Molecular Insights into the Breast and Prostate Cancer Cells in Response to the Change of Extracellular Zinc

Zinc dyshomeostasis is manifested in breast and prostate cancer cells. This study attempted to uncover the molecular details prodded by the change of extracellular zinc by employing a panel of normal and cancerous breast and prostate cell lines coupled with the top-down proteomics with two-dimensional gel electrophoresis followed by liquid chromatography-tandem mass spectrometry. The protein samples were generated from MCF-7 breast cancer cells, MCF10A normal breast cells, PC3 prostate cancer cells, and RWPE-1 normal prostate cells with or without exogenous zinc exposure in a time course (T0 and T120). By comparing the cancer cells vs respective normal epithelial cells without zinc treatment (T0), differentially expressed proteins (23 upregulated and 18 downregulated in MCF-7 cells; 14 upregulated and 30 downregulated in PC3 cells) were identified, which provides insights into the intrinsic differences of breast and prostate cancer cells. The dynamic protein landscapes in the cancer cells prodded by the extracellular zinc treatment reveal the potential roles of the identified zinc-responsive proteins (e.g., triosephosphate isomerase, S100A13, tumour proteins hD53 and hD54, and tumour suppressor prohibitin) in breast and prostate cancers. This study, for the first time, simultaneously investigated the two kinds of cancer cells related to zinc dyshomeostasis, and the findings shed light on the molecular understanding of the breast and prostate cancer cells in response to extracellular zinc variation.


Introduction
Zinc (Zn 2+ ) is essential to life.It functions in the cell as a cofactor for well over 300 enzymes and as a structural component for approximately 10% of the human proteome (∼3000 proteins) [1].Consequently, the cell has developed an elaborate molecular network over the extensive evolutionary timeline to maintain zinc homeostasis.Any disruption of such a network may lead to zinc dyshomeostasis, resulting in health problems such as cancers.Breast cancer is the most common malignancy in females worldwide [2,3], and prostate cancer in males is the second and ffth highest in incidence and mortality, respectively [2].Both breast and prostate cancers are associated with intracellular zinc dysregulation.Breast cancer cells exhibit elevated intracellular zinc levels compared to their normal epithelial cells [4], while prostate cancer cells show decreased intracellular zinc levels compared to their normal counterparts [5].Such diametrically opposite zinc profles of breast and prostate cancer cells provide an avenue for understanding the role of zinc in these two types of cancer cells.
It is well documented that cellular zinc homeostasis is maintained by Zrt/Irt-like protein (ZIP), Zn 2+ transporter (ZnT), and metallothionein (MT) [6][7][8].ZIP family contains 14 members, ZIP1-14 encoded by SLC39A1-14.Tey increase the cytoplasmic zinc level by importing zinc from the extracellular space or the intracellular organelles/vesicles into the cytoplasm.In contrast, ZnT family, which has 10 members as ZnT1-10 encoded by SLC30A1-10, reduces cytoplasmic zinc by exporting cytoplasmic zinc out of the cell or into the lumens of intracellular organelles.MT family bufers cytoplasmic zinc to maintain zinc homeostasis [9].Te elevated accumulation of intracellular zinc in breast cancer cells or the reduced intracellular zinc in prostate cancer cells is associated with the dynamic expression of ZIP, ZnT, and MT [10,11].Previous studies demonstrated that the extracellular zinc exposure resulted in the elevation of intracellular zinc [12][13][14][15].Terefore, this study attempts to prod the molecular machinery for zinc homeostasis into action by applying the extracellular zinc exposure and then uncover the dynamic changes by the proteomic approach.As intracellular zinc levels are fuctuating in the cells of living human beings, the dynamic changes in the proteomes of breast and prostate cancer cells are indeed relevant to our understanding of the zinc homeostasis in cancer cells.
Proteomics, complementary to genomics, is an established and essential platform for cancer research [16].Proteomic analysis on breast and prostate cancer tissues or cell lines or biological fuids from the cancerous individuals was employed in previous studies for the discovery and validation of the predictive, diagnostic, and prognostic markers [17][18][19][20][21][22][23][24][25][26].Diferential protein profles have been generated by the proteomics approach employing normal tissues and malignant tissues of low-or high-grade cancers [23].Comparative proteome analysis reveals changes in the proteins associated with metabolism [20,27], drug resistance, and metastasis of breast and prostate cancer cells [28,29].However, the proteomic profling has not been simultaneously carried out thus far in normal and cancerous cells of breast and prostate with or without extracellular zinc manipulation.Proteomic insights might be gained by investigating these two types of cancer cells with extracellular zinc manipulation which could prod the cells to action in response to the change of extracellular zinc.
In this study, the top-down proteomic analysis, by twodimensional gel electrophoresis (2-DE) coupled with liquid chromatography-tandem mass spectrometry (LC-MS/MS), was systematically carried out on MCF-7 breast cancer cells and MCF10A normal breast epithelial cells, PC3 prostate cancer cells, and RWPE-1 normal prostatic epithelial cells, with and without exogenous zinc exposure.Te following comparisons were performed in the data analysis: (1) the cancer cells vs the corresponding normal cells without zinc treatment (T 0 ) including MCF-7 cells vs MCF10A cells and PC3 cells vs RWPE-1 cells, (2) the cancer cells vs the respective normal cells with exogenous zinc treatment for 120 min (T 120 ) including MCF-7 cells T 120 vs MCF10A cells T 120 and PC3 cells T 120 vs RWPE-1 cells T 120 , (3) comparison of cancer cells between T 120 and T 0 including MCF-7 cells T 120 vs MCF-7 cells T 0 and PC3 cells T 120 vs PC3 cells T 0 , and (4) comparison of the normal cells between T 120 and T 0 including MCF10A cells T 120 vs MCF10A cells T 0 and RWPE-1 cells T 120 vs RWPE-1 cells T 0 .Such detailed comparative analyses revealed diferential protein expression profles of breast and prostate cells in the context of with or without extracellular zinc treatment, which provides signifcant insights and enhances our understanding of the breast and prostate cancer cells in response to extracellular zinc variation.

Materials and Methods
All the chemicals and reagents were of the highest purity grade from commercial providers as indicated in the methods.All the cell lines were purchased from American Type Culture Collection (ATCC, USA). 4 -Treated and Untreated Cells.Breast cells (MCF10A, MCF-7) and prostate cells (RWPE-1, PC3) were cultured in their standard growth media and condition described previously [10].According to the formulations of the media used here and the available data for the zinc contents in foetal bovine serum [30] and horse serum [31,32], the base level of zinc for the complete DMEM and RPMI 1640 media is approximately 5 μM, that for the complete DMEM/F12 is approximately 2 μM, and that for the complete keratinocyte serum free medium is 0.5 μM.Te mild cytotoxic dosage of ZnSO 4 for each cell line was determined by treating the cells with the individual dosages of ZnSO 4 including 0, 20, 50, 100, 150, 200, 250, 300, 350, 400, and 500 μM as described in the previous studies [10,14].Each dosage was the fnal concentration of ZnSO 4 , which was carried out by adding 10 μL of the 20X ZnSO 4 stock to the culture well containing 7000 cells in 190 μL medium (the fnal volume per well was 200 μL).Te mild cytotoxic dosage for ZnSO 4 was defned as the dosage which resulted in above 70%-85% cell viability at the end of 2 h zinc sulfate treatment.In this study, we used mild cytotoxic ZnSO 4 dosages of MCF-7 (320 μM), MCF10A (195.5 μM), PC3 (110 μM), and RWPE-1 (186.88 μM) cells for zinc treatment in proteomic analysis.Te rationale for selecting the mild cytotoxic dosages of ZnSO 4 is to obtain the datasets on diferentially expressed proteins prodded by the dosages without severely compromising the overall health of the cells in the culturing fasks of this study.Te viability of cells between 70% and 85% is ideal here, which allows the fndings to be relevant to the physiological state of the cells and provides maximum data possible.MCF-7, MCF10A, PC3, and RWPE-1 cells were grown in 75 cm 2 fasks until achieving ∼80% confuency and then the spent medium was aspirated and replaced with 11.9 mL of complete medium.ZnSO 4 at 120x stock concentration of each dosage for each cell line was prepared in sterile Milli-Q H 2 O (Milli-Q ® Advantage A10 Water Purifcation System, Merck, Australia).Te cells were treated with 100 μL of their respective ZnSO 4 stocks.Te control cells were treated with 100 μL of sterile Milli-Q water.Te cells were incubated for 120 min (T 120 ) and then the protein extraction was performed.Each treatment or control has three biological replicates, which 2

Cell Culture and Protein Extraction from ZnSO
Journal of Oncology means three protein samples for each time point of a given treatment or control.Each protein sample was prepared with three 75 cm 2 fasks of ∼80% confuence.Following the completion of incubation period, the medium was discarded, and the cells were washed and collected in 1x phosphate-bufered saline (PBS).Te cell pellets were obtained by centrifugation at 350 g for 3 min at 4 °C and washed with ice cold 1x PBS twice.Finally, the cell pellet was resuspended into 1 mL of ice cold 1x PBS and transferred into sterile 1.5 mL microfuge tubes.Te cells were centrifuged at 6000 rpm at 4 °C for 5 min and the supernatant was discarded.Te cells were snap frozen in liquid nitrogen and stored at −80 °C for protein extraction.
150-200 μL of total protein extraction bufer containing 8 M urea (Amresco, Solon, OH, USA), 2 M thiourea (Amresco, Solon, OH, USA), 4% CHAPS (Amresco, Solon, OH, USA), and 1x protease inhibitors (Sigma-Aldrich) was added to each cell pellet in a microfuge tube kept on ice.Te cells were then homogenised by using a probe sonicator (Across International, Australia) and centrifuged at 124436 g (SW 55 Ti rotor, Beckman Coulter, Indianapolis, IN, USA) at 4 °C for 1 h.Te supernatants were collected into individual tubes for either immediate analysis or storage at −80 °C.

Protein Quantifcation, Reduction, and Alkylation for 2D
Gel Electrophoresis.Te protein concentration for each sample was estimated using the EZQ ™ protein quantitation kit (Life Technologies, Eugene, OR, USA) according to the manufacturer's instructions.100 μg of each protein sample was taken in a sterile 0.65 mL microfuge tube.An equal volume of rehydration bufer (containing a mixer of carrier ampholytes (Bio-Lyte, Bio-Rad, Australia) at a fnal concentration 2% (v/v)) was added to each tube.Te sample was then mixed with 2.42 μL of reduction bufer (2 M DTT in 0.2 M TBP) and incubated at 25 °C for 1 h on a heating block (Dry Block Heater, Termoline Scientifc, Australia).Following the incubation, 5.1 μL acrylamide (5.6 M) was added to each protein sample for alkylation, vortexed, and incubated at 25 °C for further 1 h.Te protein samples were then ready for 2D separation.

First Dimension-Isoelectric Focusing (IEF)
. Te nonlinear 7 cm long immobilised pH gradient (IPG) strips with pH 3-10 gradients were hydrated with 125 μL of the above-treated protein sample (100 μg).Te IEF was then carried out by using Protean IEF apparatus (Bio-Rad, USA) with the following program: desalting at 250 V for 15 min, ramping up the voltage to 4000 V by a linear gradient for 2 h, keeping 4000 V constant for a total of 37500 Vh and then terminating the isoelectric focusing or holding at 500 V until the termination.Te temperature of the IEF apparatus during isoelectric focusing was 17 °C.Upon completion of the isoelectric focusing, the IPG strips were immediately subjected to the second dimension.

Second Dimension-Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDS-PAGE).
Te IPG strips from the frst dimension were incubated in 130 mM DTT in equilibration bufer for 10 min followed by 10 min alkylation with 350 mM acrylamide at room temperature on a gentle shaker.Instantly, the IPG strip was placed in warm agarose layer over the stacking gel (5%) which was above the resolving gel (12.5% (w/v) acrylamide, 1 mm thick 8.4 × 7 cm).Once the agarose layer was solidifed, the electrophoresis was carried out in 1x tris-glycine-SDS running bufer at 90 V and 4 °C for 3 h until the tracking bromophenol blue dye reached the bottom of the gel.Finally, the gels were taken out of the glass plates, promptly dipped into the fxatives (10% methanol with 7% acetic acid) for gel fxation, and then stained with colloidal Coomassie Brilliant Blue (cCBB) for 20 h, followed by destaining with 0.5 M NaCl thrice (15 min each).Te gels were imaged by FUJI LAS-4000 (GE Healthcare, USA).

Protein Spot Detection and Quantitative Analysis.
Te protein spots were detected and quantitatively analysed in the gel images by Delta2D (version 4.0.8,DECODON Gmbh, Germany) as described previously [33].Te protein spots in the gel images were quantitatively analysed in cancer cells compared to normal cells with or without zinc treatment.Similarly, the protein spots were analysed in each cell line with or without zinc treatment.In each comparison, the gel images were warped and fused to make master gel using "union fusion."Te spots were then transferred to each image in their group to ensure consistent spot matched (100% matching) in all biological replicates (n � 3) in each group.Te backgroundsubtracted spot volumes were described as grey values, fold changes, p values (t-test), and relative standard deviation (RSD).Based on p value (p < 0.05) and ratio of grey value, the candidate spots were considered for further proteolytic digestion and liquid chromatography-tandem mass spectrometry (LC-MS/MS) to identify the proteins.
2.6.Peptide Extraction and LC-MS/MS.Te selected protein spots were excised manually and digested with trypsin (Promega, USA) for 8 h at 4 °C.Te digested protein samples were analysed by LC-MS/MS (Mass Spectrometry Facility, Western Sydney University), using Waters nanoAcquity LC-MS/MSnanoACQUITY UPLC on a Xevo QToF mass spectrometer (Waters, USA) as described previously [34,35].Te protein identifcation was conducted employing ProteinLynx Global Server (PLGS) program (version 3.0 Waters Corporation, USA) and the UniProt (Homo sapiens, human) database with the following settings: (a) the allowed maximum missed cleavages was set to 1, (b) the allowed false discovery rate was set to 4% and the maximum protein size was set to 280 kDa, (c) the peptide modifcations were carbamidomethyl C (fxed) and oxidation M (variable), (d) the minimum fragment per peptide was 3, (e) the minimum peptide per protein was 1, and (f) the minimum fragment per protein was 7. Finally, the identifed proteins from each spot by LC-MS/MS had to meet the selection thresholds such as PLGS or protein score ≥200, sequence coverage ≥6%, and matched peptides ≥3.

Diferentially Expressed Proteins in Breast Cancer Cells (MCF
Based on the molecular functions as per literature survey and UniProt database, those 41 diferentially expressed proteins were classifed into three prominent groups, including catalytic enzymes (26%), metal ion binding proteins (16%), and molecular chaperones (11%) (Supplementary Figure 1a).PANTHER database-based protein classifcation agrees with the molecular function-based classifcation as catalytic enzyme (33%) and calcium binding-protein classes (13%) are the prominent ones (Supplementary Figure 1b).

Functional Interactions of the Diferentially Expressed
Proteins in Breast and Prostate Cells.By STRING functional protein-protein network analysis, both known and predicted functional interactions were revealed for the diferentially expressed proteins in both cancerous and normal breast and prostate cells under the experimental conditions with and without zinc exposure (Supplementary Figures 9 and 10).Heat shock protein 90 kDa α (cytosolic) class B member 1 isoform (HSP90AB1), actin cytoplasmic 1 (ACTB), and triosephosphate isomerase (TPI1) are prominent in the functional network derived from the comparison of breast cancer cells (MCF-7) and normal breast epithelial cells (MCF10A) without zinc exposure (Supplementary Figure 9a).Triosephosphate isomerase (TPI1) displays its prominence again in the functional network of the diferentially expressed proteins in MCF-7 T 120 compared to MCF10A T 120 under zinc exposure (Supplementary Figure 9b).Te metal ion binding proteins such as annexin A1 (ANXA1), annexin A5 (ANXA5), protein S100A2 (S100A2), and protein S100A13 (S100A13) are at the peripheral edge of the protein network of MCF-7 breast cancer cells without zinc exposure (Supplementary Figure 9a) and again in the network of the diferentially expressed proteins in MCF-7 with zinc exposure compared to MCF10A T 120 (Supplementary Figure 9a).Prohibitin (PHB) is prominent in the functional network of the diferentially expressed proteins in the prostate cancer cells (PC3) with zinc exposure (Supplementary Figure 10b), apart from the heat shock proteins encoded by HSPD1, HSPA1B, and HSPB1.Heat shock protein 90 kDa α (cytosolic) class B member 1 isoform (HSP90AB1), proteasome subunit α type 1 (PSMA1), elongation factor c (EEF1G), and 40S ribosomal protein SA (RPSA) are predominant in the protein network of the diferentially expressed proteins in PC3 cells without zinc exposure (Supplementary Figure 10a).

Discussion
Zinc dyshomeostasis is the hallmark of breast and prostate cancer cells.Numerous studies have focused on the zinc homeostasis of breast cancer cells or prostate cancer cells, although the current work is the frst to investigate these two kinds of cancer cells together in tandem.Furthermore, we 14 Journal of Oncology examined the total proteomic profles of breast cancer cells vs normal breast epithelial cells and prostate cancer cells vs normal prostate epithelial cells in the presence or absence of zinc exposure.Te diferentially expressed proteins with or without zinc exposure in breast cells (MCF-7, MCF10A) (Table 1) and prostate cells (PC3, RWPE-1) (Table 2) in this study are the key datasets, which enhances the understanding of the zinc homeostasis in both breast and prostate cancer cells.

Te Intrinsic Diferences between the Cancer Cells and Teir Normal Counterparts (without Zinc Exposure).
First, the analysis without extracellular zinc treatment demonstrates the intrinsic diferences between breast cancer cells MCF-7 and the normal breast epithelial cells MCF10A, as well as between prostate cancer cells PC3 and the normal counterpart RWPE-1 cells.Te proteomic results demonstrate a key feature of breast and prostate cancer cells, namely, the downregulation of tumour suppressors or antitumour proteins.Te results showed the reduction of tumour suppressor 14-3-3 protein σ and θ in MCF-7 and PC3 cancer cells compared to the normal counterparts (Tables 1 and 2), which is in agreement with the previous fndings [36,37].Te 14-3-3 proteins, including seven isoforms such as σ and θ, are associated with cell cycle, signalling, and apoptosis and are usually downregulated for cancer progression [36,37].Te tumour suppressor protein S100A2 was decreased in MCF-7 cells (Table 1) as previously reported [38,39].However, the expression of S100A2 was unchanged in PC3 cells (Table 2), which is consistent with the previous study [40].Antitumour proteins such as latexin, glutathione Stransferase P, Rho GDP-dissociation inhibitor 1, and serpin B5 were reduced in their expression in PC3 prostate cancer cells (Table 2), in agreement with the previous studies [41][42][43][44].Annexin A1 was found to be downregulated in MCF-7 and PC3 cancer cells (Tables 1 and 2), which is related to breast and prostate cancer development [45][46][47][48].Also, for the frst time, we observed a downregulated antitumour protein, glycine tRNA ligase [49], in PC3 prostate cancer cells but not in breast cancer cells.Te downregulation of glycine tRNA ligase could play a role in prostate cancer development.
Te proteomic results demonstrate another feature of breast and prostate cancer cells, that is, the upregulation of proteins related to cancer growth and metastasis.α-Smooth muscle actin (α-SMA) and tumour protein D53 (hD53) were overexpressed in MCF-7 cells (Table 1).α-SMA serves as the marker of epithelial-to-mesenchymal transition (EMT) for cancer metastasis [50,51] and hD53 promotes breast cancer cell proliferation and their expressions are correlated [51].High expression of F-actin-capping protein subunit β (CAPZB) in the breast cancer cells (Table 1) is linked with α-SMA in regulating breast cancer cell growth and motility [52,53].Overexpression of antioxidants in cancer cells enhances the cancer cell proliferation, hence cancer growth in patients.Peroxiredoxin 6, an antioxidant protein, promotes cancer cell proliferation in an oxidative stress environment [54,55].Tus, overexpressed peroxiredoxin 6 in MCF-7 cancer cells (1.8-fold, Table 1) and PC3 cancer cells (8.5-fold, Table 2) indicates its role in breast and prostate cancer development.Tis fnding also suggests that peroxiredoxin 6 (PRDX6) is a potential target for anticancer drug development.Glutathione S-transferase Mu 3 (GSTM3) is another antioxidant overexpressed in MCF-7 breast cancer cells (Table 1), while superoxide dismutase (SOD1) was overexpressed in PC3 prostate cancer cells (Table 2).D-3-Phosphoglycerate dehydrogenase, a metabolic enzyme, is involved in redox homeostasis [56].Its overexpression in MCF-7 breast cancer cells (Table 1) indicates that this enzyme is associated with breast cancer development.

Te Dynamic Expression of Proteins in Breast and Prostate
Cancer Cells in Response to Zinc Exposure.Te proteomic datasets were obtained by the comparison between breast cancer cells MCF-7 and the normal breast epithelial cells MCF10A in response to the change of extracellular zinc concentration, as well as the comparison between prostate cancer cells PC3 and the control cells in response to the change of extracellular zinc.Te analysis demonstrates that the cancer cells upregulated the proteins which are related to lysosomal activity, antioxidant activity, stress response, cancer growth, cellular structure, and metabolism.MCF-7 breast cancer cells showed overexpression of cathepsin D in response to zinc exposure (Table 1).Cathepsin D is an aspartic endoproteinase in lysosome and is well known for its roles in angiogenesis, proliferation, and invasion in breast cancer [66,67].Te change of extracellular zinc should lead to the elevation of cytoplasmic zinc in MCF-7 cells, which might in turn result in higher zinc level in lysosome and hence cathepsin D upregulation.Because zinc enhances cathepsin D activity in lysosome [68], the overexpression of this endoproteinase might be accompanied with increased proteinase activity in zinctreated MCF-7 cells.Interestingly, peroxiredoxin 6 was overexpressed only in PC3 prostate cancer cells under the zinc exposure, in contrast to its overexpression previously described in both MCF-7 and PC3 cells without zinc exposure.Additionally, peroxiredoxin 2 was also overexpressed in PC3 cells under zinc exposure.Te fndings demonstrate that peroxiredoxin 6 is related to the cancer development and stress response while peroxiredoxin 2 is likely more relevant to stress response.Antioxidant proteins, including glutathione S-transferase Mu 3 and mitochondrial NADH dehydrogenase (ubiquinone) ironsulfur protein 3, were overexpressed in breast cancer cells under zinc exposure (Table 1).A previous study showed that glutathione S-transferase Mu 3 expression has a positive relationship with zinc [69].Te molecular chaperones such as mitochondrial 60 kDa heat shock protein, heat shock 70 kDa protein 1B, and heat shock protein β1 were overexpressed in PC3 cancer cells upon zinc exposure (Table 2), which is likely a part of stress response for the prostate cancer cells.
Zinc enhances breast cancer growth.Tis is evidently supported by the increased intracellular zinc level in breast cancer cells compared to the normal breast epithelial cells [11,70,71].Te proteomic dataset showed the elevated expression of tumour protein D53 (hD53 encoded by TPD52L1) and tumour protein D54 (hD54 encoded by TPD52L2) of MCF-7 breast cancer cells in response to the change of extracellular zinc (Table 1), which explains to some extent why zinc promotes breast cancer growth.Tis fnding also suggests that hD53 and hD54 are potential targets for anticancer drug development against breast cancers.
Intriguingly, the change of extracellular zinc resulted in overexpression of prohibitin (PHB) in prostate cancer cells (PC3) (Table 2).Prohibitin can act as a tumour suppressor in prostate cancers [72].As is known, the intracellular zinc level in prostate cancer cells is lower than the normal counterparts [5,71].Te variation of extracellular zinc should lead to the increased level of zinc inside the PC3 cancer cells, which is detrimental to the prostate cancer cells.Te overexpression of prohibitin might partly explain the cytotoxicity of excess zinc for the prostate cancer cells.Moreover, the reduction of metabolic enzymes including D-3-phosphoglycerate dehydrogenase, adenylosuccinate lyase, inosine-5′monophosphate dehydrogenase, and translational elongation factor Tu under zinc exposure (Table 1) might be relevant to the decreased cell viability in MCF-7 breast cancer cells under zinc exposure [10], but the expression of these metabolic enzymes is not changed in PC3 prostate cancer cells.
Further proteomic analysis was also done by comparing breast cancer cells MCF-7 with and without zinc treatment, as well as comparing the prostate cancer cells PC3 with and without zinc treatment.Firstly, MCF-7 breast cancer cells exhibited 25 diferentially expressed proteins (Table 1) under zinc exposure compared to without zinc exposure (T 0 ), while PC3 prostate cancer cells showed only 9 diferentially expressed proteins (Table 2).Tis very fact demonstrates that breast cancer cells are more capable responders to the variation of extracellular zinc levels.Teir molecular network of zinc homeostasis might be more sophisticated than the one in prostate cancer cells.
Te fndings demonstrate that zinc upregulates the proteins related to breast cancer growth and metastasis.Zinc exposure upregulated actinin α1 and annexin A5 in MCF-7 cells (Table 1).Te cytokinetic protein actinin α1 is shown to promote tumorigenesis and epithelial-to-mesenchymal transition (EMT) in cancer via AKT/GSK3β/β catenin signalling pathways [73].Among 12 annexin A isoforms (annexin A1-11 and annexin A13), annexin A5 in particular has unphosphorylated short N-terminus which enables this protein to exhibit a wide range of functions such as signalling, cancer cell growth, and invasion [74].Te overexpression of both inorganic pyrophosphatase (PPA1) and tubulin α1c (TUBA1C) in response to the variation of exogenous zinc in MCF-7 cells (Table 1) suggests that high intracellular zinc promotes the metabolic activity of breast cancer cells, since inorganic pyrophosphatase is involved in cell metabolism, and tubulin α1c promotes glycolysis in breast cancer [75][76][77].In addition, current fnding demonstrates that heat shock 70 kDa protein was overexpressed in MCF-7 cells (Table 1), correlating well with its overexpression at the gene level [78].

Interactions of the Diferentially Expressed Proteins in
Cancer Cells.Human triosephosphate isomerase (TPI1) is a key glycolytic enzyme, and glycolysis is accelerated in cancer cells [79].Te prominence of triosephosphate isomerase in breast cancer cells (MCF-7) with and without zinc exposure (Supplementary Figures 9a and 9b) demonstrates that it is potentially associated with breast cancer development.Te marked upregulation of triosephosphate isomerase in MCF-7 cells without zinc exposure (T 0 ) and with zinc exposure (T 120 ) compared to the normal counterparts (Table 1) refects both its intrinsic expression in the breast cancers and dynamic zincresponsiveness upon zinc exposure.Triosephosphate isomerase was found to be involved in PI3K/AKT/mTOR signalling pathway and hence breast cancer development [64], which supports the signifcance of the fnding for triosephosphate isomerase in this study.Terefore, it is potentially a druggable target, and it is indeed under investigation for anticancer drug development [80].Te metal-binding proteins, S100A2 and S100A13, belong to S100 protein family, which were frst identifed by Moore in 1965 [81,82].Tere are 18 members of S100A (S100A1-S100A18).Protein S100A13 is present in the functional networks of MCF-7 with and without zinc exposure (Supplementary Figures 9a and 9b), and it was highly overexpressed both intrinsically in MCF-7 without zinc exposure and responsively to zinc exposure in MCF-7 cells.Te fndings suggest that S100A13 is involved in zinc homeostasis of breast cancer cells.Prohibitin (PHB) is a worthwhile target for future investigations according to its overexpression in the prostate cancer cells (PC3) at T 120 zinc exposure compared to the normal counterparts (Table 2) as well as its prominence in the functional protein network of PC3 T 120 vs RWPE-1 T 120 (Supplementary Figure 10b).Prohibitin is a pleiotropic chaperone/scafold tumour suppressor protein implicated in the regulation of cell proliferation and apoptosis [83].Tis study, for the frst time, demonstrated that it is also a zinc-responsive protein in the prostate cancer cells.

Conclusion
Te systematic approach of high-resolution top-down proteomics was carried out simultaneously, for the frst time, on the cancerous breast and prostate cells (MCF-7, PC3) and the normal breast and prostate cells (MCF10A, RWPE-1).Te datasets revealed the intrinsic diferences in the proteomes of cancer cells (MCF-7 and PC3) and their normal counterparts without zinc treatment, such as the downregulation of antitumour proteins (14-3-3 protein σ, protein S100A2, latexin, and annexin A1) and the upregulation of tumour protein (hD53), antioxidants (peroxiredoxin 6 and superoxide dismutase), and metabolic enzymes (dihydrolipoamide S-succinyltransferase and aldehyde dehydrogenase 1) in both breast and prostate cancer cells.Te zinc-responsive proteomes were then unravelled by their dynamic expressions prodded by the change of extracellular zinc, particularly observed were the increased expressions of tumour proteins (hD53, hD54) and triosephosphate isomerase in breast cancer cells.As the cytoplasmic zinc level is elevated in breast cancer cells, the overexpression of those zinc-responsive proteins could be involved in breast cancer development.Moreover, the upregulation of metal binding protein S100A13 likely plays a role in zinc homeostasis of breast cancer cells.Te overexpression of the tumour Journal of Oncology suppressor prohibitin (PHB) in prostate cancer cells (PC3) in response to the change of extracellular zinc provides an explanation for the inhibitory efect of zinc in prostate cancer development.Te upregulation of antioxidants in both kinds of cancer cells under zinc exposure, such as peroxiredoxin 6, would beneft cancer cell growth in response to the change of environmental conditions.Overall, the fndings here uncovered signifcant molecular targets for anticancer drug development and enhanced our knowledge as well as understanding of the role of zinc in breast and prostate cancer cells.

Figure 1 :
Figure 1: Diferentially expressed protein spots in 2-DE gels by comparisons of MCF-7 T 0 vs MCF10A T 0 and MCF-7 T 120 vs MCF10A T 120 .(a) Representative 2-DE gel images (in the left panel) of breast normal MCF10A cells (MCF10A T 0 ) and breast cancer MCF-7 cells (MCF-7 T 0 ) without zinc exposure (T 0 ).(b) Representative 2-DE gel images (in the right panel) of breast normal MCF10A cells (MCF10A T 120 ) and breast cancer MCF-7 cells (MCF-7 T 120 ) with exogenous zinc exposure for 120 min (T 120 ).Each protein extract (100 μg) was resolved based on isoelectric point (pI) and molecular weight (MW).Te diferentially expressed protein spots are shown with red circles denoting upregulation and green circles denoting downregulation.

Table 1 :
Identifed proteins in breast cancer cells (MCF-7) and normal breast epithelial cells (MCF10A) with or without exogenous zinc exposure.
2/0.0009) ↓ (0.2/0.0001) ↓ 7/0.04) ↓ III Note.MW stands for molecular weight, kDa for kilo Dalton, pI for isoelectric point, PLGS for ProteinLynx Global Server, T 0 for 0 min or without zinc exposure (control), T 120 for 120 min, ↑ for upregulation, ↓ for downregulation, MCF-7 for breast cancer cells, and MCF10A for breast normal epithelial cells.Te PLGS score, protein accession, theoretical MW/pI, matched peptides, and sequence coverage (%) were obtained using ProteinLynx Global Server (PLGS) software (version 3.0, Waters Corporation, USA) and the UniProt (Homo sapiens, human) database.Gene ID was derived from UniProt database.Te observed MW and pI were calculated according to the protein standards.Te fold changes and p values were acquired from the quantitative analysis of the gel images (each group n � 3) by Delta2D software (version 4.0.8,DECODON Gmbh, Germany).MCF-7 T 0 /MCF10A T 0 is the expression fold change of the proteins in MCF-7 cells compared to MCF10A cells without zinc exposure (T 0 ), MCF-7 T 120 /MCF10A T 120 is the expression fold change of the proteins in MCF-7 cells compared to MCF10A cells following the zinc exposure for 120 min (T 120 ), MCF-7 T 120 /MCF-7 T 0 is the expression fold change of the proteins in MCF-7 cells following zinc exposure for T 120 compared to T 0 , and MCF10A T 120 /MFC10A T 0 is the fold change of the proteins in MCF10A cells following zinc exposure for T 120 compared to T 0 .Molecular functions: I, apoptosis; II, signalling; III, catalytic activity; IV, RNA binding; V, protein synthesis; VI, protein binding; VII, structural; VIII, metal ion binding; IX, molecular chaperone; X, DNA binding; XI, DNA synthesis; XII, metabolism; XIII, lipid binding.

Figure 2 :
Figure 2: Diferentially expressed protein spots in 2-DE gels by comparisons of MCF-7 T 120 vs MCF-7 T 0 and MCF10A T 120 vs MCF10A T 0 .(a) Representative 2-DE gel images (in the left panel) of breast cancer MCF-7 cells without zinc (MCF-7 T 0 ) and with exogenous zinc exposure for 120 min (MCF-7 T 120 ).(b) Representative 2-DE gel images (in the right panel) of breast normal MCF10A cells without zinc (MCF10A T 0 ) and with exogenous zinc exposure for 120 min (MCF10A T 120 ).Each protein extract (100 μg) extract was resolved based on isoelectric point (pI) and molecular weight (MW).Te diferentially expressed protein spots are shown with red circles denoting upregulation and green circles denoting downregulation.

Figure 3 :
Figure 3: Diferentially expressed protein spots in 2-DE gels by comparisons of PC3 T 0 vs RWPE-1 T 0 and PC3 T 120 vs RWPE-1 T 120 .(a) Representative 2-DE gel images (in the left panel) of prostate normal RWPE-1 cells without zinc exposure (RWPE-1 T 0 ) and prostate cancer PC3 cells without zinc exposure (PC3 T 0 ).(b) Representative 2-DE gel images (in the right panel) of prostate normal RWPE-1 cells with exogenous zinc exposure for 120 min (RWPE-1 T 120 ) and prostate cancer PC3 cells with exogenous zinc exposure for 120 min (PC3 T 120 ).Each protein extract (100 μg) was resolved based on isoelectric point (pI) and molecular weight (MW).Te diferentially expressed protein spots are shown with red circles denoting upregulation and green circles denoting downregulation.
6/0.01) ↑ V Note.MW stands for molecular weight, kDa for kilo Dalton, pI for isoelectric point, PLGS for ProteinLynx Global Server, T 0 for 0 min or without zinc exposure (control), T 120 for 120 min, ↑ for upregulation, ↓ for downregulation, PC3 for prostate cancer cells, and RWPE-1 for prostate normal epithelial cells.Te PLGS score, protein accession, theoretical MW/pI, matched peptides, and sequence coverage (%) were obtained using ProteinLynx Global Server (PLGS) software (version 3.0, Waters Corporation, USA) and the UniProt (Homo sapiens, human) database.Gene ID was derived from UniProt database.Te observed MW and pI were calculated according to the protein standards.Te fold changes and p values were acquired from the quantitative analysis of the gel images (each group n � 3) by Delta2D software (version 4.0.8,DECODON Gmbh, Germany).PC3 T 0 /RWPE-1 T 0 is the expression fold change of the proteins in PC3 cells compared to RWPE-1 cells without zinc exposure (T 0 ), PC3 T 120 /RWPE-1 T 120 is the expression fold change of the proteins in PC3 cells compared to RWPE-1 cells following the zinc exposure for 120 min (T 120 ), PC3 T 120 /PC3 T 0 is the expression fold change of the proteins in PC3 cells following zinc exposure for T 120 compared to T 0 , and RWPE-1 T 120 /RWPE-1 T 0 is the expression fold change of the proteins in RWPE-1 cells following zinc exposure for T 120 compared to T 0 .Molecular functions: I, apoptosis; II, signalling; III, RNA binding;

Figure 4 :
Figure 4: Diferentially expressed protein spots in 2-DE gels by comparisons of PC3 T 120 vs PC3 T 0 and RWPE-1 T 120 vs RWPE-1 T 0 .(a) Representative 2-DE gel images (in the left panel) of prostate cancer PC3 cells without zinc exposure (PC3 T 0 ) and with exogenous zinc exposure for 120 min (PC3 T 120 ).(b) Representative 2-DE gel images (in the right panel) of prostate normal RWPE-1 cells without zinc exposure (RWPE-1 T 0 ) and with exogenous zinc exposure for 120 min (RWPE-1 T 120 ).Each protein extract (100 μg) was resolved based on isoelectric point (pI) and molecular weight (MW).Te diferentially expressed protein spots are shown with red circles denoting upregulation and green circles denoting downregulation.

Table 2 :
Identifed proteins in prostate cancer cells (PC3) and normal prostate epithelial cells (RWPE-1) with or without exogenous zinc exposure.