Characterizing Mammalian Zinc Transporters Using an In Vitro Zinc Transport Assay

Summary

Zinc transport has proven challenging to measure due to the weak causal links to protein function and the low temporal resolution. This protocol describes a method for monitoring, with high temporal resolution, Zn²⁺ extrusion from living cells by utilizing a Zn²⁺ sensitive fluorescent dye, thus providing a direct measure of Zn²⁺ efflux.

Abstract

Transition metals such as Zn²⁺ ions must be tightly regulated due to their cellular toxicity. Previously, the activity of Zn²⁺ transporters was measured indirectly by determining the expression level of the transporter under different concentrations of Zn²⁺. This was done by utilizing immunohistochemistry, measuring mRNA in the tissue, or determining the cellular Zn²⁺ levels. With the development of intracellular Zn²⁺ sensors, the activities of zinc transporters are currently primarily determined by correlating changes in intracellular Zn²⁺, detected using fluorescent probes, with the expression of the Zn²⁺ transporters. However, even today, only a few labs monitor dynamic changes in intracellular Zn²⁺ and use it to measure the activity of zinc transporters directly. Part of the problem is that out of the 10 zinc transporters of the ZnT family, except for ZnT10 (transports manganese), only zinc transporter 1 (ZnT1) is localized at the plasma membrane. Therefore, linking the transport activity to changes in the intracellular Zn²⁺ concentration is hard. This article describes a direct way to determine the zinc transport kinetics using an assay based on a zinc-specific fluorescent dye, FluoZin-3. This dye is loaded into mammalian cells in its ester form and then trapped in the cytosol due to cellular di-esterase activity. The cells are loaded with Zn²⁺ by utilizing the Zn²⁺ ionophore pyrithione. The ZnT1 activity is assessed from the linear part of the reduction in fluorescence following the cell washout. The fluorescence measured at an excitation of 470 nm and emission of 520 nm is proportional to the free intracellular Zn²⁺. Selecting the cells expressing ZnT1 tagged with the mCherry fluorophore allows for monitoring only the cells expressing the transporter. This assay is used to investigate the contribution of different domains of ZnT1 protein to the transport mechanism of human ZnT1, a eukaryotic transmembrane protein that extrudes excess zinc from the cell.

Introduction

Zinc is an essential trace element in the cellular milieu. It incorporates one-third of all proteins and is involved in various cellular processes, such as catalysis¹, transcription², and structural motifs³. However, despite being redox-inert, high zinc concentrations are toxic to the cell, which is why no mammalian organism has survived without the presence of mechanisms regulating zinc homeostasis. In mammals, three mechanisms are responsible for this process: (1) metallothioneins, which are cytosolic cysteine-rich proteins that bind zinc at a high affinity, thus preventing excess free cytosolic zinc⁴; (2) Zrt/Irt-like proteins (ZIPs), which are zinc transporters responsible for zinc influx into the cytosol through the plasma membrane or from intracellular organelles^4,5,6,7,8; and (3) ZnTs, which are a mammalian subset of the ubiquitous cation diffusion facilitator (CDF) family and are zinc transporters, as they extrude zinc from the cytosol across the plasma membrane or into the intracellular organelles^4,5,6,7,8,9. Due to the importance of zinc to cellular metabolism, it is vital to understand cellular zinc dynamics.

Previous methods to assess zinc dynamics depended on assessing the expression levels of mRNA under different zinc conditions by correlating them with cellular zinc measurements of fixed tissues or cells^10,11,12. These methods include chemical detection and immunohistochemistry staining. However, these methods yield only indirect measures and, thus, determine only an offline correlation between intracellular zinc concentration and the expression of zinc transporters. Consequently, these methods cannot infer any parameters requiring high temporal resolution.

A more direct measurement of Zn²⁺ transport uses radioactive isotopes of zinc¹³. This method relies on the measurement of radiolabeled Zn²⁺ to monitor zinc transport and its kinetics. However, due to the importance of zinc to cellular homeostasis, multiple cellular processes regulate intracellular zinc concentration. Among these are extracellular binding and several transport systems that work in concert to maintain tight control of intracellular Zn²⁺ levels. The combination of these processes creates considerable background noise, which makes it difficult to test individual zinc-related transport functions.

This article demonstrates a method to directly monitor the zinc transport rate by measuring the intracellular free zinc concentration using a zinc-specific fluorescent dye, FluoZin-3. The dye has high specificity for Zn²⁺ and little interference from other divalent cations, such as calcium. In addition, in its ester form, it enters the cells by nonionic diffusion and is then trapped due to the activity of intracellular di-esterase. Thus, its fluorescence is correlated primarily with the free cytosolic zinc concentration. These experiments were conducted to study the structure-function relationship of zinc transporter 1 (ZnT1), a member of the ZnT family.

Protocol

1. Cell transfection

1. Culture HEK293T cells in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 2 mM L-glutamine, and 1x penicillin/streptomycin (see Table of Materials) in a humidified incubator at 37 °C/5% CO₂ until confluence on a 10 cm plate (8.8 x 10⁶ total cells).
2. Place one 13 mm coverslip in each of the wells of a 12-well plate. Dilute 0.44 x 10⁶ trypsinized cells from step 1.1 in 12 mL of complete DMEM. Mix well by pipetting up and down three to five times. Fill each well with 1 mL of the mixed solution. Grow in a humidified incubator at 37 °C/5% CO₂ overnight.
3. Replace the complete DMEM in each well with 1 mL of serum-free DMEM (see Table of Materials) in each well. Return the 12-well plate to the humidified incubator as in step 1.1.
4. For each transfection well, dilute 1 µg of pAAV2 plasmid containing the protein of interest tagged with mCherry fluorescent protein (Supplementary File 1) in 100 µL of serum-free DMEM in a 1.5 mL tube.
5. Add 3 µg of polyethyleneimine (PEI, see Table of Materials) to 100 µL of serum-free DMEM per transfection, with each transfection in a different 1.5 mL tube. The plasmid:PEI ratio should be 1:3 (µg:µg).
6. Vortex both tubes from step 1.4 and step 1.5 for 10 s, and leave to rest at room temperature for at least 5 min.
7. Mix one part (100 µL per well) of the DNA solution from step 1.4 with one part (100 µL per well) of the PEI solution from step 1.5. Vortex the final solution for 10 s, and leave it to rest at room temperature for at least 20 min and for up to 6 h.
8. Take the 12-well plate from the incubator. Vortex the final solution from step 1.7, and add 200 µL to each of the wells in the 12-well plate from step 1.3. Return the 12-well plate to the humidified incubator (step 1.1).
9. After 3 h, replace the medium in each well with 1 mL of complete DMEM. Return the 12-well plate to the humidified incubator as in step 1.1.
10. After 2 days, take the transfected cell culture from the 37 °C/5% CO₂ incubator, and place it on an inverted fluorescence microscope using 10x magnification.
11. Using the microscope focus wheel, focus on the cells while using brightfield light.
12. Switch to the fluorescent mCherry excitation (587 nm) and emission (610 nm) wavelengths, and turn off the brightfield light. Check for the fluorescence of the cells to confirm the expression of ZnT1 mCherry.
13. Prepare the dye loading solution.
    1. Take an aliquot of 4 µL of zinc-specific fluorescent dye (see Table of Materials) in its acetoxymethyl (AM) ester form, dissolved in DMSO (1 µg/µL), from a light-protected stock stored in a −20 °C freezer. This form allows the dye to enter the cell through the plasma membrane by simple diffusion.
    2. Add 4 µL of 10% pluronic acid (or 2 µL of 20% pluronic acid, see Table of Materials) to the aliquot from step 1.13.1. Mix the solution well by pipetting up and down three times, and then add all 8 µL to a 1.5 mL tube.
    3. Add 750 µL of Ringer's solution (prepared in-house, see Table of Materials for composition) supplemented with 1 mg/mL (~0.1%) bovine serum albumin to the tube from step 1.13.2, and vortex vigorously. Add another 750 µL of the same solution, and vortex again to ensure maximum mixing.
14. Add 750 µL of the final solution from step 1.13.3 to two wells in a new 6-well culture plate. Cover with aluminum foil.
    NOTE: To avoid bleaching, the 6-well plate is always covered with aluminum foil from this step.
15. Using a fine tweezer, take up to four replicate coverslips from the transfected cell culture plate, and place them in the first filled well (up to four slides per well) of the new 6-well plate from step 1.14. Repeat this process for the second filled well.
    NOTE: All the coverslips in the same filled well must be from the same condition. However, each well can contain different conditions.
16. Cover with aluminum foil, and gently shake for 15-20 min.
17. Remove the dye loading solution, and replace it with a new washing solution of Ringer's plus albumin, as used in step 1.13.3. Cover with aluminum foil. Leave to shake again for 20 min. This allows the cleavage of the AM ester by the intracellular esterases.

2. Microscope preparation

1. Arrange the necessary tools (see Table of Materials) as mentioned: an inverted fluorescence microscope capable of detecting GFP and mCherry fluorophores, a perfusion system that allows switching between at least two solutions, a suction system, and a perfusion chamber.
2. Turn on the microscope, its light source, and the camera. Turn on the suction system.
3. Wash the perfusion system. Wash the first chamber with Ringer's solution and the second chamber with Ringer's solution containing 7 µM zinc solution supplemented with 7 µM pyrithione (zinc ionophore, see Table of Materials). To avoid air bubbles or gaps, leave some liquid in each chamber.
   NOTE: Since both containers are connected to the same tube hooked to the perfusion chamber, always ensure that the shared segment is washed with Ringer's solution.
4. Close the taps, and fill the appropriate chambers with Ringer's solution and Ringer's zinc solution, as in step 2.3.

3. Sample preparation

1. Take the perfusion chamber, and place it with the narrow side of the groove facing upward.
   NOTE: The groove is a hole in the middle of the perfusion chamber. It allows the perfusion fluid access to the cells. One side of the groove is narrow, and the opposite side is wide.
2. Apply a sealing silicone (see Table of Materials) around the groove. Clean any sealing silicone from the groove using a pipette tip.
   NOTE: Ensure there is enough silicone seal around the groove to seal a 22 mm coverslip.
3. Using a fine tweezer, take a coverslip from the washing solution, and place it on top of the groove with the cells facing down. This way, the cells will be exposed to the solution perfusing the groove during the experiment.
4. Place a 22 mm coverslip on top of the 13 mm coverslip, and tighten it using the tweezer. Take care not to crack the coverslips.
5. Flip the chamber, and press on it to release all the present liquids. Fill the groove with 100 µL of Ringer's washing solution, and press again to ensure no leakage.
   NOTE: If a leakage is detected, apply a sealing agent at the site of the leakage, and retest.
6. Mount the perfusion chamber onto the platform, and secure it. Place the perfusion and suction tubes to allow perfusion over the cells in the groove.
7. Change the perfusion rate to approximately 2 mL/min, and turn on the perfusion of the Ringer's solution. Ensure the perfusion system is working with no leakage or spillover.

4. Measurement preparation

1. Open the imaging software (see Table of Materials) by double-clicking on the icon. Log in with the relevant credentials. Choose the attached camera, and press Ok.
2. Set the microscope magnification to 10x using the button on the left side of the microscope.
3. Choosing live view and using the joystick, move the platform to focus on the cells in the groove.
4. While the perfusion is on, turn off the lights and change the wavelength to mCherry by pressing the dedicated button in the interface. Adjust the focus using the microscope focus wheel.
5. Once the microscope is focused on the cells, move the platform using the joystick to allow the selection of the appropriate cell patches.
   NOTE: A suitable area is considered an area with at least 10 cells for an ROI and an empty area for background.
6. Select the Turn ROIs On/Off drop-down menu, and choose Draw Circular ROIs.
7. Draw ROIs of cell clusters with the mCherry-expressing cells (indicates the expression of ZnT1).
8. From the same toolbar as step 4.6, click the Turn background ROIs On/Off button, and adjust the location and size of the background ROI to an area with no cells at all.
   1. Check with the mCherry and EGFP wavelengths to ensure no cells are in the selected background ROI.
      NOTE: The wavelengths were adjusted so that mCherry used an excitation wavelength of 520 nm and an emission wavelength of 610 nm and EGFP used an excitation wavelength of 470 nm and an emission wavelength of 520 nm.
9. Select the Wavelength (λ) sub-tab. Mark its checkbox, and ensure that the GFP wavelength is present and its checkbox marked. If not, add and mark it.
10. Select the Duration sub-tab. Define the measurement interval as every 5 s, and change the number of intervals to suit the experiment duration.
11. In the Focus section on the microscope panel below the eyepiece, click on the On button to enable the perfect focus system (PFS).
12. Using the PFS focus wheel, adjust the focus. In the software interface, Under ND Acquisition, ensure PFS on is ticked.
13. Click on Run now.

5. Experimental procedure

1. Start with a 90 s baseline period measurement using Ringer's solution perfusion.
2. After the baseline period ends, turn off the Ringer's solution perfusion, and then turn on the Ringer's zinc solution perfusion. Mark the switching of the solution by clicking the red flag on the bottom right of the main interface panel. A rise in fluorescence is expected to appear.
3. Once the fluorescence rise starts to saturate, change back to perfusion with Ringer's solution. Turn off the Ringer's zinc solution perfusion, and then turn on the Ringer's solution perfusion.
4. Wait until the experiment time expires. A noticeable yet steady decrease in fluorescence is expected if the ZnT1 is working properly.

6. Data export

1. To export the data with baseline subtraction, press the Subtract Baseline button, and view the changes in the "ND acquisition window".
2. Click on the Export button in the software interface in the "ND acquisition window". An excel datasheet will open. Save to the desired location.

7. Data analysis

1. For each ROI, create an average baseline fluorescence from the first 90-100 s.
2. Express the fluorescence calculated for each ROI as a percentage of the background for that ROI.
3. Create a row average of all the ROIs, and plot the result as a line graph. This creates an average of all the ROIs as a function of time.
4. Using the linear fit function, select the initial rate for the decrease in fluorescence following the wash with Ringer's solution. The slope value of the equation correlates with the transport rate.

Representative Results

ZnT1 is a mammalian zinc transporter located on the cell plasma membrane¹³. It is a member of the cation diffusion facilitator (CDF) protein family that extrudes zinc from the cytosol to the extracellular millieu¹⁴. ZnT1 has a two-domain architecture: the transmembrane domain, which transports the ions across the membrane, and a C-terminal domain¹⁴. Unlike other known CDF proteins, ZnT1 has an extended unstructured C-terminal domain (USCTD). The role of the USCTD is currently unknown. We used the protocol above to compare ZnT1 WT to ZnT1 without the USCTD to assess its involvement in zinc transport activity.

HEK 293T cells were transfected with the pAAV2 plasmid containing either ZnT1 WT or ZnT1 without the unstructured C-terminal (ΔUSCTD) and tested the transport activity rate as described above. Supplementary Figure 1 shows that both versions of ZnT1 express and localize to the plasma membrane without any issues. Figure 1 depicts the typical results for an example of such an experiment. An increase in fluorescence is the result of an increase in the intracellular Zn²⁺ concentration, while a decrease is a result of the activity of ZnT1 transporting Zn²⁺ across the cell membrane (individual graphs, including standard errors, are found in Supplementary Figure 2 and Supplementary Figure 3). It is evident from this figure that ZnT1 WT has a smoother fluorescence curve than the mutant. However, this is not an inherent differentiating feature but is related to variability in the cellular responses to zinc load, which has been observed in previous experiments. Figure 2 shows a box plot summarizing the results from 15-17 such experiments. From the figure, it is clear that ΔUSCTD presents a wider spread of transport rates compared to WT. While this can be attributed to the cellular response variability previously mentioned, it could also indicate a functional difference. For example, it may imply that the ZnT1 USCTD segment is a modulator of the transport rate. Clearly, there is no visible difference between the Zn²⁺ extruding activities of WT and ΔUSCTD. Based on the statistical analysis, the data were not normally distributed (WT dataset, Shapiro-Wilk test¹⁵, statistic = 0.72404, p value = 0.0004), so two sample non-parametric tests were used for comparisons between the datasets. The results showed that there was no statistical difference between the transport rates (Mann-Whitney¹⁶ U test: U = 132, Z = 0.15105, Asymp. Prob>|U| = 0.87994; Kolmogorov-Smirnov¹⁷ test: D = 0.27059, Z = 0.76384, Exact Prob>|D|= 0.5161).

Due to the substantial energetic cost of creating the unstructured extension (~80 amino acids), it is reasonable to assume that this domain serves a cellular function. However, a cellular function linked to this domain has not been identified so far. This domain is unique to ZnT1 and is not found in any other members of the ZnT family. As such, it may be due to the fact that, unlike other ZnTs, ZnT1 is located at the plasma membrane, while except for ZnT10, all other members are expressed in internal organelles.

The raw data and subsequent graphs shown here are based on 5 s interval measurements of the fluorescence caused by the zinc-sensitive dye. This means that in a relatively short time frame (minutes), a visualization of time-dependent zinc transport can be generated with high temporal resolution. The analysis should include a baseline period normalization since the dye is loaded into the cells prior to zinc loading and may have initial intracellular fluorescence caused by a small amount of free zinc. In addition, the zinc dye only becomes fluorescently active after cleavage by the intracellular esterase. Therefore, it is recommended to allow some time for the de-esterification process to occur for most of the dye in order to avoid irregular fluorescence patterns caused by dye ester cleavage throughout the experiment.

Figure 1: Zn²⁺ extruding activities of WT and ΔUSCTD using the zinc transport assay. The x-axis is the time in seconds. The y-axis is the fluorescence change expressed as a percentage of baseline. The line graphs are the fluorescence change as a function of time. The data are averaged over >10 ROIs. Individual graphs with standard error are shown in Supplementary Figure 2 and Supplementary Figure 3. Please click here to view a larger version of this figure.

Figure 2: Comparison of the activities of ZnT1 WT and ΔUSCTD depicted as a box plot. Each data point represents a single experiment. The black line indicates the median score. The hollow square box indicates the mean score. The y-axis indicates the change in baseline percentage per second. Please click here to view a larger version of this figure.

Supplementary Figure 1: HEK 293T cells transfected with (A) ZnT1 WT or (B) ZnT1 ΔUSCTD stimulated in the mCherry excitation (520 nm) and emission (610 nm) wavelengths. In both instances, there is expression and membrane localization of the ZnT1 variant. The microscope magnification is 20x. The exposure time is 100 ms. Please click here to download this File.

Supplementary Figure 2: ZnT1 WT Zn²⁺ transporting activity with the standard error visualized. The x-axis is the time in seconds. The y-axis is the percentage fluorescence change from the baseline fluorescence. The black line is the mean fluorescence change. Please click here to download this File.

Supplementary Figure 3: ZnT1 ΔUSCTD Zn²⁺ transporting activity with the standard error visualized. The x-axis is the time in seconds. The y-axis is the percentage fluorescence change from the baseline fluorescence. The red line is the mean fluorescence change. Please click here to download this File.

Supplementary File 1: pAAV2 plasmid sequence. Please click here to download this File.

Discussion

The above-described method allows for the direct measurement of the intracellular zinc concentration with high temporal resolution. Compared to other methods, this method involving monitoring changes in intracellular Zn²⁺ can substantially decrease background noise. In addition, the dye's selectivity for zinc eliminates potential cross-interactions with other metal cations^18,19. Finally, its lack of immediate cytotoxicity enables the testing of live cellular processes¹⁹.

However, this method has certain limitations. First, detecting intracellular dynamics can be relatively more challenging with this protocol, since the dye used in this method is designed to be trapped in the cell but cannot be directed to regions or specific processes. Other fluorescent dyes can be directed to a specific area in the cells but, unfortunately, have a much lower dynamic range²⁰. Second, the exposure of the dye to light causes a decrease in fluorescence due to bleaching, which restricts the time of exposure to light and, consequently, the active window for running the experiment. Third, the dye loading in this protocol is cumbersome and time-consuming when compared to using other dyes, especially genetically encoded dyes²¹. Finally, even after cleaving the esters, the dye leaks, leading to a Zn²⁺-independent reduction in fluorescence²².

Several steps should be taken to ensure the experiment's success. First, since transfection can be unequally distributed, the selected cells should contain the protein of interest. To avoid dye bleaching, it is important to keep the loaded cells protected from light. When running the experiment, the perfusion rate should be sufficient to complete the experiment without causing the detachment of the cells from the coverslip. In the case of multiple runs, keeping the same reservoirs for the Ringer's solution and Ringer's zinc solution is recommended. The dye loading time and wash time can be adjusted according to initial experiments. Zinc loading into the cells via perfusion with pyrithione can take from 1 min to 5 min, as assessed by monitoring the Zn²⁺-induced fluorescence with time.

Despite these difficulties, this method allows us, for the first time, to study the dynamics of the Zn²⁺ extruding process. It simplifies the process and creates a direct way to better establish causal links compared to the previous methods^23,24.

Materials

Name	Company	Catalog Number	Comments
10 cm plate	greiner bio-one	664160
12-well cell culture plate	greiner bio-one	665180
13 mm coverslips	Superior Marienfeld	111530
22 mm cover slides	Superior Marienfeld	101050
6-well culture plate	greiner bio-one	657160
Bovine serum albumin	bioWorld	22070008
Calcium chloride anhydrous, granular	Sigma Aldrich	C1016	Concentration in Ringer solution: 1 mM
D-(+)-Glucose	Glentham Life Science	GC6947	Concentration in Ringer solution: 10 mM
Dubelco’s Modified Eagle Media (DMEM)	Sartorius	01-055-1A
Eclipse Ti inverted microscope	Nikon	TI-DH	Discontinued. Replaced by Eclipse Ti2
Fetal Bovine Serum (FBS)	Cytiva	SH30088.03
Fine tweezers	Dumont	0203-55-PS
Fluozin-3AM	Invitrogen	F24195
HyClone Penicillin-Streptomycin 100x solution	Cytiva	SV30010
LED illumination system	CoolLED	pE-4000
L-glutamine	Biological Industries	03-020-1B
Magnesium chloride hexahydrate	Merck	1.05833	Concentration in Ringer solution: 0.8 mM
N[2-Hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid] (HEPES)	Formedium	HEPES10	Concentration in Ringer solution: 10 mM
Neo 5.5 sCMOS camera	ANDOR	DC-152Q-FI
NIS-Elements imaging software	Nikon	AR
Pluronic acid F-127	Millipore	540025
Pottasium chloride	Bio-Lab	163823	Concentration in Ringer solution: 5.4 mM
Pyrithione	Sigma Aldrich	H3261	Concentration in Ringer zinc solution: 7 μM
Silicone Grease Kit	Warner Instruments	W4 64-0378
Sodium chloride	Bio-Lab	190305	Concentration in Ringer solution: 120 mM
Zinc sulfate	Sigma Aldrich	31665	Concentration in Ringer zinc solution: 7 μM