Zinc (Zn) is an essential heavy metal that is incorporated into a number of human Zn metalloproteins. Zn plays important roles in nucleic acid metabolism, cell replication, and tissue repair and growth. Zn deficiency is associated with a range of pathological conditions, including impaired immunity, retarded growth, brain development disorders and delayed wound healing. Moreover, many reports have suggested that Zn is involved in cancer development and levels of Zn in serum and malignant tissues of patients with various types of cancer are abnormal. Zn may directly affect tumor cells by regulating gene expression profiles and/or cell viability, both of which are mediated in part by tumor-induced changes in Zn transporter expression. On the other hand, Zn may indirectly influence tumor cells by affecting processes within the cancer microenvironment, including immune responses; the functions and/or activity levels of immune cells that attack tumor cells are influenced by the intracellular Zn concentrations within those cells. In both cases, Zn contributes to intracellular metal homeostasis and/or signal transduction in tumor and immune cells. In this review article, we will summarize the current understanding of the roles of Zn homeostasis and signaling primarily in immune cells, with a discussion of the contributions of these processes to oncogenesis. (Cancer Sci 2008; 99: 1515–1522)
The first discussion of the critical roles of zinc (Zn) in an organism was its reported requirement for the growth of Aspergillus niger.(1) It has since been shown that Zn deficiency results in impaired growth, loss of hair, thickening and hyperkeratinization of the epidermis, and testicular atrophy in humans.(2) Fujii suggested that Zn is involved in mitosis due to its presence in nucleoli during cell division.(3,4) Prasad published a description of a syndrome that included iron-deficiency anemia, hepatosplenomegaly hypogonadism, dwarfism and geophagia, speculating that Zn deficiency may cause growth retardation and hypogonadism.(5) It has been estimated that more than 2 × 109 people have a nutritional deficiency for Zn in developing countries, which can result in growth retardation, immune dysfunction and cognitive impairment.(6) These effects are reversible with Zn supplementation. Conditioned Zn deficiencies are also known to occur in many diseases and abnormal conditions, including malabsorption syndrome, chronic liver and renal diseases, sickle cell disease, excessive intake of alcohol, malignancies and other chronic debilitating conditions.(6)
Zinc is an essential trace element and a essential structural component of a great number of proteins, including intracellular signaling enzymes and transcription factors.(7,8) In fact, Zn is required for the activity of more than 300 enzymes; as such, it participates in many enzymatic and metabolic functions in the body. More than 2000 transcription factors that work to regulate gene expression require Zn to maintain their structural integrity and bind to DNA.(6) Thus, it is not surprising that the intracellular Zn concentration is tightly controlled by Zn transporters, Zn-binding molecules and Zn sensors.(9–14) Interestingly, a number of studies have found that the expression levels of Zn transporters in human tumors correlate with their malignancy, suggesting that alteration of intracellular Zn homeostasis can contribute to the severity of cancer.(9,15,16)
How do altered Zn levels affect the behaviors of the cells, which in turn result in localized and/or systemic abnormalities? Zn is critical for the activities of various enzymes that contribute to cellular signaling pathways as well as transcription factors. Zn is believed to maintain the structures of and/or serve as a cofactor in these proteins by binding tightly to various Zn-binding motifs, including Zn finger (ZnF), ring finger, and LIM domains.(7,8) Moreover, we have recently shown that, in addition to its structural role, Zn may also function as an intracellular signaling molecule.
Physiological and pathological roles of Zn homeostasis, transporters and metallothioneins
Phenotypic expression of the rare autosomal recessive disorder, acrodermatitis enteropathica (AE), was found to be due to defects in Zn metabolism.(17) Mutations in ZIP4 were found to be causative of AE. ZIP4 encodes a member of the Zrt-Irt-like protein (ZIP)/SLC39 Zn transporter family that is expressed in intestinal organs, indicating that the intestinal absorption of Zn is a critical process.(18) Various organs are affected by severe Zn deficiency, including the epidermal, gastrointestinal, central nervous, immune, skeletal and reproductive systems.(19) Furthermore, the ZnT4 Zn transporter, which is expressed in breast epithelial cells, is responsible for the inherited Zn deficiency observed in lm mice, indicating that ZnT4-mediated supplies of Zn in breast milk are critical for infant development.(20) Moreover, studies of human breast-fed infants have revealed a potential role for ZnT2 in the Zn content of milk.(21) These findings indicate that transporter regulation of Zn homeostasis plays important cellular roles, disruption of which may induce disease states. The intracellular Zn concentration is tightly controlled by Zn importers (ZIP/SLC39),(9) exporters (ZNT/SLC30),(10) and binding proteins such as metallothioneins (Fig. 1).(11) In addition, Zn-sensing molecules, such as metal-responsive element-binding transcription factor-1 (MTF-1), regulate the expression of these molecules.(12)
Fourteen members of the ZIP family, which were first discovered in Saccharomyces cerevisiae (Zrt proteins) and Arabidopsis thaliana (Irt proteins),(15,16) have been reported in mammals, and some knockout mouse lines deficient for various ZIP proteins have been reported (Table 1). Mice lacking ZIP1, ZIP2 or ZIP3 show abnormal embryogenesis specifically under Zn-limiting conditions. Homozygous ZIP4 knockout mouse embryos die during early development, whereas heterozygosity causes hypersensitivity to Zn deficiencies, as is observed in AE patients.(18,22) ZIP6/Liv1 from zebrafish has been reported to control the epithelial–mesenchymal transition (EMT) following STAT3 activation,(23) and the Drosophila gene fear of intimacy (foi), which is similar to mammalian ZIP6 and ZIP10, is essential for proper gonad formation, E-cadherin expression and glial cell migration,(24–26) suggesting that ZIP6/Liv1 and/or ZIP10 have important roles in cell migration. In fact, it has been suggested that ZIP10 is involved in the invasive behavior of breast cancer cells.(27) We will return to this matter in the last section of this review. ZIP7 (KE4) was discovered during the characterization of the major histocompatibility complex (MHC) region on mouse chromosome 17,(28) and mapped to the human leukocyte antigen (HLA) class II region on human chromosome 6.(29) A Drosophila counterpart of ZIP7, Catsup, is reported to control melanin synthesis,(30) whereas the plant homolog IAR1 regulates root elongation by controlling auxin conjugate sensitivity.(31)
Table 1. Mutations of Zn transporters and methallothioneins
Cation diffusion facilitator (CDF) confers metal resistance to many eukaryotic cell types;(32) 10 reported members of this protein family have been reported in mammals (named the Zn transporter [ZnT] family), some of which have been targeted in knockout mouse lines (Table 1). ZnT1 knockout mice are embryonic lethal. CDF1, a nematode ZnT1 ortholog, positively regulates Ras–Raf–MEK–ERK signal transduction by promoting Zn efflux and reducing the concentration of cytosolic Zn.(33) ZnT3 knockout mice are prone to seize in response to kainic acid treatment. lm mice carry a nonsense mutation in the ZnT4 gene and produce Zn-deficient milk. ZnT5 knockout mice show poor growth, osteopenia, low bodyfat, muscle weakness and male-specific cardiac death. A point mutation in the human ZnT2 gene suggests that ZnT2 functions to enhance the Zn content of milk. A recent genome-wide association study identified the region containing ZnT8 as a risk locus for type 2 diabetes.(34) Interestingly, ZnT8 is expressed exclusively in pancreatic β cells.
Metallothioneins (MTs) – small cystein-rich proteins that bind Zn as well as other metal ions – are thought to be responsible for regulating the intracellular Zn concentration and for nonessential heavy metal detoxification. When the concentration of intracellular free Zn reaches a threshold, activation of MTF-1 induces the expression of MT, which then sequester the Zn ions.(35) Therefore, MT serve as biochemical devices that control the concentration of free Zn by sequestering Zn and releasing it in response to other biochemical events, such as oxidative signaling. MT also participate in immune responses; macrophages from MT-KO (MT-I and MT-II double knockout) mice show defects in phagocytosis and antigen presentation.(36)
Roles of Zn in immune responses
Because Zn deficiency is associated with many chronic diseases, experiments examining Zn homeostasis have employed mouse models to determine the impact of nutritional deficiency of a single element on immune function at the cellular and molecular levels.(44) It is important to point out that chronic diseases, such as gastrointestinal disorders, renal disease, sickle cell anemia, cirrhosis, some cancers, cystic fibrosis, pancreatic insufficiency and autoimmune arthritis, have been shown to lead to suboptimal Zn status in humans.(45–52) These disease states are associated with increased infections of prolonged duration, a clear indication of compromised immunity, which implies that Zn may contribute to immune cell homeostasis in vivo.(47,50) Indeed, various studies using animal models of Zn deficiency have confirmed that decreased levels of Zn induce thymic atrophy, lymphopenia, and compromised cell- and antibody-mediated immune responses. In this section, we will comment on the relationship between Zn and immunity, primarily through an examination of adaptive immunity including T cells and dendritic cells. Moreover, because immune responses against tumors may be affected by Zn deficiency, Zn may have indirect effects on tumorigenesis.
Zinc deficiency, which affects some patients with carcinomas, is reported to induce thymic atrophy. T cells, a critical antitumor population, develop in the thymus. Moreover, glucocorticoids, in particular corticosterone, are chronically elevated in Zn-deficient mice and adrenalectomies or removal of these steroids prevent the thymus from atrophying under Zn-deficient conditions.(53,54) Additionally, Zn-deficient diets cause substantial reductions in the number of CD4+CD8+ thymocytes, thymic cells known to show high rates of apoptosis in response to glucocorticoids.(55) Thus, chronic overproduction of glucocorticoids in Zn-deficient individuals may accelerate the rate of apoptosis in the thymus, which would reduce the number of peripheral T cells. On the other hand, mature CD4+ and CD8+ T cells are resistant to Zn deficiency and survive well in an otherwise atrophying thymus.(55) Interestingly, in an experimental mouse model, Zn deficiency caused an imbalance between the peripheral functions of the T-helper (Th)1 and Th2 cell populations; production of γ-interferon (IFN-γ) and interleukin (IL)-2 (products of Th1 cells) decreased, whereas production of IL-4, IL-6 and IL-10 (products of Th2 cells) were not affected.(56) Of note, Th1 cells are known to play major roles in tumor suppression.
Another mechanism by which Zn deficiency contributes to impaired immunity, including peripheral lymphopenia and compromised immune responses, is altered gene expression profiles in various cell types including T cells. Indeed, mice with modest Zn deficiencies show changes in the expression levels of 1200 genes in their T cells.(57) Altered gene expression may affect the survival and/or responses of T cells and dendritic cells, important cellular mediators of T-cell homeostasis in vivo. In fact, culturing Th0 and Th1 cell lines in low-Zn media followed by mitogenic stimulation leads to reduced expression of IL-2 and IFN-γ mRNA, which adversely affects the functional capacity of these cells.(58) In addition to T cells, other types of immune cells show specific responses to Zn deficiency. HL-60 cells, a human myeloid-like precursor cell line, survive in cultures containing low concentrations of Zn.(58) Interestingly, exposure of mouse dendritic cells to the bacterial endotoxin lipopolysaccharide (LPS), a toll-like receptor (TLR) ligand, leads to a decrease in the intracellular free Zn concentration (Fig. 2). Moreover, artificially depleting the intracellular Zn using a Zn chelator triggers dendritic cell maturation. On the other hand, artificially elevating intracellular Zn levels suppresses the ability of dendritic cells to respond to LPS. In fact, Zn appears to suppress the surface expression of MHC class II molecules because Zn is required for the endocytosis of MHC class II molecules expressed on the plasma membrane, and Zn inhibits MHC class II vesicle trafficking to the plasma membrane from the perinuclear region. LPS affects the expression of a number of Zn import and export transporter molecules, resulting in a net increase in Zn transport out of cells. Importantly, overexpression of the Zn importer ZIP6, expression of which is reduced by LPS stimulation, suppresses dendritic cell maturation followed by inhibition of the stimulatory activities of CD4+ T cells. A similar effect of Zn has been observed in vivo; injections of LPS induce reduced intracellular free Zn levels and ZIP6 expression in dendritic cells and treatment with Zn-depleting agents leads to increased dendritic cell maturation.(59) These results clearly show that intracellular Zn homeostasis is critically involved in the maturation of dendritic cells, an important step for T-cell activation (Fig. 3). In any case, intracellular Zn homeostasis in immune cells, which may be affected by tumor progression, plays a key role in responses of these cells, including those against tumors.
As discussed above, Zn is known to be a critical structural constituent of a great number of proteins, including enzymes from cellular signaling pathways and transcription factors.(7,8) In addition, Zn may also function as a signaling molecule. Microfluorescence imaging of Zn dynamics following presynaptic stimulation of hippocampal mossy fibers shows Zn release from terminal vesicles into the surrounding milieu.(60–62) The Zn is then taken up into the cytoplasm of neighboring cells through gated Zn channels. Rapid Zn influx through Ca2+ permeable AMPA/kainate (Ca-A/K) channels triggers the generation of reactive oxygen species, which are potently neurotoxic.(63) As such, Zn functions similarly to neurotransmitters, which are stored in membrane-enclosed synaptic vesicles and released by exocytosis to bind transmitter-gated ion channels and activate postsynaptic cells.(60,64–66) Using site-directed mutagenesis, Hosie et al. identified a pair of Zn-binding sites and completed the characterization of a third Zn-binding site in the GABA receptor.(67) Hirzel et al. used knockin mice carrying a D80A mutation (constructed in a Zn-binding site) in the glycine receptor (GlyR) a1 subunit gene (Glra1) to show that Zn modulates neurotransmission.(68) In the latter report, the authors showed that the hyperekplexia phenotype of the Glra1(D80A) mice was due to the loss of Zn-mediated potentiation of a1 subunit-containing GlyR. Therefore, synaptic Zn is essential for proper functioning of glycinergic neurotransmission in vivo. Zn is also known to inhibit NMDA receptor activity via two mechanisms: voltage-dependent channel blockade and voltage-independent reduction in the probability of channel opening.(69–71) Huang et al.(72) recently reported that Zn-mediated transactivation of TrkB potentiates hippocampal mossy fiber-CA3 pyramid synapses.
Thus, it is likely that Zn acts as a neurotransmitter. The role of Zn as a neurotransmitter differs from the conventional concept of a secondary messenger in cells; neurotransmitters carry information between cells, whereas secondary messengers function intracellularly. Cyclic adenosine monophosphate (cAMP) was discovered in 1957 as the first intracellular secondary messenger. A limited number of secondary messenger species have now been identified, including Ca2+, cAMP, cyclic guanylic acid (cGMP), NO, lipid mediators, G-proteins, protein kinases, protein phosphatases and nuclear receptors.(73) We hypothesize that Zn acts as a secondary messenger as well. We have shown that the STAT3-ZIP6 signaling cascade is critically involved in the EMT of zebrafish cells and is required for the nuclear localization of Snail, a ZnF-containing repressor (Fig. 4).(23) Moreover the nuclear localization of Snail is dependent on both its ZnF domain(23,74) and the Zn transporter ZIP6, suggesting that Zn may act as an intracellular signaling molecule. It is also known that Zn regulates cyclic nucleotide signaling, a conventional secondary messenger system. Zn suppresses LPS-induced tumor necrosis factor (TNF)-α and IL-1β release from monocytes. Moreover, this inhibitory effect is dependent on suppression of phosphodiesterase-mediated hydrolyzation of cyclic nucleotides into 5′-nucleotide monophosphate followed by an increase of the intracellular cGMP level. Because the NO donor S-nitrosocystein (SNOC) also inhibits LPS-induced TNF-α and IL-1β release, it is possible that increased levels of intracellular free Zn in response to SNOC play a role in this inhibition via augmentation of the cGMP level.(75) Because cytokines can induce NO production to enhance the level of intracellular free Zn,(76) Zn may act as a bridge between secondary signaling mediated by NO and cGMP. Additionally, Huberman's group showed in macrophage cell lines that the nuclear Zn concentration increases within 15 min following treatment with phorbol myristate acetate in a manner dependent on PKCβ.(77) Interestingly, we have shown that TLR-mediated signaling induces a decrease in the intracellular free Zn concentration in dendritic cells, and this decrease is required for dendritic cell activation and subsequent CD4+ T-cell activation.(59) LPS-induced decreases in the concentration of intracellular free Zn are dependent on changes in the expression profiles of Zn transporters; expression of ZIP family members are downregulated, whereas those of the ZnT transporters are upregulated. The results show that extracellular stimuli affect intracellular free Zn concentrations and these changes are critically involved in the biological expression of extracellular stimuli. Taken together, these data support a role for Zn as an intracellular signaling molecule. We have also shown that extracellular stimulation of mast cells induces an increase in intracellular free Zn levels within minutes of the stimulation, a phenomenon that we have named the ‘Zn wave’.(78) The Zn wave originates from the endoplasmic reticulum and/or the surrounding area. In mast cells, the Zn wave is dependent on Ca2+ influx and mitogen-activated protein kinase activation (Fig. 5). Because extracellular Zn does not contribute to the Zn wave and the wave is induced several minutes after FcɛR1-stimulation, Zn seems to be functioning as an intracellular secondary messenger under these conditions. This conclusion is supported by the following results: (i) extracellular stimuli, such those that cross-link FcɛRI receptors, directly induce increases in the Zn wave; (ii) an intracellular compartment, possibly the endoplasmic reticulum compartment, serves as the source of Zn; and (iii) free Zn at a level similar to that observed in the Zn wave can affect intracellular signaling molecules, such as tyrosine phosphatases, and therefore may modulate the final output triggered by extracellular stimuli.
An important difference between the Zn wave in mast cells described above and other findings in zebrafish cells(23) and dendritic cells(59) is that the Zn wave is observed several minutes after stimulation, whereas the latter observations are critically dependent on transcriptional regulation of Zn transporters and are therefore detected several hours after stimulation. Thus, it appears that intracellular Zn signaling comprises an early component, such as the Zn wave, which is directly induced by extracellular stimuli, and a late component that is dependent on transcriptional regulation of transporter expression (Fig. 6). For the former, Zn acts as a conventional intracellular secondary messenger capable of transducing an extracellular stimulus into intracellular events. Collectively, these results support the idea that Zn is a secondary messenger/signaling ion that has the potential to influence many aspects of cellular signaling through its effects on a range of Zn-binding proteins.
Zinc and tumorigenesis
In this section, we discuss the cellular effects of altered intracellular Zn homeostasis. As described in the previous sections, the intracellular Zn concentration is tightly controlled by Zn transporters, Zn-binding molecules and Zn-sensing molecules.(9–14) It has been pointed out, however, that the levels of Zn in sera and malignant tissues are abnormal in patients with various inflammatory diseases and tumors; serum Zn concentrations are lower in patients with such autoimmune diseases as rheumatoid arthritis.(79) The relationship between tumor development and Zn levels, however, appears to be complicated. Indeed, Zn levels are reduced in patients suffering from carcinomas of the liver, gallbladder, digestive tract or prostate,(80–82) whereas breast cancer patients show decreased and elevated Zn levels in sera and malignant tissues, respectively.(82–84)
For immune cells, Zn-deficient conditions result in reduced natural killer cell-mediated cytotoxic activity, antibody-mediated responses, and host defense against pathogens and tumors,(85–87) whereas excessive levels of Zn are cytotoxic; Zn induces apoptosis in lymphocytes, including T and B cells.(88,89) Zn may have similar effects on tumor cells. Tumors need Zn to survive and grow, whereas excess Zn may induce tumor cell apoptosis, although the sensitivities of the different types of tumors are likely to vary. Indeed, some tumors are resistant to high Zn concentrations. In addition, Zn levels are affected by the microenvironment surrounding the cancerous tissue: (i) many cytokines and growth factors that are produced in these micrsoenvironments, including IL-6, hepatocyte growth factor, epidermal growth factor and TNF-α, directly or indirectly affect the expression profiles of various Zn transporters; (ii) mast cells, an important cell type in cancer microenvironments, contain high levels of Zn in granules that are released into the surround milieu; and (iii) oxidation/reduction reactions in these environments markedly influence the intracellular free Zn concentration. Moreover, it is likely that the activities of the diverse set of enzymes and transcription factors that require Zn to function, including matrix metalloproteases, are affected by the altered Zn concentrations in cancer microenvironments.
The relationships between the expression levels of Zn transporters, especially ZIP family members, and tumor malignancy have recently received a great deal of attention. There are at least 14 human ZIP transporters, which are believed to allow Zn influx into the cytosol.(9,15,16) Some of these proteins are thought to be involved in cancer progression; ZIP1 is reported to be a suppressor of prostate cancer(90) and some evidence suggests ZIP6 contributes to the metastasis of breast cancer to the lymph node(91,92) although this issue is controversial.(5,8,27) Interestingly, we showed that the STAT3-ZIP6 signaling cascade is critically involved in the EMT of organizer cells and is required for the nuclear localization of Snail, a ZnF-containing repressor.(23) STAT3 is activated in the organizer of zebrafish to allow the movements of cells during gastrulation. The requirement for STAT3 is cell autonomous for the anterior migration of the organizer cells, and non-cell autonomous for the convergence of neighboring cells.(23) It is possible that ZIP6 acts as a downstream target of STAT3 to induce tumor cell migration followed by their metastasis. Moreover, we showed that ZIP6 is essential for the nuclear localization of the ZnF protein Snail, a master regulator of the EMT that suppresses the expression of cadherin.(94,95) These results establish a molecular link between STAT3, ZIP6 and Snail during the EMT, which is not only important for embryonic development including the migration of the organizer cells but also for cancer metastasis.(96) Indeed, the processes of embryonic development that rely on EMT for the generation of new tissue types have been reported to be co-opted by tumors, resulting in uncontrolled proliferation and spatial expansion.(7,81) We also demonstrated that the migratory activity of metastatic breast cancer cells was inhibited by knockdown of ZIP10 expression and Zn chelation. Importantly, analysis of clinical samples showed that breast cancers with lymph node metastases expressed significantly higher levels of the Zn transporter, ZIP10, than those without lymph node metastases.(27) Furthermore, ZIP10 was recently shown to mediate Zn uptake and to act as a membrane transporter in vivo.(97) Thus, altered intracellular Zn homeostasis due to changes in Zn transporter expression may be a key factor that determines tumor malignancy. Moreover, other types of proteins that regulate Zn levels, such as MT and MTF-1, may contribute to tumor behavior. However, because systemic changes in Zn status will regulate the activities of immune cells, which in turn will affect tumors, serum Zn concentrations may not be a useful diagnostic marker clinically. Future studies should assess the possibility of using Zn levels to control tumor metastasis or tumor malignancy.