish slc30a10 gene as an ortholog of the human SLC30A10 gene, with 42% identity and 55% similarity at the protein level (S2 Fig). Because SLC30A10 is highly expressed in the human brain and liver, where it functions as a Mn exporter, we examined the expression pattern of slc30a10 in zebrafish. We found that slc30a10 is expressed in the brain, liver, and yolk syncytial layer (YSL) of zebrafish embryos (Fig 1A), with robust expression beginning at approximately 5 dpf (Fig 1B). These similarities between humans and zebrafish with respect to their Mn-sensitive organs and SLC30A10 expression patterns (i.e., in the brain and liver) indicate that slc30a10 mutant zebrafish might serve as a suitable model for HMDPC.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Slc30a10 functions as a Mn exporter in zebrafish.

(A) In situ hybridization of wild-type zebrafish using the antisense slc30a10 probe, showing expression in the brain and liver (red arrows), as well as the YSL (red arrowheads). A sense probe was used as a negative control. (B) Semi-quantitative RT-PCR of slc30a10 mRNA in wild-type 1–7 dpf embryos, 14 dpf larvae, and 21 dpf larvae, showing the onset of expression at 5 dpf. (C) DNA and corresponding amino acid sequences of the wild-type (WT), 10-bp deletion (-10), and 1-bp insertion (+1) slc30a10 alleles following CRISPR/Cas9-based editing. (D) Summary of slc30a10 expression in WT and both slc30a10 mutant lines (n = 3 sets of 50 embryos/group). (E-G) Mn, Fe, and Zn concentration was measured in wild-type and mutant 1-week-old embryos (E; n = 6 sets of 1000 embryos/group), 3-week-old larvae (F; n = 3 sets of 50 larvae/group), and 4-month-old adults (G; n = 3 adults/group). (H-J) Heterozygous embryos were exposed to the indicated concentrations of Mn²⁺ (H), Fe³⁺ (ferric ammonium citrate, FAC; I), or Zn²⁺ (J) for 24 hours at 5 dpf. At 6 dpf, slc30a10 expression was measured; n = 3 sets of embryos/group. *p<0.05, **p<0.01, and ***p<0.001; N.S., not significant (p>0.05).

https://doi.org/10.1371/journal.pgen.1006892.g001

To create such a model, we generated two separate slc30a10 knockout zebrafish lines using the CRISPR/Cas9 system. One line contains a 10-bp frameshift deletion at codon 95, which introduces a premature stop at codon 97. The other line contains a 1-bp frameshift insertion at codon 96, which introduces a premature stop at codon 173 (Fig 1C). Both mutant lines have significantly reduced slc30a10 expression (Fig 1D; S1G Fig), and the residual transcripts correspond to the predicted mutant forms, which were confirmed by sequencing (S1H Fig). Because these two lines develop the same phenotype, all subsequent experiments were performed using the line carrying the 10-bp deletion.

To confirm that the zebrafish slc30a10 gene encodes a Mn exporter, we measured the levels of several minerals in mutant animals. Severe hypermanganesemia—the primary metabolic abnormality in HMDPC—is associated with subtle perturbations in iron metabolism in some patients [22,23]. Moreover, the ZnT (SLC30) family of proteins primarily facilitates the efflux of zinc [31–33]. Therefore, we measured Mn, Fe, and Zn levels in embryonic, larval, and adult zebrafish using ICP-MS. Surprisingly, Mn levels were lower in 1-week-old mutant embryos compared to wild-type embryos; in contrast, both Fe and Zn levels were similar between wild-type and mutant embryos (Fig 1E). Because embryos are nourished exclusively by the yolk in this stage of development, the levels of trace minerals in these embryos reflect their absorption from the yolk. In 3-week-old larvae, on the other hand, we found higher Mn levels in the mutants compared to wild-type animals; as in the embryonic stage, both Fe and Zn were similar between groups (Fig 1F). These data indicate that Mn accumulates rapidly in mutant animals during the feeding stage. Finally, in 4-month-old adults, systemic Mn levels were significantly higher in both male and female mutant animals compared to wild-type animals, and Fe levels were similar between wild-type and mutant animals. Interestingly, Zn levels were higher in mutant animals compared to wild-type animals, but only in male animals (Fig 1G). We measured the levels of several other metals in 4-month-old adults, but found no additional differences between mutants and wild-type animals (S1 Table). Taken together, these results indicate that Slc30a10 functions primarily as a Mn exporter during zebrafish development.

Slc30a10 expression is regulated by Mn in zebrafish

We measured slc30a10 expression in wild-type embryos exposed to various concentrations of MnCl₂, ferric ammonium citrate (FAC), and ZnCl₂. Even at a relatively high concentration (1 mM), none of the minerals tested caused a change in slc30a10 expression.

Next, we measured the effect of mineral exposure on slc30a10 expression in heterozygous embryos. Unlike wild-type animals, the heterozygous developed defects when exposed to Mn²⁺ or Fe³⁺ at concentrations higher than 500 μM and Zn²⁺ at concentrations higher than 250 μM.; therefore, we treated embryos with Mn²⁺, Fe³⁺, and Zn²⁺ concentrations below these levels. We found that slc30a10 expression increased in heterozygous with increasing concentrations of Mn²⁺ (Fig 1H) and was increased only with the highest concentration of Fe³⁺ (Fig 1I); in contrast, slc30a10 expression was not affected by Zn²⁺ at any concentration tested (Fig 1J). These data indicate that slc30a10 expression is regulated primarily by Mn²⁺ in heterozygous embryos.

We also found that slc30a10 expression increased in mutant adults following Mn exposure (S3A Fig). In addition, the expression of slc30a10 in mutant adults was higher than in embryos; in contrast, slc30a10 expression was unchanged between the embryonic stage and adulthood in wild-type animals (S3B Fig), suggests that the chronic accumulation of dietary Mn upregulates slc30a10 expression in adult slc30a10 mutants.

Slc30a10 mutant zebrafish develop neurological defects

Neurological disorders, including impaired gait and speech, are the most common clinical features among patients with HMDPC. Similarly, our slc30a10 mutant zebrafish begin to develop neurological defects at approximately 4 months of age. Within the first year, approximately 40% and 20% of the mutant males and females, respectively, are affected; by the second year, these percentages increase to approximately 70% and 40%, respectively. In general, the affected mutant animals are smaller and thinner than their unaffected siblings (Fig 2A and 2B). In addition, our ICP-MS analysis revealed that the affected mutants have higher levels of Mn compared to unaffected siblings (Fig 2C).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Slc30a10 mutants develop impaired neurological function.

(A) Example images of affected mutant adults and their unaffected siblings; note the thinner body shape of the affected siblings. (B) Summary of body weight of affected adults and unaffected siblings (n = 4 adults/group). (C) Affected mutants have higher Mn levels compared to their unaffected siblings (n = 4 adults/group). (D) Example movement traces of affected adults and their unaffected siblings. (E) Affected adults swim shorter distances than their unaffected siblings (n = 6 adults/group). (F) Summary of the reproductive capacity of WT and mutant adults (n = 20 adults/group). (G) Example images of a wild-type and mutant embryo, showing a darker color in the brain and liver (red arrows) in the mutant. (H) Mn exposure causes a larger increase in Mn accumulation in mutant animals compared to wild-type animals (n = 3 sets of 200 embryos/group). (I) Example images of two mutant embryos after Mn exposure. Note the distorted body shape of the Mn-treated mutant. (J) Locomotion (measured as swimming and a normal escape response) is reduced in mutant embryos following Mn treatment, whereas wild-type animals are not affected (n = 3 sets of 20 embryos/group). (K) Mutant embryos exposed to Mn swim a shorter distance than untreated mutants and wild-type embryos (n = 6 embryos/group). (L) Slc32a1 expression is reduced in both wild-type and slc30a10 embryos following Mn exposure (n = 3 sets of 20 embryos/group). (M-N) Gabb1a (M) and gabb1b (N) expression is reduced in Mn-exposed slc30a10 mutants (n = 3 sets of 20 embryos/group). (O) Western blot analysis of Gad65/67 in wild-type and mutant embryos treated with or without Mn. (P-Q) In situ hybridization of dat and th in mutant embryos following Mn exposure, showing reduced expression of both genes compared to untreated mutants and WT embryos. (R) TUNEL staining in the brain of mutant embryos following Mn exposure, showing increased apoptosis compared to untreated mutants and WT embryos. *p<0.05, **p<0.01, and ***p<0.001; in K‒O, groups with different letters differed significantly (p<0.05).

https://doi.org/10.1371/journal.pgen.1006892.g002

In addition to morphological changes, affected mutants have a distinct bradykinesia-like swimming pattern (S1 and S2 Movies). Thus, the affected mutants move slowly, less frequently, and tend to stay at the bottom of the tank. To better characterize the swimming pattern of these affected mutants, we used behavior-tracking software, which confirmed that the affected mutants stayed in a relatively limited area of the tank (Fig 2D) and swam more slowly with shorter distances (Fig 2E) than their unaffected siblings. In addition, reproductive capacity was significantly lower in affected mutants compared to unaffected mutants, leading to an overall decrease in reproduction in the entire group of mutant adults compared with age-matched wild-type animals (Fig 2F).

Interestingly, however, no defects in locomotion were observed in the mutant embryos during the hatching stage (i.e., within 7 dpf). Both wild-type and mutant embryos responded rapidly when touched on the tail (S3 and S4 Movies). We hypothesized that mutant embryos do not exhibit behavioral defects because they have not yet developed excess Mn levels, as shown by our ICP-MS results (Fig 1E); moreover, during this stage, excess metals are not imported, as they have not begun feeding. In humans, manganism gives rise to a well-defined movement disorder with symptoms resembling Parkinson's disease [34,35]. Therefore, we exposed mutant embryos to various concentrations of Mn²⁺ at specific developmental stages in order to induce manganism.

Wild-type embryos have robust slc30a10 expression starting at 5 dpf (Fig 1B), which suggests the start of a Mn-sensitive stage in development. We therefore exposed both wild-type and mutant embryos to various concentrations of Mn²⁺ at 5 dpf and analyzed the embryos at 6 dpf. At 300 μM, Mn caused a clear color change in the brain and liver of mutant embryos, but had no effect in wild-type embryos (Fig 2G). We then measured Mn levels in both Mn-treated and untreated wild-type and mutant embryos and found more accumulated Mn in mutants compared to wild-type (Fig 2H). Moreover, this accumulation in the mutant embryos persisted after the animals were transferred to Mn-free Holt buffer, suggesting that slc30a10 plays an important role in the clearance of accumulated Mn. Finally, following exposure to relatively low concentrations of Mn²⁺ (150–200 μM), mutant embryos—but not wild-type embryos—had impaired locomotion and other defects, including tremor, postural instability, disorientation, and impaired balance (Fig 2I and 2J; S5 and S6 Movies). At higher concentrations of Mn²⁺ (250–300 μM), the mutants developed more severe defects, including body distortion, rigidity, bradykinesia, and a lack of escape response (Fig 2I and 2J; S7–S10 Movies). In addition, mutant embryos exposed to 300 μM Mn swam significantly shorter distances than both wild-type embryos and untreated mutant controls (Fig 2K). At even higher doses (>300 μM), Mn was lethal to some of the mutant animals.

GABAergic neurons are the primary target affected by Mn toxicity [21]. We therefore measured the expression of GABA transporters, GABA receptors, and the glutamic acid decarboxylases Gad65 and Gad67 in Mn-treated and untreated embryos. Exposure to Mn significantly reduced the expression of the GABA transporter slc32a1 in both wild-type and mutant embryos (Fig 2L); in contrast, expression of the GABA receptors gabbr1a and gabbr1b was reduced only in the Mn-treated mutants (Fig 2M and 2N). Finally, Gad65/67 protein levels were significantly increased in Mn-treated wild-type embryos, as well as in both untreated and treated mutants (Fig 2O). Taken together, the reduced expression of the GABA transporter and receptors indicates reduced GABA uptake, while increased glutamic acid decarboxylases suggest more GABA production. These results suggest that extracellular GABA could be increased in both mutant embryos and Mn-treated embryos.

Impaired dopaminergic signaling has been attributed to the parkinsonism-like effects associated with Mn accumulation in the basal ganglia of both HMDPC patients and workers with occupational Mn exposure [8,12,23,35]. Zebrafish models of Parkinson's disease develop two classic pathological features, namely locomotor deficits and a loss of dopaminergic neurons [36]. Moreover, the escape response in zebrafish embryos requires functional dopaminergic circuits [37,38]. We therefore investigated whether the Mn-induced parkinsonism-like phenotype observed in Mn-exposed mutant embryos is due to a defect in dopaminergic signaling by measuring the expression of the dopamine transporter (dat) and tyrosine hydroxylase (th) in embryos with and without Mn exposure; both dat and th mRNA are commonly measured as molecular markers of dopaminergic neurons in zebrafish [39,40]. We found that under basal conditions, both dat and th gene expression patterns were similar between mutant and wild-type embryos. However, following exposure to Mn, mutant embryos had significantly lower levels of both dat and th mRNA (Fig 2P and 2Q; S3C and S3D Fig), suggesting decreased dopaminergic signaling. In addition, Mn-treated mutants—but not wild-type or untreated mutant embryos—had increased apoptosis markers in the brain (Fig 2R; S3E and S3F Fig), consistent with neuronal death specifically in mutant embryos following Mn exposure. Taken together, these data suggest that Mn accumulation leads to impaired GABAergic transmission and loss of dopaminergic neurons, possibly explaining the impaired locomotion observed in mutant zebrafish with Mn accumulation.

Slc30a10 mutant zebrafish develop hepatic pathology

Patients with HMDPC develop a wide range of liver complications, including steatosis, hepatomegaly, hepatic fibrosis, and—ultimately—cirrhosis [23,26]. We therefore examined whether our mutant zebrafish develop a similar hepatic phenotype. Under basal conditions, mutant embryos develop a mild fatty liver, and this effect was increased upon Mn exposure (Fig 3A). HE staining and Oil red O staining of frozen sections confirmed hepatic steatosis in Mn-exposed mutant embryos (Fig 3B and 3C). In addition, Mn exposure caused a dose-dependent darkening of the liver in mutant embryos (Fig 3D and 3E), consistent with hepatic Mn accumulation. At 4 months of age, Sirius red staining revealed hepatic fibrosis in mutant animals (Fig 3F). Moreover, the mRNA levels of the fibrosis biomarkers col1a1a and ctgfa were higher in mutant animals compared to wild-type animals (Fig 3G and 3H). These data indicate that slc30a10 mutant zebrafish develop hepatic pathology similar to the symptoms associated with HMDPC.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Slc30a10 mutant zebrafish develop liver damage.

(A) Oil red O staining shows that mutant embryos have a slightly fatty liver (red arrows), which was worsened upon exposure to Mn. (B-C) HE staining (B) and Oil red O staining (C) of frozen sections showing hepatic steatosis in mutant embryos following Mn exposure. (D) Example images of a wild-type and mutant embryo under Mn exposure, showing a darker colored liver in the mutant (red arrow). (E) Dose-response curve showing the percentage of embryos with liver color change versus Mn concentration (n = 3 sets of 20 embryos/group). (F) Image of the liver of a wild-type and mutant adult, showing severe fibrosis in the mutant liver with sirius red staining. (G-H) The fibrosis markers col1a1a and ctgfa were measured in the liver of both male and female WT and mutant animals (n = 3 sets of 20 adults/group; *p<0.05 and **p<0.01).

https://doi.org/10.1371/journal.pgen.1006892.g003

Slc30a10 mutant zebrafish develop polycythemia

Polycythemia develops early in all patients with HMDPC and is clinically distinguishable from patients with environmentally induced manganism [22,23,26]. We therefore measured the relative number of erythrocytes in both wild-type and mutant zebrafish. To visualize the erythrocytes, the animals were crossed to the Tg(globinLCR:eGFP) background, in which erythrocytes are labeled with GFP [41]; we then measured erythrocytes using FACS analysis.

Our analysis revealed an increase in GFP-positive cells in 3-week-old mutants compared with wild-type animals (Fig 4A), consistent with polycythemia. Increased systemic Mn may drive erythropoiesis by increasing the expression of erythropoietin (Epo), similar to hypoxia-stimulated erythropoietin expression [42]. We therefore measured epo expression in 3-week-old animals and found significantly increased expression in mutant animals (Fig 4B), and increased hepatic epo in 4-month-old mutants (Fig 4C). Interestingly, however, at the 1-week-old embryonic stage, neither the number of erythrocytes nor the expression of epo differed between mutant and wild-type embryos (Fig 4A and 4B). These results suggest that the polycythemia in mutant embryos develops secondary to the chronic accumulation of Mn and does not arise directly from slc30a10 deficiency.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Slc30a10 mutant zebrafish develop polycythemia, and chelation treatments improves the phenotype in mutant embryos.

(A) Summary of the relative numbers of erythrocytes in 3-week-old and 1-week-old embryos (n = 6 sets of 50 embryos/group). (B and C) Epo expression was measured in 3-week-old and 1-week-old mutant zebrafish (B) and in the liver of adult zebrafish (C); n = 3 sets of 20 adults/group. (D and E) Mn-induced locomotor defects (D) and the darker color in liver (E) were rescued by treating embryos with EDTA-CaNa₂ (chelator, n = 3 sets of 20 embryos/group). (F-G) Example images of a Mn-exposed mutant embryo, showing a darker colored liver (F), and a Mn-exposed mutant embryo following EDTA-CaNa₂ treatment (G). (H) Following exposure to 300 μM Mn, mutant embryos were transferred to either Holt buffer alone or Holt buffer containing EDTA-CaNa₂, and locomotion was measured (n = 3 sets of 20 embryos/group). *p<0.05, **p<0.01, and ***p<0.001.

https://doi.org/10.1371/journal.pgen.1006892.g004

EDTA-CaNa₂ treatment partially rescues the phenotype in slc30a10 mutant zebrafish

Chelation therapy with EDTA-CaNa₂ is commonly used to treat manganism [18,23]. We therefore tested whether chelation therapy can prevent the Mn-induced phenotype in our mutant zebrafish model by exposing mutant embryos to Mn in the presence or absence of EDTA-CaNa₂. Treating mutant animals with EDTA-CaNa₂ restored both locomotion and the escape response (Fig 4D and S11 Movies), and it reduced the Mn-induced hepatic color change (Fig 4E–4G), suggesting that EDTA-CaNa₂ chelation is an effective treatment for slc30a10 mutant zebrafish and that this model may be used to screen other chelation therapies. To test whether chelation therapy can reverse the effects of Mn (thereby mimicking better the therapeutic approach in patients), we first exposed mutant embryos to Mn; affected embryos with impaired locomotion and hepatic pathology were then transferred to Mn-free Holt buffer with or without EDTA-CaNa₂ (2 mg/ml). Treatment with EDTA-CaNa₂ improved the locomotion of Mn-exposed mutant embryos (Fig 4H), but had no effect on hepatic color change (S3G Fig), indicating that this therapy partially reverses the Mn-induced phenotype in slc30a10 mutant zebrafish.

Ferrous fumarate partially rescues the phenotype in slc30a10 mutant zebrafish

Ferrous fumarate has been used to reduce intestinal Mn absorption and relieve hypermanganesemia in patients with HMDPC [22,28,29]. Therefore, we exposed mutant embryos to Mn in the presence or absence of ferrous fumarate at 5 dpf and examined their phenotype the following day. Ferrous fumarate treatment caused the gut to turn brown in color, which suggest increased iron uptake in these embryos (S3H Fig). Importantly, however, treating mutant embryos with ferrous fumarate significantly improved locomotion (S3I Fig) and significantly reduced the rate of hepatic color change (S3J Fig), consistent with the notion that iron competes with Mn for intestinal import and can partially rescue Mn-induced hypermanganesemia in zebrafish embryos.

Increased expression of atp2c1 (spca1) in slc30a10 mutant zebrafish

In preliminary experiments, we found that exposing mutant embryos to 300 μM Mn at 2–3 dpf (prior to the onset of slc30a10 expression) had no effect, even when the embryos were exposed for several days; in contrast, the same concentration of Mn is toxic to slightly older (i.e., 5–6 dpf) mutant embryos. We therefore exposed mutant embryos to 300 μM Mn for 72 hours beginning at 3 dpf (“3–6 dpf mutants”) or for 24 hours beginning at 5 dpf (“5–6 dpf mutants”); we then examined the phenotype of both groups at 6 dpf. We found that embryos exposed beginning at 5 dpf had a clear phenotype, including reduced survival rate, locomotion defects and a change in liver color (Fig 5A–5C; S12 Movies). Strikingly, however, embryos there were exposed to Mn for a longer period (beginning at 3 dpf) were unaffected (Fig 5A–5C; S4A–S4C Fig). Using ICP-MS, we detected slightly lower Mn levels in 3–6 dpf mutants compared to 5–6 dpf mutants, despite a longer period of Mn exposure (Fig 5D). These data suggest that prior to the onset of slc30a10 expression, another exporter protein may be expressed, thereby preventing the toxic effects of Mn exposure.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. atp2c1 expression is increased in slc30a10 mutant zebrafish.

(A-C) Mutant embryos were exposed to the indicated concentration of Mn for 24 h starting at 5 dpf (“5-6dpf” group) or 72 h starting at 3 dpf (“3-6dpf” group). At 6 dpf, survival (A), locomotion (B), and dark colored live (C) were measured. (D) Summary of normalized Mn levels in 3–6 dpf and 5–6 dpf mutants. (E) Slc40a1, atp2c1, and atp132a mRNA was measured in WT and mutant embryos (left), as well as in the liver (middle) and brain (right) of WT and mutant adults. (F-G) Summary of atp2c1 expression measured in the 3–6 dpf group (F) and in the 5–6 dpf group (G); groups with different letters differed significantly (p<0.05). (H-J) Mutant embryos were injected with an atp2c1 morpholino (Mut MO) or a scrambled control (Mut Con), the embryos were exposed to the indicated concentration of Mn for 72 h starting at 3 dpf, and survival (H), locomotion (I), and liver color change (J) were measured. (K) Summary of normalized Mn concentration in Mn-exposed Mut MO and Mut Con embryos. (L) Mutant embryos were injected with an mRNA encoding human ATP2C1 containing the Q747A mutation (Mut Q747A) or the control, and 3 dpf embryos were exposed to 2 mM Mn. (M) Embryos injected with Mut Q747A mRNA have reduced Mn accumulation than control mutants following Mn exposure. A-C, E-J, and L: n = 3 sets of 20 animals/group; D, K, and M: n = 3 sets of 200 embryos/group. *p<0.05, **p<0.01, and ***p<0.001.

https://doi.org/10.1371/journal.pgen.1006892.g005

To test this idea, we measured the expression of several other exporters, including atp2c1, slc40a1, and atp13a2 [43], and examined whether these genes are upregulated in slc30a10 mutants. Our analysis revealed that only one gene—atp2c1—is upregulated in mutant embryos compared to wild-type animals (Fig 5E), suggesting that this gene may compensate for the dysfunction of slc30a10.

We also examined the effect of Mn exposure in 3 dpf wild-type and mutant embryos and found increased expression of atp2c1 (Fig 5F)—but not slc40a1 or atp13a2 (S4D and S4E Fig)—in both wild-type and mutant embryos, suggesting that atp2c1 is selectively upregulated in response to excess Mn at this early developmental stage. Interestingly, however, atp2c1 expression was unaffected by Mn exposure in either wild-type or mutant embryos at 5 dpf (Fig 5G; S4F and S4G Fig), suggesting the presence of a relatively narrow developmental window in which atp2c1 expression can respond to Mn exposure. Taken together, these data indicate that although the expression of atp2c1 is higher in slc30a10 mutant embryos compared to wild-type embryos under basal Mn conditions, atp2c1 expression responds to Mn exposure only in an early developmental stage, suggesting that Atp2c1 may play a compensatory role as an Mn exporter when slc30a10 expression is low.

To test the notion that Atp2c1 can functionally replace Slc30a10 in early development, we measured expression patterns using in situ hybridization and found that atp2c1 is expressed in the brain and abdominal viscera of zebrafish embryos (S4H Fig), which overlaps partially with the expression pattern of slc30a10 (i.e., in the brain and liver), indicating that atp2c1 and slc30a10 are expressed in similar organs. Next, we used a morpholino to block atp2c1 expression in wild-type embryos and examined the effects of 72 hours of Mn exposure beginning at 3 dpf. Although Mn-treated WT-MO embryos survived (S4I Fig), they had impaired locomotion and developed a dark-colored liver; in contrast, wild-type embryos injected with a control (scrambled) morpholino were unaffected by Mn exposure (S4J and S4K Fig). These data suggest that Atp2c1 plays a protective role against Mn toxicity in early zebrafish development.

Next, we examined the effect of blocking atp2c1 expression in slc30a10 mutant embryos exposed to Mn for 72 hours beginning at 3 dpf. We found that MO-treated mutant embryos were more sensitive to Mn exposure and had reduced survival and more severely impaired locomotion compared to control mutants injected with a scrambled morpholino (Fig 5H and 5I). In addition, MO-injected mutant embryos had a darker colored liver (Fig 5J) and higher levels of Mn compared to control mutants (Fig 5K). Taken together, these data support the notion that Atp2c1 can function as a Mn exporter under systemic Mn stress and can protect slc30a10 mutant embryos from Mn-induced toxicity selectively in early developmental stages.

Recently, Mukhopadhyay and Linstedt suggested that increasing the Mn-pumping capacity of ATP2C1 may be a feasible strategy for treating Mn overload in patients [44]. To test this strategy using our zebrafish model, we injected single-cell mutant embryos with human ATP2C1 mRNA containing the gain-of-function Q747A mutation (ATP2C1-Q747A) and measured Mn clearance at 3 dpf. Compared with control mutant embryos, ATP2C1-Q747A‒injected embryos had higher Mn resistance and lower Mn levels (Fig 5L and 5M), confirming that overexpressing ATP2C1-Q747A improves Mn clearance in zebrafish embryos early in development. Interestingly, however, ATP2C1-Q747A‒injected mutants were not rescued when exposed to Mn at 5 dpf, possibly due to a degradation of mRNA by this stage.

Atp2c1 can compensate for Slc30a10 in Mn metabolism

We found that although both 3–6 dpf and 5–6 dpf mutant embryos (i.e., mutant embryos exposed to Mn for 72 or 24 hours, respectively) developed a color change in the liver (Fig 5C), only the 5–6 dpf mutants had decreased survival and impaired locomotion (Fig 5A and 5B). We hypothesized that this difference could be due to a difference in the subcellular distribution of Mn between 3–6 dpf and 5–6 dpf mutants. Due to a lack of adequate antibodies against the zebrafish proteins, we tested this hypothesis in HeLa cells [21,44,45].

HeLa cells have barely detectable levels of SLC30A10 expression and robust levels of ATP2C1 expression (S5A Fig), which mimics the expression pattern of slc30a10 and atp2c1 in 3 dpf mutant zebrafish embryos (S5B Fig, 3 dpf). In contrast, expressing mutated SLC30A10 in HeLa cells mimics the expression pattern of 5 dpf mutant zebrafish embryos (S5B Fig, 5 dpf). We then used GPP130, a Mn sensor, to measure intra-Golgi Mn levels [21,44]. We found that exposing control-transfected HeLa cells to Mn led to increased GPP130 degradation (i.e., increased Mn transport to the Golgi apparatus) (Fig 6A); in contrast, Mn transport to the Golgi apparatus was significantly decreased in HeLa cells transfected with the mutant SLC30A10 (Fig 6B and 6C). Thus, by association we conclude that the subcellular distribution of Mn differs between 3–6 dpf mutants and 5–6 dpf mutants, possibly explaining their striking difference in phenotypes.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. Functional crosstalk between SLC30A10 and ATP2C1 in HeLa cells.

(A) HeLa cells were immunostained for ATP2C1 (green) and the Mn sensor protein GPP130 (blue). Following overnight exposure to 300 μM Mn, most of the ATP2C1 was translocated to the Golgi apparatus, and GPP130 was degraded. (B) HeLa cells were transfected with SLC30A10-Q308Stop-flag (Q308Stop, a mutant form of SLC30A10 in which codon 308 is changed to a stop codon); the cells were then immunostained for flag (red) and GPP130 (blue). Following Mn exposure, only half of the GPP130 was degraded. (C) Quantitative analysis of GPP130 immunoreactivity. (D) HeLa cells were transfected with galactosyltransferase-RFP (GalT) plasmid to label the Golgi apparatus (red); the cells were then immunostained for ATP2C1 (green). Following Mn exposure, most of the ATP2C1 was translocated to the Golgi apparatus, where it co-localized with GalT. (E-G) HeLa cells were co-transfected with the indicated constructs (E: GalT-RFP and SLC30A10-FLAG; F: GalT-RFP and SLC30A10-Q308Stop-flag; G: GalT-RFP and SLC30A10-1st-2nd-exons-deletion-flag, a mutant form of SLC30A10 in which the first and second exons are deleted); the cells were then immunostained for ATP2C1 (green) and FLAG (blue). (H) HeLa cells were mock-transfected cells or transfected with the indicated construct. After overnight exposure to Mn (500 μM), Mn concentration was measure using IPC-MS. ***p<0.001.

https://doi.org/10.1371/journal.pgen.1006892.g006

Next, we measured the subcellular localization of ATP2C1 in HeLa cells transfected with wild-type or mutant SLC30A10. We found that ATP2C1 was localized to the Golgi apparatus and throughout the cytoplasm under control conditions; however, following Mn treatment, the majority of ATP2C1 protein was localized to the Golgi apparatus (Fig 6A and 6D; S5C Fig). Wild-type SLC30A10 was located at the cell membrane (S5D Fig), whereas the mutant SLC30A10 was localized to the endoplasmic reticulum (S5D Fig). However, despite their differences in subcellular localization, both wild-type and mutant SLC30A10 disrupted the Mn-induced trafficking of ATP2C1 to the Golgi apparatus (Fig 6E–6G).

Next, we measured the levels of ATP2C1 protein in HeLa cells. We found that Mn exposure increased ATP2C1 expression in untransfected cells (S5F and S5G Fig), similar to our results in zebrafish embryos (Fig 5F). However, Mn exposure had no effect on ATP2C1 levels in HeLa cells transfected with wild-type SLC30A10 and slightly decreased ATP2C1 levels in HeLa cells transfected with mutant SLC30A10 (S5F and S5H Fig), similar to the expression pattern measured in zebrafish embryos at 5 dpf (Fig 5G). Taken together, these data indicate that ATP2C1 expression increases in response to Mn only when SLC30A10 expression is low.

Lastly, we measured Mn levels in HeLa cells following Mn exposure. We found that cells expressing mutant SLC30A10 accumulated significantly more Mn compared to control-transfected cells and cells expressing wild-type SLC30A10 (Fig 6H).

Mineral exposure

Zebrafish embryos were exposed to various concentrations of MnCl₂ or ZnCl₂, and their phenotypes and survival were recorded at different time points, depending on the specific experimental aims. Unless otherwise specified, the embryos were exposed to the indicated MnCl₂ solution for 24 hours starting at 5 dpf (days post-fertilization).

Reverse transcription and quantitative PCR

Total mRNA was isolated from the embryos and tissues using Trizol (Invitrogen). First-strand cDNA was synthesized using the PrimeScript RT reagent kit and Reverse Transcriptase M-MLV (M1701, Promega). Real-time PCR was performed using the two-step quantitative RT-PCR method with a kit (RR037A, Takara Bio), and target gene expression was normalized to β-actin mRNA levels.

Whole-mount in situ hybridization

Whole-mount in situ hybridization was performed as described previously [59].

Generation of knockout lines using CRISPR/Cas9

Mutant zebrafish lines were generated using previously reported methods [60]. In brief, a target site (GCTCTGCTTCTCCATCAGCATGG) in exon 1 of the slc30a10 gene was selected, and the guide RNA (gRNA) template was amplified from the pMD-gata5-gRNA scaffold vector. In vitro transcription was performed using 1 μg of template DNA and T7 RNA polymerase. Zebrafish codon-optimized Cas9 plasmid [61] was linearized with XbaI, and Cas9-capped mRNA was transcribed using the mMESSAGE mMACHINE T7 kit (Ambion). The size and quality of the capped mRNA and gRNA were confirmed by electrophoresis through a 2% (w/v) agarose gel. Subsequently, 100 pg of gRNA and 400 pg of Cas9 mRNA were injected directly into single-cell embryos. The following primer pairs and the restriction enzyme BstXI were used to assess the efficiency of genetic disruption: forward 5’-ACTCCTTCAACATGCTGTCGGACA-3’, reverse 5’-AGCACCTGTTTCCACCTGCT-3’. The injected embryos were raised to adulthood. F0 fish were crossed with wild-type fish to generate heterozygous F1 progeny, which were then genotyped using BstXI digestion and DNA sequencing. Heterozygous F1 fish were crossed in order to generate homozygous mutant fish. The mutant and wild-type animals were obtained from heterozygous crosses.

Inductively coupled plasma mass spectrometry (ICP-MS)

At least 1000 1-week-old embryos or 50 3-week-old larvae were collected in order to obtain the required amount of tissue (100 mg) for analysis. Single adult zebrafish were measured separately. At least three groups were measured for each stage. HeLa cells were seeded in 6-well dishes and transfected the next day. Twenty-four hours after transfection, the cells were exposed to 500 μM MnCl₂ in serum-containing growth medium for 12 hours. After Mn exposure, the cells were washed with phosphate-buffered saline (PBS) and incubated in 500 μl trypsin-EDTA at 37°C for 3 min. The trypsin was then neutralized by adding 1 ml of serum-containing growth medium, and the cells were collected by centrifugation for 3 min in a microfuge. The supernatants were discarded, and the cell pellets were washed twice in 1 ml PBS containing 10 mM EDTA, followed by centrifugation as described above. After a final wash in PBS (without EDTA), the cell pellets were re-suspended in 200 μl RIPA buffer in order to extract the proteins; 1–2 μl of protein extract was used to measure protein concentration, and the remainder was used to measure the concentration of Mn. All samples were transferred to acid-washed cans with 5 ml metal-free HNO₃ and 1 ml HF, then digested overnight at room temperature; the following day, 1 ml HClO₄ was added, and the samples were incubated at 180°C for 10 hours. The cans were then transferred to 120°C and shaken for 2 hours in order to evaporate the residual HF and HClO₄. Acid-washed vials were prepared by incubation in 20% HNO₃ for 1 day, followed by extensive washing in ultrapure water. The digested samples were then diluted to 50 ml with Milli-Q water in the acid-washed vials and measured using an Agilent 7500 series inductively coupled plasma mass spectrometer.

Western blot analysis

Whole embryos were lysed in modified RIPA buffer containing 150 mM sodium chloride, 50 mM Tris (pH 7.4), 1% Triton X-100, 1% sodium deoxycholate, 1% SDS, and proteasome inhibitor with PMSF, then centrifuged at 4°C for 15 minutes. Equal amounts of protein in each sample were resolved by SDS-PAGE on a 10% Mini-PROTEAN TGX Precast Gel (Bio-Rad) and transferred to an Immuno-Blot PVDF membrane (Bio-Rad). The membrane was blocked with 5% (w/v) non-fat milk and incubated with anti-GAD65/67 (ab11070, Abcam) and anti-actin (R1207-1, Huabio) antibodies. After washing three times for 10 minutes each with Tris-buffered saline containing Tween-20, the membranes were incubated with HRP-conjugated secondary antibodies for 1 hour at room temperature. Finally, the membranes were incubated in chemiluminescence reagent (34080, Pierce) and exposed to X-ray film.

Phenotyping

Embryos were incubated in 6-well plates, and their morphological changes and behavioral patterns were observed and recorded using a Nikon SMZ18 stereo microscope. For measuring the escape response, a soft tip was used to touch the tail of embryo, and locomotion was recorded. For behavioral analysis, a digital video tracking system running the ANY-maze software program was used. The experiments were performed in a temperature-controlled room between 1:00 pm and 5:00 pm. The fish were allowed to acclimate to the system for 1 h before the start of data acquisition. Each adult fish was placed in a 20 cm × 10 cm tank containing 1 L of water, and the total distance traveled was recorded for 30 s. Each embryo was then placed in a well of a 24-well plate containing 2 ml Holt buffer with or without Mn, and the total distance traveled was recorded for 10 min.

TUNEL apoptosis assay

Apoptosis were determined using the in situ Cell Death Detection Kit with TMR red (12156792910, Roche) in accordance with the manufacturer’s instructions. The fluorescence signal was analyzed using ImageJ software.

Frozen sections

After fixation with 4% paraformaldehyde in PBS overnight, 6 dpf embryos were washed with PBS and incubated in a 30% sucrose solution overnight. The embryos were then incubated in a 1:1 mixture of 30% (w/v) sucrose and OCT compound (Tissue-Tek) for 2 hours, followed by overnight incubation in OCT. The samples were frozen in OCT compound at -80°C, and 10-μm cryo-sections were cut using a CM1950 cryostat (Leica).

Immunofluorescence

Cells were fixed with 4% paraformaldehyde in PBS for 10 minutes, washed three times with PBS, and then permeabilized in 0.2% Triton X-100 in PBS for 5 minutes. After washing three times with PBS and blocking with PBS containing 5% fetal bovine serum (FBS), the cells were immunostained with rabbit anti-flag (2368s, Cell Signaling), rabbit anti-giantin (ab80864, Abcam), mouse anti-pan-Cadherin (ab22744, Abcam), or mouse anti-atp2c1 (WH0027032M1, Sigma) in PBS with 1% FBS at 4°C overnight. The cells were then washed three times with PBS and incubated with mouse anti-calnexin Alexa Fluor 647 (ab202572, Abcam), goat anti-rabbit Alexa Fluor 488 (A11008, Molecular Probes), donkey anti-mouse Alexa Fluor 488 (ab150105, Abcam), donkey anti-rabbit Alexa Fluor 555 (A31572, Molecular probes), and/or goat anti-rabbit Alexa Fluor 350 (A0408, Beyotime) in PBS with 1% FBS at 37°C for 1.5 hours. The slides were mounted under coverslips using ProLong Gold Antifade reagent (Thermo Fisher Scientific), and fluorescence images were obtained using an LSM-710 microscope (Zeiss, Germany) equipped with a 63X/1.4 numerical aperture oil-immersion objective.

Cell culture and transfection

HeLa cells were grown in Dulbecco's modified Eagle's medium (DMEM, Gibco) supplemented with 10% FBS (Gibco) and 1X penicillin-streptomycin (Gibco). The cells were cultured at 37°C in humidified air containing 5% CO₂. The cells were transfected using the X-tremeGENE HP DNA Transfection Reagent (Roche) in accordance with the manufacturer’s instructions. The cells were generally transfected 24 hours after plating and used 36 hours after transfection. Where indicated, freshly prepared MnCl₂ solution was added to the medium at a final concentration of 300 μM for the indicated times.

Oil red O staining

Embryos were fixed in 4% formaldehyde overnight at 4°C, washed with PBS, and stained with filtered 0.3% oil red O in 60% 2-propanol for 2 hours. After a final wash in PBS, the embryos were placed in depression wells and photographed.

Erythrocyte measurements using FACS analysis

One-week-old and 3-week-old larvae on the Tg(globinLCR:eGFP) background were crushed in Hank's buffered salt solution (HBSS, Invitrogen), followed by complete dissociation to a single-cell suspension by trituration. The cells were washed and harvested in HBSS and then examined using a Cytomics FC 500 MCL flow cytometer (Beckman Coulter, Inc.).

Microinjection of mRNA and morpholinos

The full-length wild-type human ATP2C1 cDNA was cloned using PCR, and the Q747A mutation was introduced using two-step site-directed mutagenesis. The cDNA sequence containing the mutation was inserted into the pCS2 expression plasmid, linearized, and transcribed using the mMESSAGE mMACHINE SP6 kit (Ambion). The resulting mRNA was injected into one-cell stage zebrafish embryos at 200 pg/embryo. The atp2c1 morpholino (5’-TGCATCTTCCGTCAGCTTTCGTTGC-3’) or a scrambled sequence was injected into one-cell stage embryos at 0.1 nmol/embryo.

Rescue and recovery experiments

A final concentration of 2 mg/ml EDTA-CaNa₂ was prepared and either mixed with MnCl₂ (at a final concentration of 250–300 μM) or used alone. Saturated ferrous fumarate was added to 5 ml 200 μM MnCl₂ to assess rescue efficiency. For controls, treated embryos were washed several times with fresh Holt buffer and then incubated in Holt buffer.

Statistical analyses

All summary data are presented as the mean ± the standard error of the mean (SEM). Differences between two groups were analyzed using the two-tailed Student’s t-test, and multiple groups and time points were analyzed using a two-way ANOVA. Differences with a p-value <0.05 were considered significant.

Ethics statement

Approval for the study was granted by the Institutional Animal Care and Use Committee of the Laboratory Animal Center, Zhejiang University. We followed the relevant guidelines from the Laboratory Animal Center of Zhejiang University. We anaesthetized or sacrificed the zebrafish by freezing them in ice before performing the experiments.

Data availability

All relevant data are provided within the paper and the Supporting Information.