Zinc is critical for a multitude of cellular processes, including gene expression, secretion and enzymatic activities. Cellular zinc is controlled by zinc-chelating proteins and by zinc transporters. The recent identification of zinc permeability of the lysosomal ion channel TRPML1 (transient receptor potential mucolipin 1), and the evidence of abnormal zinc levels in cells deficient in TRPML1, suggested a role for TRPML1 in zinc transport. In the present study we provide new evidence for such a role and identify additional cellular components responsible for it. In agreement with the previously published data, an acute siRNA (small interfering RNA)-driven TRPML1 KD (knockdown) leads to the build-up of large cytoplasmic vesicles positive for LysoTracker™ and zinc staining, when cells are exposed to high concentrations of zinc. We now show that lysosomal enlargement and zinc build-up in TRPML1-KD cells exposed to zinc are ameliorated by KD of the zinc-sensitive transcription factor MTF-1 (metal-regulatory-element-binding transcription factor-1) or the zinc transporter ZnT4. TRPML1 KD is associated with a build-up of cytoplasmic zinc and with enhanced transcriptional response of mRNA for MT2a (metallothionein 2a). TRPML1 KD did not suppress lysosomal secretion, but it did delay zinc leak from the lysosomes into the cytoplasm. These results underscore a role for TRPML1 in zinc metabolism. Furthermore, they suggest that TRPML1 works in concert with ZnT4 to regulate zinc translocation between the cytoplasm and lysosomes.
Transition metals, including zinc, are widely recognized for their toxic effects caused by acute and chronic exposure, despite most of the transition metals being indispensable for proper cell function. Zinc has the widest repertoire of known biological roles, which include being a cofactor in several enzymes and DNA-interacting proteins [1,2]. Zinc deficiency, as well as excess, has been shown to be deleterious for cells [3,4], prompting the need for tight regulation of its cellular levels. Such regulation is carried out by a system of transporters, chelating proteins and zinc-sensitive transcriptional responses [5–7].
Zinc enters the body through enterocytes, primarily through Zip4, a member of the Zip [Slc39a (solute carrier family 39a)] family of zinc transporters [8,9]. From the enterocyte cytoplasm, zinc moves into the bloodstream through a series of transporters, including those belonging to the ZnT (Slc30a) family [6,10]. In general, Zip transporters are responsible for the influx of zinc into the cytoplasm, whereas ZnT transporters are responsible for its efflux. In the bloodstream, zinc is taken up by cells via Zip transporters, and by the endocytosis of zinc bound to plasma proteins. After entering the cytoplasm, zinc is chelated by cytoplasmic proteins, such as MTs (metallothioneins) , and extracted into organelles or across the cell membrane by ZnT transporters. For example, ZnT2 and ZnT4 transporters are localized in intracellular vesicular compartments, including lysosomes. They play important roles in vesicular zinc accumulation [12–14]. The reason for zinc chelation and extraction is two-fold. First, high levels of zinc are damaging to cells [3,4]. Secondly, a number of biological processes requiring zinc take place inside organelles, necessitating zinc transport across the organellar membranes [6,15]. MTs also serve as zinc reservoirs, releasing chelated zinc under low-zinc conditions .
The balance between ZnTs, Zips and MTs is responsible for the net zinc flux and for the free cytoplasmic zinc concentration in the given cell type. Zinc spikes, driven by variations in dietary zinc uptake, are read by zinc-responsive transcription factors, including MTF-1 (metal-regulatory-element-binding transcription factor-1), which binds cytoplasmic zinc through its zinc-finger motifs, leading to its nuclear translocation . Upon activation by cytoplasmic zinc, MTF-1 induces the transcription of genes coding for MTs and some ZnTs, such as ZnT1 and ZnT2 [17,18]. This feedback system allows cells to rapidly quench spikes in free cytoplasmic zinc. The combined activity of zinc transporters and chelators keeps the free zinc concentration in the nanomolar range.
The recent identification of zinc permeability through the lysosomal ion channel TRPML1 (transient receptor potential mucolipin 1)  raised the possibility that this ion channel is involved in zinc transport. This notion was later supported by the finding that zinc builds up in the lysosomes of TRPML1-KD (knockdown) cells . TRPML1 is an ion channel residing in the later portions of the endocytic pathway due to the presence of lysosomal localization signals on its C- and N-termini [21,22]. TRPML1 is encoded by the MCOLN1 gene, which was identified as the gene mutated in the lysosomal storage disease MLIV (mucolipidosis type IV) [23,24]. MLIV is associated with the build-up of cytoplasmic storage bodies, motor dysfunction and developmental delays [25–27]. The characterization of TRPML1 activation by PtdIns(3,5)P2  suggested its activation by delivery to the PtdIns(3,5)P2-rich lysosomes and late endosomes . The discussion of TRPML1 function has been focused on its role in the fusion of vesicles in the endocytic pathway (reviewed in ), and, recently, lysosomal secretion . At the same time, the evidence of iron and zinc permeability through TRPML1  raises an interesting possibility that it plays a role in regulating cellular transition metal levels.
In the present study, we sought to delineate the mechanism through which TRPML1 participates in cellular zinc homoeostasis. We based our search on the assumption that, if TRPML1 is directly involved in the regulation of zinc transport, then cellular zinc trafficking and zinc-dependent processes should change as a function of TRPML1 status. Our assays show that both of these assumptions are true, strongly suggesting that TRPML1 is involved in cellular zinc homoeostasis. We found that the effects of zinc in TRPML1-KD cells included a characteristic lysosomal enlargement. The zinc-sensing cytoplasmic transcription factor MTF-1 plays a key role in this process, as MTF-1 KD reversed zinc-dependent lysosomal enlargement in TRPML1-KD cells. Suppressing the expression of the vesicular zinc transporter ZnT4 eliminated the zinc-induced lysosomal enlargement and zinc retention in TRPML1-deficient cells. Pulse–chase experiments with extracellular zinc show that zinc clearing from vesicular structures including lysosomes is delayed in TRPML1-KD cells. However, zinc secretion was not decreased in these cells; instead, it was somewhat increased. We conclude that TRPML1 works in concert with ZnT4 to regulate lysosomal zinc trafficking, perhaps by providing a lysosomal zinc leak pathway. This is supporting evidence of the novel role of TRPML1 in regulating cellular transition metals and TRPML1 as a new target in transition metals toxicity.
HeLa cells were maintained in DMEM (Dulbecco's modified Eagle's medium; Sigma–Aldrich) supplemented with 7% FBS (fetal bovine serum), 100 μg/ml penicillin/streptomycin and 5 μg/ml Plasmocin prophylactic (Invivogen). For siRNA (small interfering RNA) and cDNA transfection, antibiotic-free medium was used. Metals were added directly to DMEM. Zinc (100 μM) was added to antibiotic-free medium, containing serum, 24 h after transfection, for 48 h.
siRNA-mediated KD and plasmid transfection
ON-TARGET plus siRNA were designed as described previously [30,31] and synthesized by Dharmacon. The TRPML1 siRNA probe targeting the sequence 5′-CCCACATCCAGGAGTGTAA-3′ in MCOLN1 was used for all TRPML1 KDs. The MTF-1 siRNA (catalogue number SASI_Hs01_00177112), ZnT2 siRNA (catalogue number SASI_Hs01_00055662) and ZnT4 siRNA (catalogue number SASI_Hs00225995) were from Sigma. Non-targeting control siRNA#1 (Sigma) was used as a negative control. Cells were transfected using Lipofectamine™ 2000 (Invitrogen) as described by the manufacturer using 600 nM siRNA per 35 mm well. All KDs were confirmed using SYBR Green (Fermentas)-based qPCR (quantitative PCR). For DNA transfections, 2 μg of HA (haemagglutinin)-tagged human ZnT2 or ZnT4 were used in parallel with 1 μg of GFP (green fluorescent protein)-tagged TRPML1 constructs.
Reverse transcription and qPCR
RNA was isolated from cells using TRIzol® (Invitrogen) according to the manufacturer's protocol. cDNA was synthesized using the GeneAmp RNA PCR system (Applied Biosystems) with oligo(dT) priming. qPCR was performed using SYBR Green. The amount of cDNA loaded was normalized to starting RNA concentrations, with a final concentration of 9 ng (40 ng in ZnT experiments) of RNA loaded per experimental well. Six-point standard curves were generated for each primer using 1:2 dilutions of cDNA. cDNA for the genes below were amplified using the indicated primers (IDT).:
MCOLN1forward, 5′-TCTTCCAGCACGGAGACAAC-3′; reverse, 5′-AACTCGTTCTGCAGCAGGAAGC-3′. MT2a: forward, 5′-AAGTCCCAGCGAACCCGCGT-3′; reverse, 5′-CAGCAGCTGCACTTGTCCGACGC-3′. MTF-1: forward, 5′-GCCATTTCGGTGCGATCACGAT-3′; reverse, 5′-TTTCACCAGTATGTGTACGAACGTGAGT-3′. ZnT2 (SLC30A2): forward, 5′-GCAATCCGGTCATACACGGGAT-3′; reverse, 5′-CAGCTCAATGGCCTGCAAGT-3′. ZnT4 (SLC30A4): forward, 5′-CACATACAGCTAATTCCTGGAAGTTCATCT-3′; reverse, 5′-GCCTGTAACTCTGAAGCTGAATAGTACAT-3′.
All primers were designed to span exons and negative reverse transcription controls were tested to ensure amplification of cDNA only. qPCR was performed using the Standard Curve method on the 7300 Real Time System (Applied Biosystems). Reactions were run with the following parameters: 2 min at 50°C, 10 min at 95°C, and 40 cycles at 95°C for 15 s followed by 60°C for 1 min. All experimental samples were run in triplicate and normalized to a β-actin endogenous control.
Cells were seeded on glass coverslips and loaded with dyes for 15 min at 37°C in buffer containing 150 mM NaCl, 5 mM KCl, 1 mM CaCl2, 1 mM MgCl2, 10 mM Hepes, pH 7.4, and 1 g/l glucose. The loading was followed by 15 min washout in all cases except TSQ (6-methoxy-8-p-toluenesulfonamido-quinoline). LysoTracker™ Red DND-99, FluoZin-3,AM (acetoxymethyl ester) and TSQ (Sigma) were used at 0.1, 2 and 150 μM respectively. Confocal microscopy was performed using Leica TCS SP5 and Bio-Rad 3000 confocal microscopes. Live cells were treated as above. For co-localization experiments, cells were fixed for 5 min in 4% paraformaldehyde at room temperature (23°C) and permeabilized using 0.1% Triton X-100 for 5 min on ice. Following washout and blocking in BSA and goat serum, cells were treated with primary anti-HA antibody (Roche, 12CA5) overnight. Cells were then incubated in fluorescent-tagged secondary antibodies for 1 h. Images were analysed using ImageJ (NIH).
β-Hexosamindase activity assay
Untreated control and TRPML1 siRNA-transfected cells were washed with 37°C-heated PBS and 1 ml of 37°C-heated serum-free DMEM was added to each 35 mm dish. For each sample, 100 μl of the supernatant was incubated with 400 μl of 1 mM 4-nitrophenyl N-acetyl-β-d-glucosaminide (Sigma, N9376) for 1 h at 37°C in 0.1 M citrate buffer (pH 4.5) (Sigma, C2488). This volume was replaced with fresh 100 μl of 37°C-heated DMEM to the dish. Samples were collected every 1 h for 4 h. Reactions were stopped by adding 500 μl of borate buffer (100 mM boric acid, 75 mM NaCl and 25 mM sodium borate, pH 9.8) and the absorbance was measured in a spectrophotometer at 405 nm. To determine total cellular content of β-hexosamindase, cells were lysed with 1 ml of 1% Triton X-100 and, after a 10000 g spin for 5 min at 4°C, 100 μl of the cell extracts was used for the enzyme activity reaction. Enzyme activity was determined as the amount of 4-nitrophenol produced per mg of protein (Bradford method). Absorbance was calibrated with different amounts of 4-nitrophenol (Sigma, N7660) in 0.1 M citrate buffer.
Zinc secretion assay
Control, TRPML1 and MTF1 siRNA-transfected cells were plated on a 12-well plate and, after 48 h, were pulsed with 100 μM zinc for 3 h, washed twice with warm PBS, and chased in 1 ml of DMEM per well. Duplicate samples were collected after 0, 5, 15, 30 or 60 min. For each sample, 50 μl of supernatant was placed in a 96-well plate. Zinc content was measured by incubating the supernatants with 10 μM cell-impermeant FluoZin-3 tetrapotassium salt (Molecular Probes, Invitrogen F-24194) for 15 min at 37°C. The 96-well plate was read via a FujiFilm FLA-5100 fluorescent image analyser. After the last time point, cells were washed with PBS, 200 μl of detergent solution was added to lyse the cells and fluorescence was normalized to total protein in each sample.
Statistical significance was calculated using a one-tailed unpaired t test with P ≤ 0.05 considered significant. Data are presented as means±S.E.M.
TRPML1, zinc and lysosomal enlargement
Since zinc dyshomoeostasis has been reported in TRPML1-KD cells , we sought to delineate mechanisms linking TRPML1 loss, zinc dysregulation and vacuolar enlargement in TRPML1-KD cells. In order to test how TRPML1 loss affects cells exposed to zinc, we used TRPML1-specific siRNA as described previously [30,31]. The siRNA-mediated KD resulted in more than an 85% KD of TRPML1 mRNA, as confirmed by qPCR with TRPML1-specific primers (Supplementary Figure S1 at http://www.biochemj.org/bj/451/bj4510155add.htm). In the previous studies utilizing the same system, the same siRNA resulted in almost complete elimination of TRPML1 protein, within the same time scale [30,31].
TRPML1-KD cells were previously reported to contain enlarged zinc-positive organelles . We performed live-cell confocal analysis of control and TRPML1-KD HeLa cells stained with the zinc-selective dyes TSQ and FluoZin-3,AM, in conjunction with the lysosomal dye LysoTracker™. Figure 1(A) shows that, although zinc is barely detectable in LysoTracker™-positive compartments in untreated control cells, such compartments displayed TSQ fluorescence in cells treated with 100 μM zinc for 48 h. We did not consistently detect enlarged LysoTracker™-positive compartments in TRPML1-KD cells until these cells were treated with zinc. Following zinc treatment, large compartments positive for zinc (TSQ) and LysoTracker™ were abundant in TRPML1-KD cells (Figure 1A). Figure 1(B) shows similar analysis of zinc in control and TRPML1-KD cells, performed using LysoTracker™ and another zinc dye, FluoZin-3,AM. The enlarged LysoTracker™-positive vesicles in zinc-treated TRPML1-KD cells also displayed bright FluoZin-3 fluorescence, consistent with high concentrations of zinc. Interestingly, FluoZin-3 signal was also detected in LysoTracker™-negative vesicles, consistent with our previous observations of non-lysosomal zinc pools in mammary cells [32,33], which was more commonly observed in zinc-treated TRPML1-KD cells than in zinc-treated control cells (Figure 1B). The lysosomal enlargement trend in zinc-treated TRPML1-KD cells persisted over a 96-h period (results not shown).
Zinc accumulation and enlargement of LysoTracker™-positive compartments in TRPML1-deficient zinc-treated cells
The fact that TRPML1-KD cells treated with zinc and stained with TSQ show cytoplasmic fluorescence (Figure 1A), whereas FluoZin-3,AM-stained cells do not (Figure 1B), can be explained by the nature of these dyes’ fluorescence and interaction with zinc. In the cytoplasm, zinc is bound to proteins such as MTs and very little zinc is present in free ion form. MTs bind large amounts of zinc (up to seven ions per molecule), with very high affinity (picomolar range) . Both TSQ and FluoZin-3 are sensitive to zinc in the mid- to high-nanomolar range. However, unlike FluoZin-3, which only fluoresces when bound to more labile zinc, TSQ fluoresces even when in a complex with metalloproteins such as MT–Zn [35,36]. Therefore the difference between TSQ and FluoZin-3 stains may reflect large amounts of Zn2+ ions bound in zinc-treated TRPML1-KD cells to cytoplasmic proteins such as MT2a.
Using LysoTracker™ fluorescence images, we calculated the average gain in the lysosomal size in zinc-treated TRPML1-KD cells. The same LysoTracker™ concentration, laser intensity, pinhole and gain values were used throughout the experiments. The greyscale LysoTracker™ images were filtered using 50% intensity threshold yielding binary black-and-white images, which were then treated using the Watershed segmentation algorithm implemented as an ImageJ plugin (Supplementary Figure S2A at http://www.biochemj.org/bj/451/bj4510155add.htm). Large clusters of signal not resolved using this algorithm were manually eliminated (Supplementary Figure S2B). Next, the ImageJ ‘Analyze particles’ algorithm was used, which yielded numbers and sizes of LysoTracker™-positive particles in each image (Supplementary Figure S2C). For particle counting, the following options were used: minimal size, 0.09 μm2; minimal circularity, 0.5. The same algorithm was used in all subsequent analysis in the present study.
Zinc-treated TRPML1-KD cells showed a tendency towards increasing the number of larger LysoTracker™-positive particles (Supplementary Figure S3 at http://www.biochemj.org/bj/451/bj4510155add.htm). In zinc-exposed TRPML1-KD cells, lysosomes were approximately 80% larger than in zinc-exposed control cells or in untreated TRPML1-KD cells (Figure 2). LysoTracker™-positive particles in zinc-treated TRPML1-KD cells averaged 176.15±12.50% (eight independent experiments, three to five images each, one to three cells on each image, 214–573 LysoTracker™-positive particles in each experiment, P<0.05 relative to control) of the average size of individual lysosomal areas in zinc-treated control siRNA-transfected cells (100±4%, eight independent experiments, three to five images each, one to three cells on each image, 99–1100 LysoTracker™-positive particles in each experiment). The use of threshold preserves the relative sizes of compartments, but makes evaluation of their absolute sizes unreliable. With this reservation in mind, the average size of LysoTracker™-positive compartment in zinc-treated control cells was 2.09±0.30 μm2 (statistics as above; two experiments were omitted from this analysis due to different microscope settings). In TRPML1-KD zinc-treated cells it was 3.56±0.51 μm2. Beyond agreeing with the previous findings on zinc build-up in TRPML1-KD cells , these data suggest that, in the absence of TRPML1, a zinc-dependent process leads to the enlargement of LysoTracker™-positive (ostensibly lysosomal) compartments.
Statistical analysis of lysosomal enlargement in TRPML1-deficient zinc-treated cells
Nickel, used at the same concentration, did not induce lysosomal enlargement (Supplementary Figure S4 at http://www.biochemj.org/bj/451/bj4510155add.htm), suggesting that this aspect of the TRPML1-KD phenotype is zinc-specific.
The above results confirm, in a different system, the previous findings on lysosomal enlargement and zinc retention in TRPML1-KD cells . A logical question following observation of this phenomenon is the source and means of zinc build-up in the lysosomes of TRPML1-KD cells. Since zinc is retained by the lysosomes in TRPML1-KD cells, TRPML1 must be involved in the clearance of zinc from the lysosomes. Where does the zinc retained in TRPML1-KD cells come from? Zinc retained by the lysosomes: (i) may come through endocytosis; or (ii) may be taken up by cells through cell-membrane transporters and then exported into lysosomes across their membranes. In both cases TRPML1 dissipates zinc accumulation. The difference between the two scenarios is that the latter entails: (i) a lysosomal zinc transporter that works in concert with TRPML1 to regulate zinc traffic between cytoplasm and lysosomes; and (ii) dependence of the lysosomal zinc accumulation in TRPML1-KD cells on a cytoplasmic step.
TRPML1 and MTF-1
Cytoplasmic zinc is tightly regulated, and therefore the expression of proteins involved in zinc maintenance is managed by zinc-dependent transcription factors such as MTF-1. Zinc binding to MTF-1, followed by MTF-1 nuclear import, leads to the expression of proteins involved in zinc handling, such as MTs. Their mRNA levels can thus be used to monitor cytoplasmic zinc levels.
Figure 3(A) shows a successful siRNA-mediated KD of over 90% of MTF-1 mRNA in untreated cells. This effect had a clear physiological significance, as MTF-1 KD severely suppressed mRNA levels of MT2a, which is the most ubiquitous and best described MT, well known to be regulated by MTF-1 (Figure 3B). In unstimulated cells, MTF-1 siRNA decreased MT2a mRNA levels to 19.12±5.71% of control levels (n=3, P<0.05). This established a toolkit for testing the role of MTF-1 in TRPML1-dependent effects of zinc.
MTF-1 is involved in the enlargement of LysoTracker™-positive compartments in zinc-treated TRPML1-deficient cells
MTF-1 suppression eliminated the build-up of enlarged vacuoles in TRPML1-KD cells (Figures 2 and 3C, and Supplementary Figure S4). In the double TRPML1/MTF-1-KD zinc-treated cells, the average lysosomal area was 101.78±7.06% (three independent experiments, three to four images each, one to three cells on each image, 127–1035 LysoTracker™-positive particles in each experiment) of its value in zinc-treated control cells, a statistically indistinguishable value. These data suggest that these TRPML1-dependent aspects of zinc phenotype in HeLa cells are mediated by MTF-1. The fact that suppression of a zinc-dependent step in the cytoplasm eliminates lysosomal enlargement and zinc retention in TRPML1-KD cells strongly suggests that the source of zinc retained in TRPML1-deficient lysosomes is not mediated through endocytosis, but is derived directly from a cytoplasmic route. This route is discussed below. We asked whether dysregulation of this route affects cytoplasmic zinc and which zinc transporters are involved in this process.
TRPML1 and MT2a
As discussed above, the presence of a cytoplasmic zinc- and TRPML1-dependent process suggests that the expression of zinc-dependent genes in cells exposed to zinc is also affected by TRPML1 activity. In this regard, MTF-1/zinc-dependent genes can be used as cytoplasmic zinc reporters. In order to test this idea, we first screened mRNA levels of a well-known MTF-1-dependent gene product, MT2a. Figure 4 shows that up-regulation of MT2a mRNA in response to zinc is significantly augmented in TRPML1-KD cells, compared with cells transfected with control siRNA. In control cells, 48 h stimulation with zinc increased MT2a mRNA levels to 676.8±152.3% of control levels (n=3, P<0.05), whereas in TRPML1-KD cells, the response to zinc was significantly greater (2992.7±678.5% of control levels; n=3, P<0.05). We conclude that a zinc-handling deficit in TRPML1-KD cells leads to a persistent dysregulation of zinc management.
MT2a response to zinc changes is amplified in TRPML1-deficient cells
It should be noted that, in our system, the difference in MT2a mRNA response is only seen under zinc exposure (Figure 4) and not under the basal conditions. Therefore the up-regulation of MT2a mRNA is a reaction to the change in cytoplasmic zinc buffering in the absence of TRPML1. Together with the MTF-1-KD assays, these experiments established that the lysosomal zinc build-up and enlargement in TRPML1-KD cells depends on a cytoplasmic step, and that TRPML1 loss increases cytoplasmic zinc concentration (also consistent with the data shown in Figure 1). We propose that TRPML1 works in concert with a lysosomal zinc transporter to maintain normal zinc levels in the cytoplasm and in the lysosomes. The loss of such concerted activity leads to cytoplasmic and lysosomal zinc buildup. Next, we set out to identify such a transporter.
TRPML1 and ZnT4
Of the Slc30 (ZnT) family, ZnT2, ZnT3, ZnT4 and ZnT8 have been implicated in zinc loading within endocytic compartments [37–39]. However, ZnT3 and ZnT8 expression seems to be restricted to the brain and pancreas [40,41]. For this reason, ZnT2 and ZnT4 became the first targets of our investigation. Recombinant ZnT2 and ZnT4 both partially co-localized with TRPML1 (Figure 5), indicating that they might work in concert with TRPML1. If either ZnT2 or ZnT4 are necessary for the TRPML1-dependent aspect of zinc homoeostasis, then KD should affect the zinc phenotype in TRPML1-KD cells. The predicted direction of zinc transport through ZnT transporters is into the lysosomes. Therefore ZnT KD should negate the effects of TRPML1 KD. We chose to modulate ZnT levels using siRNA.
ZnT2 and ZnT4 co-localize with TRPML1
Figure 6(A) shows qPCR confirmation of ZnT2 and ZnT4 KD (over 70% and 95% control mRNA levels in untreated cells respectively). ZnT2 siRNA decreased ZnT2 mRNA levels to 29.88±21.31% of control levels (n=3, P<0.05), whereas ZnT4 siRNA decreased ZnT4 mRNA levels to 4.36±3.73% of control levels (n=3, P<0.05). siRNA-mediated KD of ZnT4, but not ZnT2, rescued lysosomal swelling and lysosomal zinc build-up in zinc-treated TRPML1-KD cells (Figures 6B and 6C). In the double TRPML1/ZnT4-KD cells exposed to zinc, the average lysosomal area decreased to approximately 50% of the value reported in TRPML1-KD cells. It averaged 89.06±7.08% of the lysosomal size in control untreated cells (four independent experiments). In double TRPML1/ZnT2-KD cells, it remained largely the same as in TRPML1-KD cells exposed to zinc (167.22±15.71% of control cells, four independent experiments, P<0.05). The fact that ZnT4 KD abolished lysosomal enlargement under TRPML1 suppression has a clear physiological significance, because it identifies the lysosomal zinc transporter that, in concert with TRPML1, forms a lysosomal ‘zinc sink’, which absorbs the cytoplasmic zinc (Figure 7). In direct agreement with this, zinc-treated TRPML1- and ZnT4-KD cells have large amounts of zinc within the cytoplasm compared with ZnT2 KD (Figure 6B). Whether or not this phenomenon is specific to HeLa cells remains to be established, but this is consistent with zinc accumulation in lysosomes in ZnT4-overexpressing mammary cells . We note that the triple TRPML1/ZnT2/ZnT4 KD was extremely cytotoxic even at the control conditions, which precluded its analysis.
ZnT4 KD rescues enlarged lysosomes under TRPML1 suppression
A model of the ‘zinc sink’
Zinc secretion and lysosomal zinc leak in TRPML1-KD cells
The results described above suggest that ZnT4 loads lysosomes with zinc and that, in the absence of TRPML1, zinc is trapped in the lysosomes. Furthermore, clearance of zinc from the lysosomes in TRPML1-KD cells was delayed compared with control cells (Supplementary Figure S5 at http://www.biochemj.org/bj/451/bj4510155add.htm). The reason for zinc retention in TRPML1-deficient lysosomes was the subject of our next set of experiments.
TRPML1 has been implicated in lysosomal secretion. Therefore it is possible that suppression of lysosomal secretion due to TRPML1 loss traps zinc in the lysosomes. Alternatively, it is possible that TRPML1 is a lysosomal zinc channel, responsible for the zinc leak from lysosomes into the cytoplasm. The role of TRPML1 in zinc secretion was analysed using two assays. The premise of this experiment was: if zinc retention in the TRPML1-KD cells is due to retarded secretion, then zinc secretion in the TRPML1-KD cell will be significantly lower than in control cells. This is directly addressed in Figure 8(A). We measured zinc release from zinc-preloaded control and TRPML1-KD cells using FluoZin-3 tetrapotassium salt. Cells, grown on 12-well plates, were incubated in culture medium with 100 μM zinc for 3 h (pulse). Next they were washed in PBS and placed into fresh culture medium (chase), whose samples were taken 0, 5, 15 and 60 min later. Samples were analysed using FluoZin-3 (cell-impermeant salt) fluorescence, which was normalized to the total protein content of the well containing the given sample of cells. Figure 8(A) shows that the zinc content of the medium during the chase phase was not lower, but indeed higher, in TRPML1-KD than in control cells.
Zinc clearance and secretion in TRPML1-KD cells
Next, total constitutive lysosomal secretion was recorded using β-hexosaminidase release over a 4 h time interval. A previously described protocol was used . Figure 8(B) shows that there was no detectable difference in constitutive exocytosis of β-hexosaminidase between TRPML1-KD and control cells. These data are consistent with similar secretion rates, but indicate a higher lysosomal zinc content in TRPML1-KD cells. They argue against a difference in secretion being the main cause of zinc retention in the lysosomes of TRPML1-KD cells.
In the course of the present study, we found that, in agreement with a previously study , there is a significant dysregulation of zinc handling in TRPML1-KD cells. Our findings provide further insight into this process by showing that TRPML1 loss affects processes beyond lysosomes, and by identifying some of the components of the lysosomal ‘zinc sink’ responsible for export of cytoplasmic zinc into the lysosomes.
We show that the lysosomal enlargement and zinc build-up are alleviated by suppression of the cytoplasmic transcription factor MTF-1. These data clearly establish that a cytoplasmic step is involved in the loading of lysosomes with zinc and their enlargement in TRPML1-KD cells exposed to zinc. They largely rule out endocytosis as the other source of lysosomal zinc retention. What is this step and what is the role of MTF-1 in this process?
We found that the loss of ZnT4 abolishes zinc build-up and lysosomal enlargement in TRPML1-KD cells exposed to zinc. The fact that ZnT2 does not appear to have an effect in our system is puzzling. It probably reflects the specifics of zinc handling by HeLa cells. ZnT2 expression requires several factors . It is possible that ZnT2 expression, or its response to zinc, is defective in HeLa cells. Furthermore, it is possible that ZnT2 and TRPML1 do not work in the same functional space. A more exciting possibility is that TRPML1 regulates localization, or activity of ZnTs, specifically ZnT2 (Figure 7). Accordingly, a dependence of ZnT2 activity and structural integrity on lysosomal pH was shown recently .
Finally, we demonstrate that zinc leak from the lysosomes into the cytoplasm is delayed in TRPML1-KD cells. It is unlikely and that such a delay is due to problems with zinc secretion in TRPML1-KD cells, since there was no reduction in zinc secretion in these cells compared with control cells (Figure 8). The lack of an effect of TRPML1 on β-hexosaminidase secretion in our system does not refute TRPML1's role in secretion. It is possible that TRPML1 is involved in another aspect of secretion, such as regulated secretion, which was beyond the scope of the present study.
Although MTF-1 was shown to regulate ZnT1 expression, no evidence of ZnT4 regulation by MTF-1 has been found to date. It is important to remember that MTF-1 regulates MT expression and that MTs are indispensable for zinc transport. Several lines of evidence show that MTs deliver and load zinc on the transporters and that, in the absence of MTs, zinc transport is severely compromised [44,45]. On the basis of this, we suggest that, in our system, MTF-1 KD affected zinc transport into lysosomes due to the loss of MT2a, and possibly other MTs. This probably affected the loading of lysosomal zinc transporters with zinc. We hypothesize that, as suggested previously , MT2a transfers cytoplasmic zinc on to ZnTs, such as ZnT4, promoting its export into lysosomes. Such a lysosomal ‘zinc sink’ may be a defence mechanism against deleterious zinc spikes during toxic levels of zinc exposure.
At the same time, MT2a mRNA response to zinc in TRPML1-KD cells is dramatically increased. This is consistent with increased cytoplasmic zinc levels in TRPML1-KD cells exposed to zinc. What is the role of TRPML1 in this process? The fact that TRPML1 loss leads to lysosomal zinc retention indicates that TRPML1 dissipates lysosomal zinc. TRPML1 may transport zinc from the lysosomes into the cytoplasm, as suggested by our data in Supplementary Figure S5. It is important to remember that the up-regulation of zinc chelation and extraction from cells persistently exposed to zinc is likely to lead to a fall in cytoplasmic zinc concentration after the initial spike induced by the addition of zinc. It is possible that the function of TRPML1 is to dissipate zinc accumulated in the lysosomes due to ZnT4 activity during zinc spike. TRPML1 loss would result in weak, but prolonged, zinc leak through other transporters leading to low, but chronic, zinc elevation, perhaps sufficient to chronically activate MTF-1.
The central finding of the present study is the identification of the source of zinc build-up in TRPML1-KD cells as cytoplasmic as well as the identification of components that are necessary for the TRPML1-deficient ‘zinc sink’ to function properly: ZnT4 and cytoplasmic proteins, such as the transcription factor MTF-1. These components are involved in the cellular response to exposure to zinc against the background of the loss of a lysosomal ion channel, whose function remains unknown to date. Our findings shed light on TRPML1 function and its importance for proper cell function and health. Since the loss of TRPML1 is the cause of MLIV, further studies will show whether or not dysregulation of cytoplasmic zinc in TRPML1-KD cells is a contributing factor in the key aspects of MLIV pathogenesis, such as the build-up of storage bodies or cell death. Beyond MLIV, the fact that TRPML1 is involved in zinc trafficking potentially affects other neurodegenerative diseases. Zinc is an essential transition metal that needs to be highly regulated and zinc dysregulation has been linked to neuronal death following a seizure or ischaemic episode [47,48], formation of β-amyloid plaques associated with Alzheimer's disease  and pancreatic β-cell degeneration . It will be interesting to see what the role of TRPML1 is in those processes.
Dulbecco's modified Eagle's medium
green fluorescent protein
mucolipidosis type IV
metal-regulatory-element-binding transcription factor-1
small interfering RNA
solute carrier family
transient receptor potential mucolipin 1
Ira Kukic designed, executed and analysed the experiments, and co-wrote the paper. Jeffrey Lee and Jessica Coblentz provided technical expertise at the early stages of the project. Shannon Kelleher co-wrote the paper and provided expertise in data interpretation. Kirill Kiselyov designed the experiments, analysed the data and co-wrote the paper.
We thank Dr Haoxing Xu, Dr Shmuel Muallem, Dr Bruce Pitt and Dr Ora Weisz for fruitful discussions.
This work was supported by the National Institutes of Health [grant numbers HD058577 and ES01678 (to K.K.) and HD058614 (to S.L.K.)].