Systematic characterization of zinc in a series of breast cancer cell lines reveals significant changes in zinc homeostasis

Abstract

An optimal amount of zinc (Zn²⁺) is essential for proliferation of human cells; Zn²⁺ levels that are too high or too low cause cell cycle exit. Tumors of the breast have been characterized by high levels of total Zn²⁺. Given the role of Zn²⁺ in proliferation of human cells and elevation of zinc in breast cancer tumors, we examined the concentration of total and labile Zn²⁺ across a panel of five breast cancer cell lines compared to the normal MCF10A cell line. We found that three cell lines (MDA-MB-231, MDA-MB-157, and SK-Br-3) showed elevated labile Zn²⁺ in the cytosol, while T-47D showed significantly lower Zn²⁺, and MCF7 showed no change compared to MCF10A cells. There was no change in total Zn²⁺ across the cell lines, as measured by ICP-MS, but we did observe a difference in the cells ability to accumulate Zn²⁺ when Zn²⁺ in the media was elevated. Therefore, we examined how proliferation of each cell line was affected by increases and decreases in the media. We found striking differences, where three cancer cell lines (MDA-MB-231, MDA-MB-157, and MCF7) showed robust proliferation in high Zn²⁺ at concentrations that killed MCF10A, T-47D, and SK-Br-3 cells. We also discovered that four of the five cancer cell lines demonstrate compromised proliferation and increased cell death in low Zn²⁺, suggesting these cells may be addicted to Zn²⁺. Overall, our study suggests significant differences in Zn²⁺ homeostasis and regulation in different types of breast cancer cells, with consequences for both proliferation and cell viability.

Keywords: breast cancer cell lines; genetically encoded fluorescent reporter; proliferation; viability; zinc.

Figure 1

Quantification of cytosolic Zn²⁺ in selected breast cancer cell lines using the genetically encoded FRET sensor ZapCV2. A, selected breast cancer cell lines included in this study and their respective molecular subtypes. Each cell line was stably transfected with both NES-ZapCV2 and H2B-HaloTag. B, schematic of the ZapCV2 sensor illustrating that Zn²⁺ binding increases FRET from ECFP to cpVenus. The ratio of FRET to CFP is proportional to the Zn²⁺ concentration. C, a representative ZapCV2 calibration in MCF7 cells. Cells are imaged in Hepes-buffered Hank’s balanced salt solution (HHBSS) then treated with R_min solution containing the Zn²⁺ chelator TPA to acquire the FRET ratio of the fully desaturated sensor. The chelator is then washed out and an R_max solution containing buffered Zn²⁺ and pyrithione is added to acquire the FRET ratio of the fully saturated sensor. D, representative images of each phase of the calibration in MCF7 cells. R_rest corresponds to the FRET ratio of cells in HHBSS with no treatment. R_min corresponds to cells treated with 50 μM TPA. R_max corresponds to cells treated with buffered Zn²⁺ and pyrithione. The scale bar represents 20 μm. E, the calculated cytosolic Zn²⁺ for each cell line in nM. Box and whisker plots showing the distribution of individual data points, with a line at the median and whiskers extending to minimum and maximum values. Statistical significance was determined via Brown-Forsythe and Welch ANOVA test with Dunnett’s T3 comparison test (^∗∗p < 0.01; ^∗∗∗p < 0.001; ^∗∗∗∗p < 0.0001) for n ≥ 3 biological replicates. ECFP, enhanced cyan fluorescent protein; TPA, tris(2-pyridylmethyl)amine.

Figure 2

Breast cancer cells have variable amount of total zinc in different zinc media. A, ICP-MS analysis of total zinc pool in noncancerous MCF10A cells and different breast cancer cells grown for 48 h in minimal media (MM), zinc-rich (ZR) conditions 30ZR (MM + 30 μM ZnCl₂) and 150 ZR (MM + 150 μM ZnCl₂). Box and whisker plots showing the distribution of individual data points, with a line at the median and whiskers extending to minimum and maximum values. All measurements were performed in four biological replicates (N = 4). Significance was determined via ANOVA with Tukey’s multiple comparison test (^∗∗∗p < 0.001; ^∗∗∗∗p < 0.0001). Black and colored asterisks represent statistical significance of differences with respect to MCF10A MM condition and MM conditions of each cell types, respectively. MCF10A cells were all dead under 150ZR conditions after 48 h. B, table reflects the average values of total zinc pool (fg/cell) in each cell types. Errors represent SD from mean. ICP-MS, inductively coupled plasma mass spectrometry

Figure 3

Proliferation of breast cancer cells changes in response to the available Zn²⁺ in minimal media. Results of resazurin cell proliferation assay on MCF10A (A), MDA-MB-231 (B), MDA-MB-157 (C), SK-BR-3 (D), MCF7 (E), and T-47D (F) cells grown for 48 h in minimal media supplemented with varying levels of TPA and ZnCl₂. Box and whisker plots showing the distribution of individual data points, with a line at the median and whiskers extending to minimum and maximum values. Each dot represents one well of cells in a 96-well plate. The fluorescence intensity of resorufin in each well is normalized to the average fluorescence intensity of resorufin in the minimal media condition and statistical significance was determined with respect to the minimal media (MM) condition for each cell type. 3ZD = 3 μM TPA, 2ZD = 2 μM TPA, 30ZR = 30 μM ZnCl₂, 100ZR = 100 μM ZnCl₂, 150ZR = 150 μM ZnCl₂, 250ZR = 250 μM ZnCl₂, and 500ZR = 500 mM ZnCl₂. Statistical significance was determined viaBrown–Forsythe and Welch ANOVA with Dunnett’s T3 multiple comparison test (∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗∗p < 0.0001) for n = 8 to 16 wells per condition per experiment. Each experiment was repeated for 2 to 3 times separately. TPA, tris(2-pyridylmethyl)amine.

Figure 4

Viability of breast cancer cells changes with available Zn²⁺ in media. Results of live/dead assay on MCF10A (A), MDA-MB-231 (B), MDA-MB-157 (C), SK-BR-3 (D), MCF7 (E), and T-47D (F) cells grown for 48 h in media containing 1.5% chelex-100 treated serum (minimal media, MM) supplemented with 2 μM TPA (2ZD), 30 μM ZnCl₂ (30ZR), and 150 μM ZnCl₂ (150ZR). Each dot represents one well of cells in a 96-well plate. All measurements were performed in n = 8 to 16 wells per condition per experiment. Each experiment was repeated for two separate times for a total of 16 to 32 replicates. Statistical analysis was performed viaBrown–Forsythe and Welch ANOVA with Dunnett’s T3 multiple comparison test (Significance: ^∗p < 0.05, ^∗∗∗p < 0.001; ^∗∗∗∗p < 0.0001). Error bars represent SD. TPA, tris(2-pyridylmethyl)amine.

Figure 5

Proliferation of breast cancer cells changes in response to available Zn²⁺ in full-growth media. Results of resazurin cell viability assay on MCF10A (A), MDA-MB-231 (B), MDA-MB-157 (C), SK-BR-3 (D), MCF7 (E), and T-47D (F) cells grown for 48 h in full growth media supplemented with varying levels of TPA and ZnCl₂. Box and whisker plots showing the distribution of individual data points, with a line at the median and whiskers extending to minimum and maximum values. Each dot represents one well of cells in a 96-well plate. The fluorescence intensity of resorufin in each well is normalized to the average fluorescence intensity of resorufin in the full growth media condition. All measurements were performed in n = 8 to 16 wells per condition per experiment. Each experiment was repeated for 2 to 4 times separately and statistical analysis was performed viaBrown–Forsythe and Welch ANOVA with Dunnett’s T3 multiple comparison test (Significance∗∗∗∗ = p < 0.001, ∗∗∗ = p < 0.001, ∗∗ = p < 0.01, ∗ = p < 0.05.). TPA, tris(2-pyridylmethyl)amine.

Figure 6

Metallothionein (MT) expression in breast cancer cell lines following Zn²⁺ enrichment via western blotting. Representative western blots showing MT expression in MDA-MB-231 (A), MDA-MB-157 (B), SK-BR-3 (C), MCF7 (D), and T-47D (E) cells, compared to MCF10A cells cultured for 48 h in minimal media (MM) or MM supplemented with either 40 μM ZnCl₂ (MCF10A, 40ZR) or 150 μM ZnCl₂ (150ZR). MT band (∼15–40 kDa) intensities and total protein levels were quantified using ImageJ from MT immuno-stained and Ponceau-stained membrane images. For each lane, the integrated intensity of the MT band was divided by the total protein intensity to calculate the MT/total protein ratio. These ratios were then normalized to the average MT/total protein ratio of the MCF10A MM condition (technical replicates) within each western blot experiment. F, quantification of MT/total protein ratios normalized to the MCF10A MM reference across various breast cancer cell lines. Each data point represents an individual technical replicate. We have excluded technical replicates that lacked sufficient legibility for accurate intensity measurements from the analysis (labeled as x). Statistical analysis was conducted across all biological replicates viaBrown–Forsythe and Welch ANOVA with Dunnett’s T3 multiple comparison test (Significance ∗∗∗p < 0.001; n ≥ 2 biological replicates, technical replicates ≥ 4). Error bars represent SD.

Figure 7

Relative protein expression of metallothionein. (A), slc39A transporters (B), slc30A transporters (C) reported as log2 (fold change) in expression relative to the average across 375 cancer cell lines. Data from ref (42). Data are from 31 breast cancer cell lines (gray). Cell lines used in this study: MDA-MB-231 (cyan), MDA-MB-157 (blue), MCF7 (pink), and T-47D (green). Data from Table S2 of ref (42).