The Tumor Suppressor, P53, Decreases the Metal Transporter, ZIP14

Loss of p53’s proper function accounts for over half of identified human cancers. We identified the metal transporter ZIP14 (Zinc-regulated transporter (ZRT) and Iron-regulated transporter (IRT)-like Protein 14) as a p53-regulated protein. ZIP14 protein levels were upregulated by lack of p53 and downregulated by increased p53 expression. This regulation did not fully depend on the changes in ZIP14’s mRNA expression. Co-precipitation studies indicated that p53 interacts with ZIP14 and increases its ubiquitination and degradation. Moreover, knockdown of p53 resulted in higher non-transferrin-bound iron uptake, which was mediated by increased ZIP14 levels. Our study highlights a role for p53 in regulating nutrient metabolism and provides insight into how iron and possibly other metals such as zinc and manganese could be regulated in p53-inactivated tumor cells.


Introduction
Over 50% of identified human cancers are caused by mutations in the gene encoding tumor suppressor p53 [1][2][3][4]. P53 has a well-documented function as a transcription factor [5]. It also possesses transcription-independent cytosolic functions, including inhibition of autophagy [6][7][8][9], regulation of mitochondrial function [10,11] and regulation of apoptosis [12]. Inactivation of p53 contributes not only to cancer initiation and progression, but also to chemotherapy resistance [13,14]. Tumor cells are characterized by uncontrolled proliferation. Rapidly proliferating cells need considerably more nutrients than normal cells. Tumor metabolism is altered to accommodate for the increased nutrient requirement and metabolic demand. Elucidation of p53's function in regulating nutrient transport is important to the understanding of how p53 influences tumor metabolism.
Iron, an essential nutrient, is required by nearly all cells to maintain energy metabolism, DNA synthesis and cell proliferation. Rapidly growing and frequently dividing tumor cells need considerably more iron than quiescent cells. Iron promotes tumor initiation by enhancing free radical formation and accelerates tumor growth by fostering the cell proliferation [14]. Increasing cellular iron content by repressing the iron storage protein ferritin leads to increased tumor cell proliferation [15]. Tumor cells also have elevated iron uptake to meet the increased iron need [14]. Under most physiological conditions, transferrin-bound iron (TBI) is the major iron source and is mediated through the endocytosis of transferrin (Tf)/transferrin receptor 1 (TfR1) complex [16]. Under conditions of increased iron-bound Tf (Tf-saturation) or decreased Tf availability, such as conditions in cancer patients undergoing chemotherapy, levels of plasma non-transferrin-bound iron (NTBI) gel, transferred to nitrocellulose and incubated for 1 h in blocking buffer (5% nonfat dry milk in Tris-buffered saline-Tween 20, TBST). Blots were incubated for 1 h at room temperature in blocking buffer containing mouse anti-FLAG, anti-FLAG-HRP, M2 (1:10,000, Sigma, St. Louis, MO, USA), rabbit anti-DMT1 (Proteintech, Rosemont, IL, USA, 1:5000), or mouse anti-TfR1 (Thermo Scientific, Waltham, MA, USA, 1:5000). After four washes with TBST, blots were incubated with a 1:5000 goat anti-mouse secondary antibody conjugated to horseradish peroxidase (HRP, Millipore, Burlington, MA, USA). To confirm equivalent loading, blots were stripped for 15 min in Restore PLUS Western Blot Stripping Buffer (Thermo Scientific, Waltham, MA, USA), blocked for 1 h in blocking buffer, and reprobed with mouse anti-actin (Millipore, Burlington, MA, USA, 1:10,000) or rabbit anti-tubulin (Rockland, Limerick, PA, USA, 1:5000) followed by HRP-conjugated goat anti-mouse (Millipore, Burlington, MA, USA) or donkey anti-rabbit (GE Healthcare, Little Chalfont, UK) secondary antibody. For loading control of plasma membrane proteins, mouse anti-Na + , K + ATPase antibody (1:2000, Santa Cruz, Dallas, TX, USA) followed by HRP-conjugated secondary antibodies were used. After two washes with TBST and TBS, bands were visualized by using enhanced chemiluminescence (SuperSignal West Pico, Thermo Scientific, Waltham, MA, USA) and X-ray film. For quantification, after primary antibody incubation, blots were probed with infrared fluorescent dye (IRDye 800) conjugated rabbit anti-mouse or Alexa Fluor 680 conjugated goat anti-rabbit secondary antibody (Thermo Scientific, Waltham, MA, USA) and visualized using a Licor Imaging System (LI-COR, Lincoln, NE, USA). HepG2 cell with endogenously FLAG tagged ZIP14 (HepG2-ZIP14-FLAG cells) were used for immunoprecipitation analysis. The post-nuclear supernatant fractions of the cell lysates were incubated with anti-FLAG (M2) agarose beads (Sigma, St. Louis, MO, USA) for 1 h at 4 • C. The beads were washed three times for 10 min in NETT buffer. The protein complex was eluted from the beads with elution buffer (0.5 mg/mL triple FLAG peptide in TBS with protease inhibitor). The elution sample was separated into two halves and analyzed by immunoblotting. One half was probed for FLAG-ZIP14 and Actin. Another half was probed by anti-ubiquitin and anti-p53 antibodies.

Cellular Iron Uptake Assay
The iron uptake analysis was performed as previously described [26]. Briefly, for non-transferrinbound iron uptake, HepG2-ZIP14-FLAG cells grown in six-well plates were washed three times with serum free media (SFM) and incubated for 1 h in SFM. Cells were incubated with 2 µM 55 Fe (ferric-citrate) for 2 h and then washed three times with cell membrane-impermeable iron chelator solution to remove cell surface-bound iron. Cells were solubilized with lysis buffer (0.1% Triton X-100, 0.1% NaOH) and cell-associated radioactivity was determined by a scintillation counter. Iron uptake was calculated as cpm/mg of protein and expressed as percent of control.

Measurement of Iron Levels by Inductively Coupled Plasma Mass Spectrometry (ICP-MS)
The cellular iron level was determined by ICP-MS. Briefly, HepG2 cells were transfected with p53-specific siRNA or negative control siRNA for 48 h in a six-well plate. Cells were washed four times with ice-cold PBS-EDTA (2 mM) and solubilized with 400 µL lysis buffer (0.2 M NaOH, 0.2% SDS). The protein concentration was determined by using the RC DC Protein Assay (Bio-Rad). The cell lysates were digested in nitric acid at a concentration of 12%. Digestion was carried out at 85 • C for 16 h and 95 • C for an additional 2 h. The digested lysates were diluted in Milli-Q H 2 O to a final concentration of 1% nitric acid. The iron concentration was measured by an Agilent 7700 Series ICP-MS instrument. The ICP-MS analyses were performed by the Arizona Laboratory for Emerging Contaminants (ALEC) at the University of Arizona.

Statistical Analysis
Data were analyzed by one-way ANOVA or unpaired t-test with GraphPad Prism software, version 5 (GraphPad Software, La Jolla, CA, USA). Tukey's post hoc comparisons tests were performed with multiple comparisons. p-values < 0.05 were considered to be statistically significant.

Suppression of P53 Increases Endogenous Levels of ZIP14, but Not DMT1 in HepG2 Cells
To examine whether changes in p53 alters ZIP14 or DMT1 levels, we chose to use the HepG2 hepatoma cell line because it expresses both ZIP14 and DMT1 [33]. Importantly, HepG2 cells express WT p53 and represent an ideal cell type to test p53 function [34,35]. A previously generated HepG2 cell line with endogenous FLAG-tagged ZIP14 was used [26,30]. This cell line keeps the endogenous gene-regulatory machinery intact and is a convenient cell model to detect ZIP14 protein by the well-validated anti-FLAG antibody [26,30]. Immunoblot analyses revealed that ZIP14 levels increased significantly after siRNA-mediated p53 knockdown ( Figure 1A-C). To demonstrate the specific effect of p53 on ZIP14 protein, another p53-targeting siRNA (p53 siRNA-2) was used. Similar to p53 siRNA-1, siRNA-2 treatment efficiently reduced the expression of p53 and resulted in elevated ZIP14 levels ( Figure 1D). Quantitative RT-PCR (qRT-PCR) analysis indicated that ZIP14 mRNA levels correlated with the increased ZIP14 protein when p53 expression was suppressed ( Figure 1E). To immunodetect DMT1 protein in HepG2 cells, we first validated the anti-DMT1 antibody in HepG2 cells by using gene specific siRNA. Immunoblot analysis indicated that the DMT1-immunoreactive band decreases with three different siRNAs targeting DMT1, confirming the detected band migrating at 60 kDa is DMT1 ( Figure 1F). In contrast to ZIP14, DMT1 did not change when p53 expression was suppressed ( Figure 1G). These results suggest that the WT p53 plays a role in regulating ZIP14, but not DMT1 in HepG2 cells.

Statistical Analysis
Data were analyzed by one-way ANOVA or unpaired t-test with GraphPad Prism software, version 5 (GraphPad Software, La Jolla, CA, USA). Tukey's post hoc comparisons tests were performed with multiple comparisons. p-values < 0.05 were considered to be statistically significant.

Suppression of P53 Increases Endogenous Levels of ZIP14, but Not DMT1 in HepG2 Cells
To examine whether changes in p53 alters ZIP14 or DMT1 levels, we chose to use the HepG2 hepatoma cell line because it expresses both ZIP14 and DMT1 [33]. Importantly, HepG2 cells express WT p53 and represent an ideal cell type to test p53 function [34,35]. A previously generated HepG2 cell line with endogenous FLAG-tagged ZIP14 was used [26,30]. This cell line keeps the endogenous gene-regulatory machinery intact and is a convenient cell model to detect ZIP14 protein by the well-validated anti-FLAG antibody [26,30]. Immunoblot analyses revealed that ZIP14 levels increased significantly after siRNA-mediated p53 knockdown ( Figure 1A-C). To demonstrate the specific effect of p53 on ZIP14 protein, another p53-targeting siRNA (p53 siRNA-2) was used. Similar to p53 siRNA-1, siRNA-2 treatment efficiently reduced the expression of p53 and resulted in elevated ZIP14 levels ( Figure 1D). Quantitative RT-PCR (qRT-PCR) analysis indicated that ZIP14 mRNA levels correlated with the increased ZIP14 protein when p53 expression was suppressed ( Figure 1E). To immunodetect DMT1 protein in HepG2 cells, we first validated the anti-DMT1 antibody in HepG2 cells by using gene specific siRNA. Immunoblot analysis indicated that the DMT1immunoreactive band decreases with three different siRNAs targeting DMT1, confirming the detected band migrating at 60 kDa is DMT1 ( Figure 1F). In contrast to ZIP14, DMT1 did not change when p53 expression was suppressed ( Figure 1G). These results suggest that the WT p53 plays a role in regulating ZIP14, but not DMT1 in HepG2 cells. for 48 h. Cell lysates were analyzed by immunoblotting for ZIP14. After stripping, blots were reprobed for p53 and β-actin; (E) ZIP14 transcript levels after p53 knockdown were quantified and normalized to those of β-actin. Experiments were repeated at least three times with consistent results; (F) cells were incubated with three different DMT1-targeting siRNA for 48 h. Cell lysates were analyzed by immunoblotting for DMT1. After stripping, blots were re-probed for β-actin; (G) cells were treated with p53-targeting siRNA #1 for 48 h. Cell lysates were analyzed by immunoblotting for DMT1. After stripping, blots were re-probed for p53 and β-actin. NC: universal scrambled negative control siRNA, * p < 0.05, ** p < 0.001, *** p < 0.0001, compared with control.

Knockdown of P53 Does Not Change TfR1 and Ferritin Levels in HepG2 Cells
Previous studies demonstrated that ZIP14 levels are upregulated by iron overload in HepG2 cells [30]. To examine whether the increase in ZIP14 levels after p53 knockdown results from changes in cellular iron content, we first measured protein levels of TfR1 and ferritin. TfR1 is upregulated by iron deficiency and downregulated by iron overload. In contrast, ferritin is increased in iron-loaded cells and decreased by iron depletion. Knockdown of p53 by siRNA did not significantly change the levels of TfR1 (Figure 2A) or ferritin ( Figure 2B), suggesting unchanged cellular iron levels. However, unaltered ferritin and TfR1 levels could be the consequence of regulatory effects mediated by p53. For example, a study in H1299 lung cells demonstrated that p53 induction resulted in decreased TfR1 expression and increased ferritin expression [27]. To further examine the effect of p53 knockdown on cellular iron in HepG2 cells, we directly measured iron levels by ICP-MS analysis. We found that p53 knockdown resulted in a ~33% increase in cellular iron ( Figure 2C), which could contribute to elevated ZIP14 when p53 expression is suppressed. These results also suggest that p53 knockdown in HepG2 cells results in no apparent changes in the levels of TfR1 or ferritin in spite of increased cellular iron. Cell lysates were analyzed by immunoblotting for ZIP14. After stripping, blots were re-probed for p53 and β-actin; (E) ZIP14 transcript levels after p53 knockdown were quantified and normalized to those of β-actin. Experiments were repeated at least three times with consistent results; (F) cells were incubated with three different DMT1-targeting siRNA for 48 h. Cell lysates were analyzed by immunoblotting for DMT1. After stripping, blots were re-probed for β-actin; (G) cells were treated with p53-targeting siRNA #1 for 48 h. Cell lysates were analyzed by immunoblotting for DMT1. After stripping, blots were re-probed for p53 and β-actin. NC: universal scrambled negative control siRNA, * p < 0.05, ** p < 0.001, *** p < 0.0001, compared with control.

Knockdown of P53 Does Not Change TfR1 and Ferritin Levels in HepG2 Cells
Previous studies demonstrated that ZIP14 levels are upregulated by iron overload in HepG2 cells [30]. To examine whether the increase in ZIP14 levels after p53 knockdown results from changes in cellular iron content, we first measured protein levels of TfR1 and ferritin. TfR1 is upregulated by iron deficiency and downregulated by iron overload. In contrast, ferritin is increased in iron-loaded cells and decreased by iron depletion. Knockdown of p53 by siRNA did not significantly change the levels of TfR1 ( Figure 2A) or ferritin ( Figure 2B), suggesting unchanged cellular iron levels. However, unaltered ferritin and TfR1 levels could be the consequence of regulatory effects mediated by p53. For example, a study in H1299 lung cells demonstrated that p53 induction resulted in decreased TfR1 expression and increased ferritin expression [27]. To further examine the effect of p53 knockdown on cellular iron in HepG2 cells, we directly measured iron levels by ICP-MS analysis. We found that p53 knockdown resulted in a~33% increase in cellular iron ( Figure 2C), which could contribute to elevated ZIP14 when p53 expression is suppressed. These results also suggest that p53 knockdown in HepG2 cells results in no apparent changes in the levels of TfR1 or ferritin in spite of increased cellular iron. for 48 h. Cell lysates were analyzed by immunoblotting for ZIP14. After stripping, blots were reprobed for p53 and β-actin; (E) ZIP14 transcript levels after p53 knockdown were quantified and normalized to those of β-actin. Experiments were repeated at least three times with consistent results; (F) cells were incubated with three different DMT1-targeting siRNA for 48 h. Cell lysates were analyzed by immunoblotting for DMT1. After stripping, blots were re-probed for β-actin; (G) cells were treated with p53-targeting siRNA #1 for 48 h. Cell lysates were analyzed by immunoblotting for DMT1. After stripping, blots were re-probed for p53 and β-actin. NC: universal scrambled negative control siRNA, * p < 0.05, ** p < 0.001, *** p < 0.0001, compared with control.

Knockdown of P53 Does Not Change TfR1 and Ferritin Levels in HepG2 Cells
Previous studies demonstrated that ZIP14 levels are upregulated by iron overload in HepG2 cells [30]. To examine whether the increase in ZIP14 levels after p53 knockdown results from changes in cellular iron content, we first measured protein levels of TfR1 and ferritin. TfR1 is upregulated by iron deficiency and downregulated by iron overload. In contrast, ferritin is increased in iron-loaded cells and decreased by iron depletion. Knockdown of p53 by siRNA did not significantly change the levels of TfR1 ( Figure 2A) or ferritin ( Figure 2B), suggesting unchanged cellular iron levels. However, unaltered ferritin and TfR1 levels could be the consequence of regulatory effects mediated by p53. For example, a study in H1299 lung cells demonstrated that p53 induction resulted in decreased TfR1 expression and increased ferritin expression [27]. To further examine the effect of p53 knockdown on cellular iron in HepG2 cells, we directly measured iron levels by ICP-MS analysis. We found that p53 knockdown resulted in a ~33% increase in cellular iron ( Figure 2C), which could contribute to elevated ZIP14 when p53 expression is suppressed. These results also suggest that p53 knockdown in HepG2 cells results in no apparent changes in the levels of TfR1 or ferritin in spite of increased cellular iron.

Increased Expression of P53 Decreases ZIP14 in HepG2 Cells
We next tested the effect of p53 overexpression on ZIP14 in HepG2 cells by transfecting 1 or 2 µg of p53 encoding vector (pCMV-p53) into HepG2 cells for 72 h. As a control, cells were transfected with 2 µg of an empty vector (pCMV). Immunoblot analysis demonstrated that p53 protein levels increased in a dose-dependent manner with the transfected p53 plasmid ( Figure 3A,B). Results showed a trend of decrease in ZIP14 levels when cells were transfected with 1 µg of p53-expressing vector and a significant decrease in ZIP14 levels (about 36%) in cells transfected with 2 µg of p53 vector ( Figure 3C). Consistently, qRT-PCR analysis demonstrated that the ZIP14 mRNA levels decreased by about 26% when cells were transfected with 2 µg of p53 vector ( Figure 3D). No changes in TfR1 levels were observed in cells transfected with p53 ( Figure 3E), suggesting that the decrease in ZIP14 after p53 overexpression only contributed to a minor amount of total iron uptake in these cells.

Increased Expression of P53 Decreases ZIP14 in HepG2 Cells
We next tested the effect of p53 overexpression on ZIP14 in HepG2 cells by transfecting 1 or 2 µg of p53 encoding vector (pCMV-p53) into HepG2 cells for 72 h. As a control, cells were transfected with 2 µg of an empty vector (pCMV). Immunoblot analysis demonstrated that p53 protein levels increased in a dose-dependent manner with the transfected p53 plasmid ( Figure 3A,B). Results showed a trend of decrease in ZIP14 levels when cells were transfected with 1 µg of p53-expressing vector and a significant decrease in ZIP14 levels (about 36%) in cells transfected with 2 µg of p53 vector ( Figure 3C). Consistently, qRT-PCR analysis demonstrated that the ZIP14 mRNA levels decreased by about 26% when cells were transfected with 2 µg of p53 vector ( Figure 3D). No changes in TfR1 levels were observed in cells transfected with p53 ( Figure 3E), suggesting that the decrease in ZIP14 after p53 overexpression only contributed to a minor amount of total iron uptake in these cells.

Increased Expression of P53 Decreases ZIP14 in HepG2 Cells
We next tested the effect of p53 overexpression on ZIP14 in HepG2 cells by transfecting 1 or 2 µg of p53 encoding vector (pCMV-p53) into HepG2 cells for 72 h. As a control, cells were transfected with 2 µg of an empty vector (pCMV). Immunoblot analysis demonstrated that p53 protein levels increased in a dose-dependent manner with the transfected p53 plasmid ( Figure 3A,B). Results showed a trend of decrease in ZIP14 levels when cells were transfected with 1 µg of p53-expressing vector and a significant decrease in ZIP14 levels (about 36%) in cells transfected with 2 µg of p53 vector ( Figure 3C). Consistently, qRT-PCR analysis demonstrated that the ZIP14 mRNA levels decreased by about 26% when cells were transfected with 2 µg of p53 vector ( Figure 3D). No changes in TfR1 levels were observed in cells transfected with p53 ( Figure 3E), suggesting that the decrease in ZIP14 after p53 overexpression only contributed to a minor amount of total iron uptake in these cells.

Knockdown of Endogenously Expressed P53 Increases Endogenous ZIP14 in HEK293 ZIP14-FLAG Cells
To validate the effect of p53 on ZIP14 in another cell type, we used the HEK293 human embryonic kidney cell line. Similar to HepG2 cells, HEK293 cells express wild-type p53 [36,37]. In order to detect endogenous ZIP14 protein in HEK293 cells, the CRISPR-Cas9 mediated genome editing approach was employed to add a 3× FLAG epitope near the C-terminus just before the stop codon of ZIP14 to generate endogenous HEK 293 ZIP14-FLAG cells ( Figure 4A). Cell clones containing 3× FLAG insertion were identified by PCR-based genotyping ( Figure 4B). Successful inframe insertion of the 3× FLAG epitope at the C-terminus of ZIP14 was confirmed by sequencing. Transfection of cells with ZIP14 siRNA decreased the band detected by anti-FLAG antibody ( Figure 4C), indicating that we could use anti-FLAG antibody to detect endogenous ZIP14 protein in HEK293 cells. To demonstrate that adding a 3× FLAG epitope does not alter ZIP14's regulation, we incubated cells with the iron chelator DFO to induce iron deficiency. Iron deficiency resulted in significantly lowered ZIP14 levels ( Figure 4D), which is consistent with our previous results for endogenous ZIP14 in HepG2 cells and stably transfected ZIP14 in HEK293 cells [30]. These results indicate that the regulation of ZIP14 is preserved in HEK293-ZIP14-FLAG cells. We then investigated the effect of p53 suppression or overexpression on ZIP14 in these cells. Knockdown of endogenous p53 increased ZIP14 protein ( Figure 4E) and mRNA levels ( Figure 4F), and p53 overexpression resulted in decreased ZIP14 at both protein and mRNA levels ( Figure 4G,H). These results confirm the negative regulation of ZIP14 by p53 and demonstrate that this regulation is not limited to one specific cell type.

Knockdown of Endogenously Expressed P53 Increases Endogenous ZIP14 in HEK293 ZIP14-FLAG Cells
To validate the effect of p53 on ZIP14 in another cell type, we used the HEK293 human embryonic kidney cell line. Similar to HepG2 cells, HEK293 cells express wild-type p53 [36,37]. In order to detect endogenous ZIP14 protein in HEK293 cells, the CRISPR-Cas9 mediated genome editing approach was employed to add a 3× FLAG epitope near the C-terminus just before the stop codon of ZIP14 to generate endogenous HEK 293 ZIP14-FLAG cells ( Figure 4A). Cell clones containing 3× FLAG insertion were identified by PCR-based genotyping ( Figure 4B). Successful in-frame insertion of the 3× FLAG epitope at the C-terminus of ZIP14 was confirmed by sequencing. Transfection of cells with ZIP14 siRNA decreased the band detected by anti-FLAG antibody ( Figure 4C), indicating that we could use anti-FLAG antibody to detect endogenous ZIP14 protein in HEK293 cells. To demonstrate that adding a 3× FLAG epitope does not alter ZIP14's regulation, we incubated cells with the iron chelator DFO to induce iron deficiency. Iron deficiency resulted in significantly lowered ZIP14 levels ( Figure 4D), which is consistent with our previous results for endogenous ZIP14 in HepG2 cells and stably transfected ZIP14 in HEK293 cells [30]. These results indicate that the regulation of ZIP14 is preserved in HEK293-ZIP14-FLAG cells. We then investigated the effect of p53 suppression or overexpression on ZIP14 in these cells. Knockdown of endogenous p53 increased ZIP14 protein ( Figure 4E) and mRNA levels ( Figure 4F), and p53 overexpression resulted in decreased ZIP14 at both protein and mRNA levels ( Figure 4G,H). These results confirm the negative regulation of ZIP14 by p53 and demonstrate that this regulation is not limited to one specific cell type.

Knockdown of Endogenously Expressed P53 Increases Endogenous ZIP14 in HEK293 ZIP14-FLAG Cells
To validate the effect of p53 on ZIP14 in another cell type, we used the HEK293 human embryonic kidney cell line. Similar to HepG2 cells, HEK293 cells express wild-type p53 [36,37]. In order to detect endogenous ZIP14 protein in HEK293 cells, the CRISPR-Cas9 mediated genome editing approach was employed to add a 3× FLAG epitope near the C-terminus just before the stop codon of ZIP14 to generate endogenous HEK 293 ZIP14-FLAG cells ( Figure 4A). Cell clones containing 3× FLAG insertion were identified by PCR-based genotyping ( Figure 4B). Successful inframe insertion of the 3× FLAG epitope at the C-terminus of ZIP14 was confirmed by sequencing. Transfection of cells with ZIP14 siRNA decreased the band detected by anti-FLAG antibody ( Figure 4C), indicating that we could use anti-FLAG antibody to detect endogenous ZIP14 protein in HEK293 cells. To demonstrate that adding a 3× FLAG epitope does not alter ZIP14's regulation, we incubated cells with the iron chelator DFO to induce iron deficiency. Iron deficiency resulted in significantly lowered ZIP14 levels ( Figure 4D), which is consistent with our previous results for endogenous ZIP14 in HepG2 cells and stably transfected ZIP14 in HEK293 cells [30]. These results indicate that the regulation of ZIP14 is preserved in HEK293-ZIP14-FLAG cells. We then investigated the effect of p53 suppression or overexpression on ZIP14 in these cells. Knockdown of endogenous p53 increased ZIP14 protein ( Figure 4E) and mRNA levels ( Figure 4F), and p53 overexpression resulted in decreased ZIP14 at both protein and mRNA levels ( Figure 4G,H). These results confirm the negative regulation of ZIP14 by p53 and demonstrate that this regulation is not limited to one specific cell type.  PCR results demonstrate that both sgRNAs successfully lead to the knock-in of 3× FLAG sequence in HEK293 cells. We then identified single cell clones to establish the cell line as HEK293-ZIP14-FLAG cells and these cells were used from C to H; (C) cells were treated with ZIP14targeting siRNAs for 48 h. Cells lysates were analyzed by immunoblotting for ZIP14. After stripping, blots were probed for tubulin as a loading control; (D) to confirm that endogenously tagged 3× FLAG ZIP14 in HEK293 cells retains its biological function and regulation, cells were treated with iron chelator desferrioxamine (DFO) (100 µM) for 24 h to induce iron deficiency. Cell lysates were analyzed by immunoblotting for ZIP14. After stripping, blots were probed for TfR1 as an iron treatment control and tubulin as a loading control; (E) cells were transfected with p53targeting siRNA for 48 h. Cell lysates were analyzed by immunoblotting for ZIP14. After stripping, blots were probed for p53 and tubulin; (F) transcript levels of ZIP14 after p53 siRNA treatment were measured and normalized to those of β-actin. Data represents three independent experiments with consistent results. NC: negative control; (G) cells were transfected with an empty pCMV (CON) or pCMV-p53 vector for 48 h. Cell lysates were analyzed by immunoblotting; (H) transcript levels of ZIP14 after p53 overexpression were measured and normalized to those of β-actin. Data represents Figure 4. Knockdown of p53 increases endogenous ZIP14 protein level in human embryonic kidney HEK293 cells. We generated a derivative HEK293 cell line in which the endogenous ZIP14 locus was modified so as to incorporate a 3× FLAG tag at the ZIP14 C-terminus by using the Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)/CRISPR associated protein 9 (CRISPR/Cas9)-mediated genome editing approach (HEK293-ZIP14-FLAG cells). We design the single-stranded oligodeoxynucleotides (ssODN) to include about 50 bp of left and right homologous regions (HR) at each side of the ZIP14 stop codon, and the nucleotides encoding 3× FLAG sequence. (A) Schema for genotyping by PCR analysis: primer pair was designed to amplify a region of about 200 bp in length to cover the upstream and downstream area of ZIP14 s stop codon. After ssODN mediated homologous recombination, the knock-in allele will have codons for 3× FLAG sequence, resulting in a PCR fragment of about 260 bp; (B) two single-guide RNA (sgRNA) were cloned into the px330 vector. PCR results demonstrate that both sgRNAs successfully lead to the knock-in of 3× FLAG sequence in HEK293 cells. We then identified single cell clones to establish the cell line as HEK293-ZIP14-FLAG cells and these cells were used from C to H; (C) cells were treated with ZIP14-targeting siRNAs for 48 h. Cells lysates were analyzed by immunoblotting for ZIP14. After stripping, blots were probed for tubulin as a loading control; (D) to confirm that endogenously tagged 3× FLAG ZIP14 in HEK293 cells retains its biological function and regulation, cells were treated with iron chelator desferrioxamine (DFO) (100 µM) for 24 h to induce iron deficiency. Cell lysates were analyzed by immunoblotting for ZIP14. After stripping, blots were probed for TfR1 as an iron treatment control and tubulin as a loading control; (E) cells were transfected with p53-targeting siRNA for 48 h. Cell lysates were analyzed by immunoblotting for ZIP14. After stripping, blots were probed for p53 and tubulin; (F) transcript levels of ZIP14 after p53 siRNA treatment were measured and normalized to those of β-actin. Data represents three independent experiments with consistent results. NC: negative control; (G) cells were transfected with an empty pCMV (CON) or pCMV-p53 vector for 48 h. Cell lysates were analyzed by immunoblotting; (H) transcript levels of ZIP14 after p53 overexpression were measured and normalized to those of β-actin. Data represents two independent experiments with consistent results. NC: universal scrambled negative control siRNA. * p < 0.05, ** p < 0.001, compared with control.

P53 Suppression Increases ZIP14, Whereas P53 Overexpression Decreases ZIP14 Levels in HEK293 Cells Stably Transfected with ZIP14
To further examine p53's role in regulating ZIP14 protein, we used stably transfected HEK293 cells, which express FLAG-tagged ZIP14 without its 5 and 3 untranslated mRNA regions (HEK293-ZIP14-Stable cells) [30]. Data obtained from two individual stable cell clones indicated that knockdown of p53 increased ( Figure 5A), whereas overexpression of p53 decreased ZIP14 protein levels ( Figure 5B). Interestingly, qPCR analysis by specific primers amplifying the FLAG sequence demonstrated that ZIP14 mRNA levels increased under both p53 suppression ( Figure 5C) and p53 overexpression conditions ( Figure 5D). Different from the endogenous ZIP14 mRNA levels in HEK293 cells ( Figure 4H), the increased mRNA level ( Figure 5D) observed in HEK293-ZIP14-Stable cells overexpressing p53 does not correlate with the decreased protein level ( Figure 5B). This lack of correlation has been observed in our previous study when the same two cell lines were treated with iron chelator to induce iron deficiency [30]. Here, in HEK293-ZIP14-Stable cells, the decreased ZIP14 protein without a decrease in ZIP14 mRNA suggests a post-transcriptional effect of p53 on ZIP14 protein levels. These results lead us to test how p53 altered ZIP14 independently of its well-known transcriptional regulation. two independent experiments with consistent results. NC: universal scrambled negative control siRNA. * p < 0.05, ** p < 0.001, compared with control.

P53 Suppression Increases ZIP14, Whereas P53 Overexpression Decreases ZIP14 Levels in HEK293 Cells Stably Transfected with ZIP14
To further examine p53's role in regulating ZIP14 protein, we used stably transfected HEK293 cells, which express FLAG-tagged ZIP14 without its 5′ and 3′ untranslated mRNA regions (HEK293-ZIP14-Stable cells) [30]. Data obtained from two individual stable cell clones indicated that knockdown of p53 increased ( Figure 5A), whereas overexpression of p53 decreased ZIP14 protein levels ( Figure 5B). Interestingly, qPCR analysis by specific primers amplifying the FLAG sequence demonstrated that ZIP14 mRNA levels increased under both p53 suppression ( Figure 5C) and p53 overexpression conditions ( Figure 5D). Different from the endogenous ZIP14 mRNA levels in HEK293 cells (Figure 4H), the increased mRNA level ( Figure 5D) observed in HEK293-ZIP14-Stable cells overexpressing p53 does not correlate with the decreased protein level ( Figure 5B). This lack of correlation has been observed in our previous study when the same two cell lines were treated with iron chelator to induce iron deficiency [30]. Here, in HEK293-ZIP14-Stable cells, the decreased ZIP14 protein without a decrease in ZIP14 mRNA suggests a post-transcriptional effect of p53 on ZIP14 protein levels. These results lead us to test how p53 altered ZIP14 independently of its well-known transcriptional regulation.  A and B). Transcript levels of ZIP14 after p53 siRNA transfection were measured and normalized to those βactin (C and D). Data represents three independent experiments. * p < 0.05, compared with control.  ZIP14, p53 and β-actin (A,B). Transcript levels of ZIP14 after p53 siRNA transfection were measured and normalized to those β-actin (C,D). Data represents three independent experiments. * p < 0.05, compared with control.

Evidence That P53 Interacts with ZIP14 to Alter the Stability of ZIP14
Previous studies demonstrated that internalized plasma membrane ZIP14 is degraded through the proteasome-mediated pathway that involves the ubiquitination of ZIP14 [30]. To further explore the mechanism underlying the regulation of ZIP14 by p53, we tested whether p53 suppression affected the stability of ZIP14. HepG2-ZIP14-FLAG cells were treated with p53-targeting or negative control siRNA for 48 h. The amount of ZIP14 on the cell surface was measured by biotinylation with an impermeable biotinylation reagent, NHS-SS-biotin. Similar to the changes in ZIP14 in the total cell lysate, cell-surface ZIP14 levels increased with p53 knockdown ( Figure 6A). To test whether the increased cell-surface ZIP14 is due to decreased protein degradation, we analyzed the fate of the cell-surface ZIP14 by the biotin pulse-chase experiment. Cells were incubated with biotinylation reagent at 4 • C and then chased at 37 • C for 2 and 4 h to allow internalization. We found that knockdown of p53 inhibited degradation of cell surface-derived ZIP14 ( Figure 6B). Interestingly, p53 co-immunoprecipitated with ZIP14 suggesting that they form a complex ( Figure 6C  Previous studies demonstrated that internalized plasma membrane ZIP14 is degraded through the proteasome-mediated pathway that involves the ubiquitination of ZIP14 [30]. To further explore the mechanism underlying the regulation of ZIP14 by p53, we tested whether p53 suppression affected the stability of ZIP14. HepG2-ZIP14-FLAG cells were treated with p53-targeting or negative control siRNA for 48 h. The amount of ZIP14 on the cell surface was measured by biotinylation with an impermeable biotinylation reagent, NHS-SS-biotin. Similar to the changes in ZIP14 in the total cell lysate, cell-surface ZIP14 levels increased with p53 knockdown ( Figure 6A). To test whether the increased cell-surface ZIP14 is due to decreased protein degradation, we analyzed the fate of the cell-surface ZIP14 by the biotin pulse-chase experiment. Cells were incubated with biotinylation reagent at 4 °C and then chased at 37 °C for 2 and 4 h to allow internalization. We found that knockdown of p53 inhibited degradation of cell surface-derived ZIP14 ( Figure 6B). Interestingly, p53 co-immunoprecipitated with ZIP14 suggesting that they form a complex ( Figure 6C  Total cell lysates were collected and cell-surface proteins were isolated by streptavidin gel. ZIP14 levels in both the total cell lysate and cell-surface fractions were analyzed by immunoblotting; (B) cell-surface proteins were labeled with biotin at 4 °C and then chased for 2 and 4 h at 37 °C to allow for endocytosis. Cells were lysed and biotin-labeled cell-surface proteins were isolated by using streptavidin gel and were eluted with 50 mM DTT. Samples were analyzed by immunoblotting. Plasma membrane type Na + , K + ATPase Figure 6. Knockdown of p53 decreases the degradation of cell surface ZIP14. (A) HepG2-ZIP14-FLAG cells were incubated with p53-targeting siRNA for 48 h and then cell-surface proteins were labelled with cell membrane-impermeable biotin at 4 • C for 30 min. Total cell lysates were collected and cell-surface proteins were isolated by streptavidin gel. ZIP14 levels in both the total cell lysate and cell-surface fractions were analyzed by immunoblotting; (B) cell-surface proteins were labeled with biotin at 4 • C and then chased for 2 and 4 h at 37 • C to allow for endocytosis. Cells were lysed and biotin-labeled cell-surface proteins were isolated by using streptavidin gel and were eluted with 50 mM DTT. Samples were analyzed by immunoblotting. Plasma membrane type Na + , K + ATPase was used as a control for cell-surface protein; (C) cells were treated with p53-targeting siRNA for 48 h before the immunoprecipitation procedure. Half of the eluted fraction was probed for ZIP14 and the other half was probed with anti-ubiquitin antibody.

Knockdown of P53 Increases Non-Transferrin-Bound Iron Uptake in HepG2 Cells through Elevated ZIP14
To test the functional consequences of p53 knockdown in HepG2 cells, we evaluated the cellular iron transport by measuring NTBI uptake. We found that 55 Fe-citrate (NTBI) uptake increased significantly after cells were incubated with p53-targeting siRNA ( Figure 7A). To test whether the increased NTBI uptake is due to elevated ZIP14 levels, we performed double siRNA knockdown. We found that knockdown of ZIP14 abolished the effect of p53 suppression on cellular iron uptake ( Figure 7B). Western blotting analysis validated the efficient knockdown of both ZIP14 and p53 ( Figure 7C), and confirmed that ZIP14 knockdown does not alter the levels of DMT1, which is another iron import protein at the cell surface ( Figure 7D). Taken together, our results in HepG2 and HEK293 cells indicate that ZIP14 is negatively regulated by p53 and suggest that ZIP14 may contribute to increased cellular NTBI uptake in p53-inactivated tumor cells.
was used as a control for cell-surface protein; (C) cells were treated with p53-targeting siRNA for 48 h before the immunoprecipitation procedure. Half of the eluted fraction was probed for ZIP14 and the other half was probed with anti-ubiquitin antibody.

Knockdown of P53 Increases Non-Transferrin-Bound Iron Uptake in HepG2 Cells through Elevated ZIP14
To test the functional consequences of p53 knockdown in HepG2 cells, we evaluated the cellular iron transport by measuring NTBI uptake. We found that 55 Fe-citrate (NTBI) uptake increased significantly after cells were incubated with p53-targeting siRNA ( Figure 7A). To test whether the increased NTBI uptake is due to elevated ZIP14 levels, we performed double siRNA knockdown. We found that knockdown of ZIP14 abolished the effect of p53 suppression on cellular iron uptake ( Figure 7B). Western blotting analysis validated the efficient knockdown of both ZIP14 and p53 ( Figure 7C), and confirmed that ZIP14 knockdown does not alter the levels of DMT1, which is another iron import protein at the cell surface ( Figure 7D). Taken together, our results in HepG2 and HEK293 cells indicate that ZIP14 is negatively regulated by p53 and suggest that ZIP14 may contribute to increased cellular NTBI uptake in p53-inactivated tumor cells.

Discussion
Nutrient uptake and metabolism in tumor cells are controlled by genetic mutations and cellular responses to the tumor microenvironment. A common metabolic phenotype of diverse tumors is enhanced nutrient uptake [38]. If the nutrient transporters, that are specifically induced in tumor cells are identified, molecules that can inhibit the pathways of their induction or block the function of the induced transporters will have significant potential as chemotherapeutic compounds. In the identified human cancers, the tumor suppressor p53 is the most frequently mutated gene. Oncogenic pathways controlling cell growth and survival are often activated by loss of p53's proper function. Alterations in nutrient uptake and metabolic processes are also consequences of p53 mutations [39][40][41]. Elucidation of the underlying mechanisms that contribute to these alterations is critical for the understanding of tumor metabolism and for the development of cancer treatment.
As a well characterized transcription factor, p53's function depends largely on its localization in the nucleus and its ability to trans-activate other genes causing cell cycle arrest or cell apoptosis [42]. In the present study, by using two wild-type p53 cell lines, we identified ZIP14 as a new p53-regulated protein. We found that ZIP14 is downregulated by p53 overexpression and upregulated by p53 suppression. Interestingly, the results indicate that the effect of p53 on ZIP14 protein is, at least partially independent of changes in ZIP14's mRNA levels. Moreover, we demonstrated that p53 co-precipitates with ZIP14 and affects ZIP14's ubiquitination and degradation. We have disclosed a new mechanism by which plasma membrane ZIP14 is regulated and highlighted the function of p53 in the regulation of cellular iron metabolism.
Our results suggest that loss of p53 may accelerate NTBI uptake through ZIP14, providing insight into the mechanisms of altered nutrient metabolism in p53-related cancers and implying potential clinical significance in patients undergoing chemotherapy. Chemotherapy is a widely used treatment for cancer patients and patients with other disorders, such as blood and autoimmune diseases [43,44]. By eliminating rapidly dividing tumor cells, chemotherapeutics also result in significantly elevated level of NTBI in the plasma [18,45,46]. Plasma NTBI is rapidly taken up by the liver, mainly through ZIP14 [47]. As a result, the extracellular NTBI will increase in the tumor microenvironment and provides a significant iron source for tumor growth, promoting therapy resistance. By using human liver hepatoma HepG2 cells, our study demonstrated that loss of p53 resulted in elevated ZIP14 levels, which increased NTBI uptake into cells, providing insight into how cells lacking p53 may acquire more iron to satiate increased growth demand when NTBI occurs within the body. Since ZIP14 can mediate the transport of other essential nutrients, including zinc and manganese [22,48], our results also suggest a potential role of ZIP14 in regulating the metabolism of these nutrients in p53-inactivated tumor cells.
Growing interest in cancer and ZIP14 has led researchers to investigate the expression level of ZIP14 in patient samples. For example, by exon array analysis, alternative splicing between exon 4A and exon 4B of ZIP14 gene was identified in colorectal tumors, and found to be regulated by the Wnt signaling pathway [49]. In this study, the authors found that exon 4A of ZIP14 was expressed about 50% lower in colorectal tumor mucosa samples than in normal samples, whereas exon 4B levels were mildly elevated in tumor samples compared to normal mucosa samples. Therefore, the authors suggested that alternatively spliced ZIP14 isoforms could be used as a cancer biomarker. A recent study, by using immunohistochemistry analysis, revealed a decreased ZIP14 expression in human prostate cancer tissues compared to that of normal prostate tissues [50]. However, the status of p53 in these patient tissue samples was not examined. Our present study provides evidence that ZIP14 is downregulated by p53. Further studies in p53-null animals and in p53-inactivated patient samples will further elucidate ZIP14's regulation by p53.