Zinc is an essential micronutrient and cytoprotectant involved in the host response to inflammatory stress. We tested whether zinc transporters, the critical regulators that maintain intracellular zinc concentrations, play a role in cell survival, particularly in lung epithelia, during inflammation. Initially, mRNA transcripts were quantitatively measured by RT-PCR for all known human zinc transporters, including 14 importers (SLC39A1–14) and 10 exporters (SLC30A1–10), in primary human lung epithelia obtained from multiple human donors and BEAS-2B cell cultures under baseline and TNF-α-stimulated conditions. While many zinc transporters were constitutively expressed, only SLC39A8 (Zip8) mRNA was strongly induced by TNF-α. Endogenous Zip8 protein was not routinely detected under baseline conditions. In sharp contrast, TNF-α induced the expression of a glycosylated protein that translocated to the plasma membrane and mitochondria. Increased Zip8 expression resulted in an increase in intracellular zinc content and coincided with cell survival in the presence of TNF-α. Inhibition of Zip8 expression using a short interfering RNA probe reduced cellular zinc content and impaired mitochondrial function in response to TNF-α, resulting in loss of cell viability. These data are the first to characterize human Zip8 and remarkably demonstrate that upregulation of Zip8 is sufficient to protect lung epithelia against TNF-α-induced cytotoxicity. We conclude that Zip8 is unique, relative to other Zip proteins, by functioning as an essential zinc importer at the onset of inflammation, thereby facilitating cytoprotection within the lung.
- tissue injury
- cell death
zinc is an essential dietary element and known cytoprotectant (32, 34). We recently reported that zinc protects the lung epithelium following activation of the extrinsic apoptosis pathway, a contributing factor in acute lung injury (1, 2). To identify the mechanism(s) responsible for zinc-mediated cytoprotection, we investigated zinc transporters, the primary regulators of zinc homeostasis in mammals. Zinc transporters comprise a family of multiple transmembrane-spanning domain proteins that are encoded by two solute-linked carrier (SLC) gene families: SLC30 (also known as ZnT) and SLC39 (also known as Zip), from here on in referred to as ZnTs and Zips. Our own evaluation of the human genome (http://genome.ucsc.edu) identified 10 ZnT and 14 Zip family members, corroborating two recent reports (7, 22). In general, ZnT and Zip family members have opposite roles in regulating cellular zinc homeostasis. ZnT transporters reduce cytosolic zinc bioavailability by promoting zinc efflux in conditions of excess, whereas Zip transporters function by increasing cytosolic zinc during deficient states. Members of both families exhibit tissue-specific expression and possess differential responsiveness to dietary zinc as well as to physiological stimuli, including cytokines and hormones [for reviews, see Kambe and Liuzzi (16) and Liuzzi and Cousins (22)]. The expression of certain zinc transporters is controlled by cytosolic zinc concentrations through interactions with metal-responsive transcription factor 1 (MTF1) or by influencing mRNA stability, whereas much less is known regarding the molecular events that govern their response to cytokines. Recently, murine ZIP14 expression was reported to be induced through an IL-6-dependant pathway at the onset of the systemic acute phase response. Induction of the expression resulted in the mobilization of zinc from the plasma into hepatocytes, thereby suggesting a role for ZIP14 in zinc-mediated cytoprotection (23). In addition, Lang and colleagues (21) reported that zinc transporter gene expression is attenuated in favor of cellular zinc uptake in the lung following allergic-mediated airway inflammation in a mouse model of asthma. Since the molecular pathogenic events that define acute lung injury remain uncertain and through the observation that zinc is a cytoprotecant in lung epithelia (3, 6, 25), we hypothesized that zinc transporters have a vital role in mediating cytoprotection in the lung at the onset of acute systemic inflammation.
Sepsis or the exogenous administration of LPS in humans results in a transient decrease in plasma zinc concentration that occurs as a result of bioredistribution without loss of whole body zinc content (8, 26). It has been proposed that this innate host response increases intracellular zinc availability for protein synthesis, neutralization of reactive nitrogen and oxygen species, and prevention of microbial invasion (19). If correct, then deficits in zinc, as a consequence of dietary insufficiency or transporter dysfunction, may increase host susceptibility to injury.
The present study addressed whether and to what extent zinc transporters contribute to a cytokine stress response in human lung epithelia. By measuring the expression of all 24 zinc transporters, we observed that expression of only SLC39A8/Zip8 is substantially upregulated by TNF-α, a prototypical inflammatory cytokine expressed at the onset of the acute phase response. Previous work by Begum and colleagues (4) demonstrated that Zip8 expression is upregulated in human monocytes in response to LPS, TNF-α, and live bacteria (4). However, the question whether this is a unique and biologically relevant process was not determined. Our investigation extends upon these initial observations by demonstrating that the induction of Zip8 expression in human lung epithelia is indispensable and essential for the mobilization of zinc into the cell at the onset of inflammation, thereby protecting the cell from injury and death.
Primary, differentiated, polarized, human lung epithelial cells (hLECs) were isolated and cultured as previously reported (3, 17). hLECs were maintained in 1:1 DMEM and Ham's F-12 media (DMEM-F-12) supplemented with 2% Ultroser G (BioSepra, Villeneuve, France) and antibiotics unless otherwise stated. Zinc was not deliberately added to the culture medium. Human lungs were collected with approval from The Ohio State University Institutional Review Board. The human BEAS-2B upper airway epithelial cell line (catalog no. CRL-9609, American Type Culture Collection) was cultured under DMEM with 10% FBS and antibiotics. To further minimize zinc contamination, cells were maintained in serum-free medium for 24 h before and during the course of study.
The conventional standard used to identify human zinc transporter genes is SLC30A or SLC39A, whereas reference to the corresponding proteins is identified as either ZnT or Zip family members. To simplify the terminology in this investigation, when referring to the gene and protein, we identify both SLC39A8 and Zip8 exclusively as Zip8.
Evaluation of zinc transporters by quantitative RT-PCR.
Total RNA was isolated from primary hLECs or BEAS-2B cells using TRIzol reagent (Invitrogen). One microgram of total RNA was converted to cDNA using reverse transcriptase. Twenty-four human zinc transporter family members were identified using the University of California-Santa Cruz genome database (14 importers Zip1–14 and 10 exporters ZnT1–10) (http://genome.ucsc.edu). Primer pairs were designed for all 24 human zinc transporters using Primer Express software version 2.0 (Applied Biosystems) with similar melting point temperatures for concomitant use on a 96-well plate. Zinc transporter gene expression was assessed by quantitative PCR (qPCR) using Power SYBR Green Master Mix (2×, Applied Biosystems, Foster City, CA) and an ABI 7000 automated thermocycler.
Zinc transporter mRNA expression for 12 human donor cultures was first determined for all 24 transporters under baseline culture conditions. Transporter gene expression was then analyzed following exposure to TNF-α for 4 and 24 h. TNF-α stimulation was evaluated using seven separate donor cultures and cultures of BEAS-2B cells. All experiments were performed in duplicate, and all cycle threshold (Ct) values were standardized to β-actin for each sample before further calculations. The equation to calculate fold change was as follows: fold change = 2ABS(Ct baseline − Ct stimulated), where ABS is absolute absorbance value. mRNA expression was reported as the average relative copy number (RCN) as follows: 2−ΔCt× 100, where ΔCt is the Ct value standardized to β-actin (9). Dissociation curves were also initially evaluated for each qPCR sample to qualitatively confirm assay integrity.
Zip8 plasmid construction.
The full-length reference sequence cDNA for human Zip8 (GenBank Accession No. NM_022154) was purchased packaged in a mammalian plasmid vector (pCMV6-XL4) (OriGene). The plasmid was propagated per standard protocols and then subcloned into the pDSRed-N1 mammalian expression vector (Clontech) at EcoR1 and BamHI restriction sites to generate a COOH-terminal DsRed-tagged fusion protein. In-frame ligation was verified by sequence analysis prior to transfection. BEAS-2B cells were established under standard zinc-deficient culture conditions and transiently transfected with DsRedZip8 plasmid or control plasmid (DsRed Monomer Vector) using Lipofectamine 2000 Reagent (Invitrogen).
Antibody production and Zip8 protein expression.
A polyclonal rabbit anti-peptide antiserum to Zip8 corresponding to amino acid residues 225–243 was custom developed (Covance) similar to a previous report (4). Primary hLECs or BEAS-2B cells were dissociated with Cellstripper enzyme solution (Mediatech-Cellgro, Herndon, VA), and cell lysates or fractions were isolated. Samples (5 μg protein) were then denatured at 37°C for 30 min. Cytosolic, membrane, and nuclear fractions were analyzed by Western blot analysis with the anti-Zip8 antiserum compared with preimmune antiserum to verify specificity of the antiserum and quantify the protein in cell compartments. Protein expression was evaluated with and without TNF-α stimulation or following transfection with control pCMV vector versus wild-type Zip8.
BEAS-2B cells were immunostained to determine Zip8 protein location using an anti-human Zip8 antiserum. 4′,6-Diamidino-2-phenylindole (DAPI) was used to visualize the nucleus. Immunostaining experiments were performed at baseline and after transfection with DsRed control or DsRedZip8 plasmid. Protein location was evaluated using green fluorescent protein (green), red fluorescent protein (red), or DAPI (blue) filters with a disc scanning confocal microscope (Olympus BX61). Images were initially taken with all three filters engaged and then viewed separately to assess individual fields. The magnification of all images involved use of ×10 (WHN10x) and ×60 (Olympus 60×/1.42 Oil PlaneApon or 60×/0.90N LUMPLANF1) objectives. z-Section images were routinely obtained at room temperature for the analysis of fixed slides and at 37°C for live cell images using a ×60 immersion objective. The imaging medium included the use of Olympus oil for slides and DMEM without phenol red for live culture. The fluorochromes were as follows: DsRed (from BD Biosciences, excitation/emission: 551.5/583 nm), JC-1(from Molecular Probes, excitation/emission: 514/529 and 590 nm), MitoTracker green (from Molecular Probes, excitation/emission: 490/516 nm), FluoZin-3(from Molecular Probes, excitation/emission: 494/516 nm), and DAPI (from Sigma, excitation/emission: 340/488 nm).
Determination of Zip8 glycosylation.
Membrane protein was incubated with 1% Nonidet P-40 and 50 mM sodium phosphate (pH 7.5, 25°C) and PNGase F or endoglycosidase H (EndoH; 750 units, New England Biolabs, Ipswich, MA) at 37°C for 2 h. Samples were then denatured with SDS-PAGE loading buffer (37°C for 30 min) and run on a 7.5% SDS-PAGE gel before being immunoblotted to determine the extent of protein glycosylation. The Quentix Western Blot Signal Enhancer kit (Pierce) was used to increase signal intensity.
Cell surface biotinylation and immunolocalization.
Cell surface biotinylation was conducted to determine the presence of Zip8 expression at the cell membrane. BEAS-2B cells were first transfected with DsRed control or DsRedZip8 plasmid and cultured as described above. Cells were then rinsed with PBS-containing 0.1 mM CaCl2 and 1 mM MgCl2 followed by cold PBS (pH 8.0). Cells were exposed to sulfo-NHS-LC biotin and then washed with PBS containing 1% BSA. Cells were disadhered (Cellstripper enzyme, Mediatech-Cellgro, Herndon, VA), collected, and washed with PBS (pH 7.4). Biotinylated protein was immunoprecipitated with streptavidin beads and purified prior to being resolved on a 7.5% Tris·HCl gel (Bio-Rad, Hercules, CA). An equal amount of immunoprecipitant was loaded for each sample. Resolved proteins were then transferred to a nitrocellulose membrane and detected with anti-human Zip8 antiserum.
Translocation of exogenous zinc into the cytosol.
Zinc uptake into lung epithelial cultures was measured using the cell-permeable fluorescent dye FluoZin-3 (Molecular Probes). Optimization of zinc was determined empirically at baseline following the addition of zinc sulfate (0–100 μM) to cells preloaded with FluoZin-3 and also compared with TNF-α-treated cultures. The intracellular zinc concentration was semiquantitatively measured by the mean intracellular fluorescence obtained from confocal microscopy for each field. The same conditions were evaluated by the addition of the zinc chelator N,N,N′,N′-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN; 20 μM) following the removal of excess zinc from the culture medium to confirm the specificity of zinc detection by FluoZin-3.
Modulation of Zip8 expression: short interfering RNA-mediated knockdown.
A 21mer short interfering (si)RNA target sequence directed against human Zip8 was purchased (QIAGEN) and transfected into BEAS-2B cells using HiPerFect transfection reagent (QIAGEN). The siRNA was compared against a nonsilencing control siRNA (Alexa fluor488) and Hs-MAPK1 control siRNA (positive control) in TNF-α-treated and nontreated cultures. Western blot analysis of membrane protein isolates were evaluated to determine the efficiency of Zip8 inhibition. The effect of siRNA-mediated Zip8 knockdown on cell survival was then evaluated following TNF-α exposure. BEAS-2B cultures were again transfected with siRNA control or siRNA targeted against Zip8. Cells were either exposed to TNF-α or no treatment for 24 h. All groups were preloaded with FluoZin-3, and one group was then exposed to 40 μM zinc sulfate for 30 min. Cytosolic zinc content was quantitatively evaluated within z-stack analysis by confocal fluorescent microscopy. Cell viability was also compared between control siRNA- and Zip8 siRNA-transfected cells following exposure to TNF-α by measuring lactate dehydrogenase (LDH) release into culture supernatants. Mitochondrial viability was simultaneously determined using a JC-1 MitoProbe Assay Kit and Flow Cytometry (Molecular Probes, Invitrogen) to assess mitochondrial membrane depolarization (using the manufacturer's instructions). Briefly, green fluorescence records an increase in monomer formation (normal), whereas red fluorescence records an increase in aggregate formation (consistent with mitochondrial membrane depolarization).
Localization of Zip8 in mitochondria.
Mitochondria of TNF-α-treated, nontransfected, and Zip8-transfected BEAS-2B cells were purified via a mitochondria isolation kit (Pierce). Cytosolic or mitochondrial proteins were isolated, quantified by protein assay (Bio-Rad), and analyzed by Western blot analysis with the anti-Zip8 antiserum. Samples were further probed with antibodies against E-cadherin (a plasma membrane-specific protein) and cytochrome oxidase II (a mitochondria-specific protein) to confirm sample purity. In addition, BEAS-2B cells were transfected and then stained with Mitotracker Green to visualize mitochondria (Molecular Probes). Colocalization of Zip8 and mitochondria was then evaluated by confocal microscopy.
Statistical significance was determined by calculating the P value using Student's t-test. A P value of <0.05 was considered significant unless otherwise stated. SDs were calculated for all data sets and are represented by error bars in figures.
Evaluation of zinc transporter gene expression.
Steady-state mRNA levels for each zinc transporter were evaluated in human lung epithelia using a quantitative RT-PCR-based screening assay. Lung epithelial mRNA expression profiles were determined in samples extracted from 12 different human donor cultures under baseline conditions (Table 1). The relative amount of RNA transcript was quantitatively evaluated for each transporter within each donor and compared between donors. We observed that many zinc transporters are expressed, to varying degrees, and with striking consistency in primary hLEC cultures across different donors. Eight of ten ZnT family members were consistently expressed in nonstimulated primary hLECs, with ZnT9 being the highest. Exporters ZnT3 and ZnT4 were expressed but at the detection threshold. Eleven of fourteen Zip family members were consistently constitutively expressed in nonstimulated hLECs at variable magnitudes. The highest expressed were Zip6 and 7 whereas the importers Zip3, Zip4, and Zip13 were barely detectable in most donor samples. Transporter expression was below the detection threshold for some transporters, resulting in fewer data points for certain transporters, as shown in Table 1.
Next, primary cultures were treated with TNF-α for 4- and 24-h intervals, and mRNA expression profiles were quantitatively evaluated. TNF-α was chosen as a prototypical proinflammatory cytokine since it appears early in the circulation at the onset of acute systemic inflammation (5, 30). Marginal changes were observed at 4 h for many transporters (data not shown); however, only Zip8 continued to show increased expression out to 24 h, whereas most others had returned toward baseline levels (Fig. 1A). Results are also shown as the average fold change from baseline obtained from 12 different human donor cultures after 24 h of TNF-α exposure for Zip (Fig. 1B) and ZnT (Fig. 1C) family members. Transporters that do not appear had no detectable change above baseline following 24-h TNF-α exposure or the expression was unable to be detected. An identical pattern of zinc transporter mRNA expression was observed with BEAS-2B cultures following exposure to TNF-α for 24 h, showing a substantial increase only in Zip8 relative to all other zinc transporters (Fig. 1D).
Increased vesicular zinc in response to TNF-α.
Since Zip8 mRNA expression was significantly modified by TNF-α, we next determined whether this may impact intracellular zinc concentrations. Initially, we observed that cultures maintained in zinc-free culture medium in the absence of TNF-α had relatively low detectable amounts of available zinc, as recorded by FluoZin-3 staining (Fig. 2A). Zinc supplementation into the culture medium resulted in a dose-responsive increase in cytosolic zinc as measured by fluorescent densitometric quantitation and by confocal microscopy. TNF-α-treated cultures compared with non-TNF-α-treated cultures had a consistently higher amount of cytosolic zinc at 0, 25, and 50 μM zinc (Fig. 2B). A decline in zinc content was observed at the highest zinc concentration studied (100 μM ZnSO4) in TNF-α-treated cells. Detection of available zinc by FluoZin-3 was quenched following the addition of the zinc chelator TPEN, thereby confirming that the fluorescence was specific for zinc. Cell viability and morphology did not change as a result of either zinc or TNF-α exposure as measured by trypan blue staining and LDH release (data not shown). Taken together, the increase in intracellular zinc correlates with increased Zip8 mRNA expression and demonstrates that a net gain in available zinc occurs in response to TNF-α exposure.
Induction of Zip8 protein.
Knowing that TNF-α significantly increases Zip8 mRNA expression and correlates with cytosolic zinc accumulation, we then characterized Zip8 protein expression. Bioinformatic analysis of the 459-amino acid protein sequence predicted an 8-transmembrane-spanning domain protein that includes 3 potential N-linked glycosylation sites and a rough endoplasmic reticulum retention motif. First, we compared cytosolic, membrane, and nuclear fractions obtained from primary hLECs or BEAS-2B cultures under baseline and TNF-α-treated conditions by Western blot analysis for Zip8 expression. Strikingly, we observed the expression of a highly inducible band at an approximate molecular mass of 140 kDa exclusively in the membrane fraction of TNF-α-stimulated cultures. The same molecular species was not routinely detected in nonstimulated cultures (Fig. 3A). An ∼55-kDa band was also routinely detected in the cytosolic fraction; however, its expression did not appear to be significantly influenced by TNF-α. No evidence of the 140-kDa band was observed in the nuclear fraction or in membrane fractions immunoblotted with preimmune antiserum. In addition, detection of the 140-kDa band disappeared following coincubation with the original peptide antigen used to generate the antiserum. Next, Zip8 was overexpressed as a DsRed-containing fusion protein in BEAS2B cultures followed by immunostaining of fixed cultures with the anti-hZip8 antiserum. No fluorescence was observed following immunostaining with the control preimmune antiserum (Fig. 3B,a). In contrast, z-axis analysis by confocal microscopy confirmed localization of Zip8 expression at the plasma membrane as well as expression within cytosolic vesicles in Zip8-transfected cultures (Fig. 3B,b, c, and d), indicating that Zip8 translocates to the plasma membrane as well as other locations in the cytosol. In addition, wild-type human Zip8 was overexpressed in confluent BEAS-2B cultures followed by an evaluation of cytosolic and membrane fractions by Western blot analysis. In agreement with previous observations, a 140-kDa band was consistently detected only in the membrane fraction of Zip8-transfected cells.
Localization of Zip8 and posttranslational processing.
Next, we conducted surface biotinylation labeling experiments to confirm the localization of Zip8 at the plasma membrane. Surface biotinylation of Zip8-transfected BEAS-2B cultures resulted in the detection of a dominant 140-kDa protein in the membrane fraction, thereby confirming the presence of the transporter at the plasma membrane (Fig. 4A). As expected, biotinylated Zip8 was not observed in the cytosol or in vector-control treated cultures. Since the mass of the membrane-associated protein that we observed was significantly greater than the predicted mass of the core protein, we determined whether Zip8 is subject to glycosylation. PNGase treatment of cell membrane extracts obtained from Zip8-transfected cultures consistently reduced the 140-kDa band to ∼80 kDa, consistent with a pattern expected for a N-linked glycoprotein (Fig. 4B). In addition, EndoH digestion reduced the presence of the 140-kDa protein, although we were unable to detect lower molecular weight species (Fig. 4B). Next, we exposed BEAS-2B cultures to the glycosylation inhibitor tunicamycin prior to TNF-α stimulation or Zip8 transfection and then evaluated membrane and cytosolic fractions by Western blot analysis. As shown previously, a major band of ∼140 kDa was observed in the membrane fraction of TNF-α-treated and Zip8-transfected cultures (Fig. 4C). Treatment with the glycosylation inhibitor tunicamycin significantly decreased the detection of the 140-kDa band in the membrane fraction, providing further evidence that Zip8 is glycosylated. Furthermore, tunicamycin treatment resulted in a moderate but consistent increase in a 55-kDa protein, consistent with the predicted mass of nonglycosylated Zip8, in the cytosolic fraction.
Impact of Zip8 siRNA-mediated knockdown.
The ability of siRNA to suppress Zip8 protein expression was first evaluated in nontreated and TNF-α-treated BEAS-2B cultures. Again, TNF-α significantly increased Zip8 expression in the membrane fraction. Expression of the 140-kDa band was significantly decreased in Zip8 siRNA-treated cells compared with control siRNA or non-siRNA-treated cultures. All densitometric analysis was standardized against β-actin (Fig. 5A). We then evaluated cytosolic zinc concentrations following incubation with FluoZin-3 under the same conditions in nontransfected, control siRNA-treated, and Zip8 siRNA-treated cultures. The Zip8 siRNA-transfected cells consistently had a significant reduction in fluorescent intensity, presumably due to decreased zinc import, as determined by confocal fluorescent analysis compared with the siRNA control (Fig. 5B).
Inhibition of Zip8 expression adversely affects epithelial viability.
Having observed that many zinc transporters are constitutively expressed in human lung epithelia but that only Zip8 is substantially induced in response to TNF-α, we wanted to determine whether Zip8 expression is an indispensable component of the lung epithelial response to inflammatory stress. To evaluate this, we first suppressed Zip8 expression in BEAS-2B cultures and then exposed cultures to TNF-α for a 24-h period. We observed that the inhibition of Zip8 expression in the presence of TNF-α resulted in a significant increase in LDH release compared with nontransfected or control siRNA-transfected cultures (Fig. 5C). Since LDH release is a relatively nonspecific measure of cytotoxicity, we next evaluated mitochondrial viability. Consistent with the increase in LDH release, suppression of Zip8 expression in the presence of TNF-α had a negative impact on mitochondrial function, as measured by a decrease in mitochondrial membrane potential compared with nontransfected and control siRNA-transfected cultures (Fig. 6). Specifically, Zip8 suppression in the presence of TNF-α resulted in a larger shift in cells from high to intermediate or low mitochondrial membrane potential, consistent with cell death, as determined by JC-1 staining. Pooled data from three independent experiments demonstrated a significant shift in mitochondrial membrane polarization under the influence of TNF-α (Fig. 6B). Knowing that Zip8 impacts mitochondrial function, we then evaluated whether Zip8 may have a direct role in causing this effect. First, using confocal microscopy of z-sections, we observed the presence of Zip8, expressed as a fusion protein, within the mitochondria of lung epithelia (Fig. 7A), which, upon further inspection of mitochondrial membrane preparations, was confirmed by the detection of the 140-kDa protein in both TNF-α-treated and Zip8-transfected cultures (Fig. 7B). Collectively, these findings indicate that in addition to increasing zinc import across the plasma membrane in response to TNF-α, Zip8 may also be involved in mitochondrial function.
Zip8 facilitates zinc-mediated cytoprotection.
To our knowledge, this is the first study to demonstrate that a human zinc transporter is directly involved with cytoprotection. We also observed that Zip8 is not naturally abundant within human lung epithelia but is significantly induced in response to a proinflammatory cytokine (TNF-α). This observation is noteworthy when considering that upon screening all Zip and ZnT family members, Zip8 was the only one of 24 family members that was substantially upregulated across multiple human donor specimens and the related human BEAS-2B upper airway epithelial cell line. This is consistent with the recent work of Liuzzi and colleagues (23), which demonstrated a rapid induction of murine ZIP14, an ortholog to human Zip8, in vivo in response to acute inflammation and infection. Based on this work and ours, we predict that Zip8 is unique, relative to other zinc transporters, in that it does not contribute significantly to dietary zinc maintenance but rather functions as a vital component of the innate host response against stress, thereby facilitating zinc-mediated protection. Furthermore, it is remarkable that upregulation of Zip8 alone is sufficient to protect the lung epithelium against the cytotoxic effects of TNF-α, a potent cytokine known to invoke cell death across different cells.
Zip8 is induced as part of the innate host response.
The first recorded annotation for Zip8 was by the name of Bacillus calmette-guerin-induced gene in monocyte clone 103 (BIGM103), showing that gene expression is induced in primary human monocytes following exposure to the Bacillus calmette-guerin cell wall skeleton, a pathogen associated molecular pattern derived from the gram-negative bacterium (4). Interestingly, BIGM103 was not constitutively abundant but induced by inflammatory mediators including LPS and TNF-α. Gene expression was also observed to be abundant in the lung compared with most other tissues, although this was not further examined. Our work expands upon this observation and demonstrates that Zip8 is the only zinc transporter strongly induced by TNF-α in primary differentiated human lung epithelia, a parenchymal cell type that plays a vital role in maintaining tissue integrity in response to stress. Furthermore, we demonstrate that Zip8 is glycosylated, in accord with a previous report showing that overexpression of ZIP8, the murine ortholog, yielded a high-molecular-weight, membrane-bound protein that facilitated zinc import into cells. Interestingly, ZIP8 expression was also reported to localize within other intracellular compartments (14). In our investigation, we also consistently observed a specific 55-kDa band that was associated exclusively with cytosol preparations. Whether this represents a nonglycosylated monomeric form of Zip8 or a posttranslationally modified form will require further investigation. In this investigation, we demonstrate that the induction of Zip8 expression resulted in a net gain of available zinc within the cytosol that occurs at least in part following increased expression of Zip8 at the plasma membrane. Also consistent with previous reports, the zinc is associated with vesicles (11, 12). This is consistent with previous studies (6, 27) that also identified compartmentalized expression of zinc transporters within vesicles. Whether the newly formed zinc pool is being temporarily stored, sequestered to prevent toxicity, or redirected to serve cytoprotective functions within organelles, such as mitochondria, remains to be resolved.
Zinc transport and mitochondrial protection.
We provide novel evidence suggesting that Zip8 also physically translocates to mitochondria. Taken together, it suggests that Zip8-mediated zinc transport, the mitochondria, and cytoprotection are connected. Our findings support the possibility for a dynamic process in which Zip8 functions at the plasma membrane to increase available zinc while also playing a role in modulating zinc content in mitochondria. Consistent with these observations are recent studies demonstrating that the movement of zinc between cytosolic and mitochondrial pools is functionally significant (28) and that it occurs via a transporter-mediated process, although a specific transporter was not identified (10, 15, 24). Previous studies (20, 29) have also demonstrated expression and function of zinc transporters in other cell organelles in vertebrates, thereby supporting our findings that Zip8 may have a direct role in maintaining zinc concentrations within the mitochondria. Elucidation of the mechanism(s) to account for Zip8-mediated cytoprotection and maintenance of mitochondrial membrane polarization and function will require further study.
Zinc transport and signal transduction: a potential connection.
We (3) recently reported that zinc sustains the lung epithelium under conditions of inflammatory stress and that zinc depletion enhances cell susceptibility to apoptosis and mechanical dysfunction, specifically apoptosis initiated by cytokines. This supports a previous investigation (31) demonstrating the detrimental effects of zinc deficiency in lung epithelia exposed to oxidant-mediated stress. We also observed that cellular zinc uptake rapidly activates signal transduction pathways, including the phosphatidylinositol 3-kinase/Akt pathway, that protect the epithelium, whereas zinc depletion suppresses these pathways, thereby promoting cell death. These observations support work by others showing zinc-induced activation of p70 S6 kinase (18), MAPK (13), and the EGF receptor pathway (33) across different cell types. Based on this, we postulate that in addition to zinc transport, Zip8 may provide additional vital regulatory functions that directly or indirectly involve the activation of signal transduction pathways that regulate cell survival. Taken together, these results suggest that zinc transport is highly coordinated and connected to internal signaling networks designed to sense the outside environment and rapidly direct the innate host response toward cytoprotection.
We anticipate that further inspection of Zip8, and possibly other zinc transporters, may establish a critical link between susceptibility to tissue injury (by virtue of heterogeneity within genes that regulate metal homeostasis) and the environment, namely, dietary zinc insufficiency, which is estimated to affect millions within the United States and even larger populations abroad (32). Therefore, it will be of interest to determine if zinc deficiency, in conjunction with alterations in genetic composition, are predisposing factors that contribute to cellular dysfunction and lung pathogenesis.
This work was supported by National Institutes of Health (NIH) Grants R01-HL-086981-01 (to D. L. Knoell) and F32-HL-086186-01 (to B. Besecker) and in part by NIH Grant GM-61390 (subcontract to W. Sadee). The authors also give special thanks to Lifeline of Ohio Tissue Procurement Agency.
We are indebted to Timothy Dalton, Mikhail Gavrilin, and Mark Wewers for intellectual support and technical advice.
↵* B. Besecker and S. Bao contributed equally to this work.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society