Zinc ions have a potential to attenuate both Ni ion uptake and Ni ion-induced inflammation

Nickel ions (Ni2+) are eluted from various metallic materials, such as medical devices implanted in human tissues. Previous studies have shown that Ni2+ enters inflammatory cells inducing inflammation. However, the regulation of Ni2+ uptake in cells has not yet been reported in detail. In the present study, we investigated the effects of various divalent cations on Ni2+ uptake and Ni2+-induced interleukin (IL)-8 production in the human monocytic cell line, THP-1. We demonstrated that ZnCl2, MnCl2, and CoCl2 inhibited the Ni2+ uptake, while CuCl2, FeCl2, MgCl2, and divalent metal transporter (DMT)-1 inhibitor, Chlorazol Black, did not. Furthermore, ZnCl2 inhibited Ni2+-induced IL-8 production, correlating with the inhibition of Ni2+ uptake. These results suggested that Ni2+ uptake occurred through Zn2+, Mn2+, and Co2+-sensitive transporters and that the inhibition of Ni2+ uptake resulted in the inhibition of IL-8 production. Furthermore, using an Ni wire-implanted mouse model, we found that Ni wire-induced expression of mouse macrophage inflammatory protein-2 (MIP-2) and cyclooxygenase-2 (COX-2) mRNA in the skin tissue surrounding the wire were enhanced by low Zn conditions. These results suggested that the physiological concentration of Zn2+ modulates Ni2+ uptake by inflammatory cells, and a Zn deficient state might increase sensitivity to Ni.

ions. Each of these members exhibits specificity toward a specific metal. However, the metal specificity of the transporter involved in Ni 2+ uptake remains unclear.
Ni 2+ uptake in cells and nuclei in the human monocytic cell line, THP-1, has already been reported 18 . THP-1 cells also have the ability to produce IL-8 by treatment with Ni compounds 19,20 . Therefore, using THP-1 cells, we examined whether the competition between Ni 2+ and other ions affected IL-8 production. Especially, to assess the accumulation of metals in the cells and Ni 2+ elution in the tissues precisely, we used inductively coupled plasma mass spectrometry (ICP-MS), a highly sensitive and efficient analysis technique for detecting various metal ions. In this study, we found that the physiological concentration of Zn 2+ affected the uptake of Ni 2+ by THP-1 cells and the sensitivity of mouse to Ni 2+ .

Results
NiCl 2 -stimulated increase in Ni 2+ content and IL-8 production in THP-1 cells. THP-1 cells were treated with various concentrations of NiCl 2 for 24 h and the Ni 2+ content in the cells and IL-8 level in the medium were determined. Both Ni 2+ content and IL-8 production increased in a NiCl 2 concentration-dependent manner ( Fig. 1a and b). As IL-8 production was significantly induced by NiCl 2 at the concentration of ≥0.2 mM (Fig. 1b), 0.2 mM NiCl 2 was used in all the experiments. Ni 2+ content in the cells increased in a time-dependent manner (Fig. 1c), consistent with the concentration-dependent increase in the cells, and IL-8 level in the medium increased significantly from the 4-h mark (Fig. 1d). The incubation of THP-1 cells in 0.2 mM NiCl 2 for 24 h did not affect the viability as determined by the MTT assay (data not shown).
Effects of metal ions on the uptake of Ni ions. THP-1 cells were treated with 0.2 mM NiCl 2 in the presence of various divalent cations (0.03 mM), including Zn 2+ , Mg 2+ , Fe 2+ , Co 2+ , Cu 2+ , or Mn 2+ , added as dichloride salts. The Ni 2+ content in the cells after 24 h of incubation was determined by ICP-MS. The increase in the intracellular Ni 2+ content was inhibited by ZnCl 2 , CoCl 2 , and MnCl 2 (Fig. 2a). In contrast, the increase in Ni 2+ content was not inhibited by the divalent metal transporter 1 (DMT1) inhibitor, Chlorazol Black (Fig. 2b). Because Ni 2+ activates Toll-like receptor 4 (TLR4), the effects of the TLR4 inhibitor, TAK-242, on Ni 2+ uptake were determined. The results suggested that TAK-242 did not affect Ni 2+ uptake (Fig. 2c), suggesting that TLR4 activation was not involved in Ni 2+ uptake. To confirm whether ZnCl 2 also inhibits Ni 2+ uptake in the other cell lines, a human monocytic cell line, U937 (Fig. 2d), and a human embryonic kidney cell line, HEK293 (Fig. 2e) were treated with 0.2 mM NiCl 2 in the presence of 0.03 mM ZnCl 2 . Ni 2+ content in these cells was increased by NiCl 2 treatment, and this increase was reduced by ZnCl 2 . These findings suggested that Ni 2+ uptake occurred generally via a Zn 2+ -sensitive transporter. Cellular compartmentalization of Ni ions and the effects of ZnCl 2 . To confirm whether Ni 2+ entered the cells or was bound to the cell membrane, the cellular compartmentalization of Ni 2+ was determined by the fluorescence indicator, Newport Green. This compound was used to detect Ni 2+ in the immune cells in a previous study 21 . Although Newport Green could bind to both Zn 2+ and Ni 2+ , the concentration of ZnCl 2 used in this experiment, 0.01 mM, did not apparently increase the fluorescence. In contrast, treatment with 0.2 mM NiCl 2 increased the fluorescence in the cells, indicating that Ni 2+ entered the cells. Consistent with the data of ICP-MS, treatment with ZnCl 2 inhibited the NiCl 2 -induced increase in fluorescence (Fig. 3), indicating that even at a low concentration, Zn 2+ inhibited Ni 2+ uptake.
Effects of ZnCl 2 and MnCl 2 on Ni 2+ -induced IL-8 production. To clarify whether the inhibition of Ni 2+ uptake resulted in the inhibition of IL-8 production, the cells were treated with 0.2 mM NiCl 2 in the presence of 0.01 and 0.03 mM ZnCl 2 and MnCl 2 . The increase in the Ni 2+ content was reduced by ZnCl 2 and MnCl 2 in a concentration-dependent manner ( Fig. 4a and d). Treatment with ZnCl 2 did not affect the Zn 2+ content in the cells, but that with MnCl 2 increased the Mn 2+ content. In these conditions, IL-8 production was also inhibited by these cations (Fig. 4c and f). MnCl 2 at 0.03 mM concentration slightly induced IL-8 production by itself, both in the presence and absence of NiCl 2 (Fig. 4f), indicating that Mn 2+ has a weak ability to induce IL-8 production by itself.

Effects of ZnCl 2 on CoCl 2 -and LPS-induced IL-8 production.
To confirm the selectivity of the action of ZnCl 2 , the effects of ZnCl 2 on CoCl 2 -and LPS-induced IL-8 production were examined. Treatment with 0.2 mM CoCl 2 increased Co 2+ content in the cells and IL-8 production. ZnCl 2 (0.01 and 0.03 mM) inhibited this increase in a dose-dependent manner ( Fig. 5a and b). In contrast, the same concentrations of ZnCl 2 and MnCl 2 did not inhibit LPS-induced IL-8 production ( Fig. 5c and d), indicating that Zn 2+ did not affect the signaling pathway inducing IL-8 expression in this case.

Enhancement of Ni wire-induced inflammation in a Zn-deficient state. Finally, we examined
whether the physiological concentration of Zn 2+ affects Ni 2+ -induced inflammation in low Zn diet-fed mice. Consumption of the low-Zn diet for two weeks reduced Zn 2+ levels in the serum to one third of the normal levels ( Fig. 6a), but the level in the skin tissues was unchanged (Fig. 6b). As previously reported, implantation of the Ni wire on the back of mice induced inflammation, visible as vasodilation/erythema (Fig. 6c), edema (Fig. 6d), and the expression of inflammatory proteins such as MIP-2 ( Fig. 6e) and COX-2 ( Fig. 6f). In the mice fed with low-Zn diet for 2 weeks, interestingly, the Ni 2+ -induced expression of MIP-2 and COX-2 was significantly higher than that in the control group ( Fig. 6e and f). The concentration of Ni 2+ in the serum and skin tissues was also higher in the low-Zn diet group than in the control group ( Fig. 6g and h), indicating that enhanced inflammation promoted Ni 2+ elution.

Discussion
In this study, we found that Ni 2+ entered the THP-1 cells in a Zn 2+ , Mn 2+ , and Co 2+ -sensitive manner, and that Zn 2+ inhibited Ni 2+ uptake, resulting in reduced IL-8 production. More importantly, we showed that Ni 2+ -induced inflammation was enhanced in a systemic low-Zn state. Our findings suggest that maintaining a normal level of Zn 2+ is important to reduce the incidence of Ni-induced inflammation and allergy.
As expected, the incubation of THP-1 in the presence of NiCl 2, elicited an increase in intracellular Ni 2+ level and IL-8 production. The accumulation of Ni 2+ in THP-1 cells was induced rapidly until 4 h and then it accumulated gradually. The findings, consistent with those in the previous report 18 , suggested that the increase was regulated by Ni 2+ influx and efflux balance. The increase in Ni 2+ level in the cells was antagonized by Zn 2+ , Mn 2+ , and Co 2+ , indicating the involvement of transporter(s) sensitive to these divalent cations. The antagonizing effects of ZnCl 2 and MnCl 2 were observed at concentrations lower than those of NiCl 2 , indicating that the affinity of Zn 2+ and Mn 2+ was much higher than that of Ni 2+ to the transporter. The putative transporters were DMT1 and ZIPs. Although DMT1 has an affinity to Ni 2+16 , it was likely to contribute minimally to Ni 2+ uptake in THP-1 cells, because the DMT1 inhibitor, Chlorazol Black 22,23 , did not decrease Ni 2+ uptake. The ZIP family consists of several members and some of them have an affinity to Ni 2+24-26 . All ZIPs except for ZIP12 were expressed in THP-1 cells 27 , and ZIP2 25,28 , ZIP3 26 , ZIP8, and ZIP14 24,29,30 have been shown to have an affinity to Zn 2+ , Mn 2+ , Co 2+ . In addition, ZIPs are known to be induced by the stimulation of TLR4 31 . However, the possibility that Ni 2+ induced Zn transporters via the stimulation of TLR4 was rejected, because TAK-242 did not affect the increase in Ni content in the cells incubated for 24 h. These findings suggested that the Ni 2+ entered via constitutively expressed ZIP-type transporters. However, because several family members might be involved in Ni 2+ uptake and because they have no specific inhibitors, it was difficult to identify the one responsible in this case. We started screening the specific inhibitors of Ni 2+ influx to identify the transporter.
We, for the first time, also showed that antagonizing Ni 2+ uptake by Zn 2+ resulted in the inhibition of IL-8 production. Zn 2+ also inhibited Co 2+ uptake and Co 2+ -induced IL-8 production whereas Zn 2+ did not inhibit LPS-induced IL-8 production, indicating that Zn 2+ did not affect the signaling pathway for IL-8 expression. In contrast, although Mn 2+ inhibited Ni 2+ uptake, Mn 2+ itself induced IL-8 production. These findings were consistent with the observation that Mn 2+ as well as Ni 2+ could activate HIF-1α 12 . These findings also suggested that Zn 2+ has the ability to attenuate Ni 2+ and Co 2+ -induced inflammation.
The protective effects of Zn 2+ at physiological concentrations were also observed in an in vivo model. We had reported that Ni 2+ elution from the Ni wire induced inflammatory events, such as neutrophil infiltration and prostaglandin and histamine production 5,6 , and that the initial inflammatory responses induced further elution of Ni 2+5 . Using the Ni wire-implanted mouse model, we showed that Ni 2+ -induced inflammation was enhanced in a Zn-deficient state. Additionally, the mice fed with Zn-deficient diet for 2 weeks showed an enhanced Ni wire-induced expression of MIP-2, a neutrophil chemokine, and COX-2. The elution of Ni 2+ was also enhanced, probably via augmentation of the inflammation, as consistent with the previous study. The severe Zn deficiency causes various defects in the function of the skin, such as barrier function. However, in our condition, although Zn 2+ concentration in the serum was apparently decreased, that in the skin was unchanged, indicating that functions of the skin were not impaired. Even though the Ni 2+ elution and Ni 2+ -induced cytokine expression were enhanced, this suggested that the concentration of Zn 2+ in the serum and/or in the intercellular fluids affected the Ni 2+ uptake of leukocytes infiltrated from the blood stream. These results suggested that Ni 2+ -induced  The present in vitro and in vivo findings suggested that Zn 2+ modulated Ni 2+ uptake and the activation of the inflammatory cells. Our findings also suggested the need to issue a warning that a Zn-deficient state may exacerbate medical device-induced inflammation. A recent report indicated that the prevalence of Zn deficiency in Japanese adult males and females increased with increasing age, and that infants were also susceptible to Zn deficiency 32 . Therefore, it is important to ascertain whether people with Zn-deficiency are susceptible to Ni allergy, and to determine Zn 2+ levels to avoid the induction of Ni-induced inflammation in people implanted with medical devices.
Mice. Four-week-old male ICR mice were purchased from SLC (Shizuoka, Japan). They were fed a standard diet (CE-2, CLEA, Tokyo, Japan) (control diet group, n = 12) or a Zn-deficient diet (CLEA, Tokyo, Japan) (low-Zn diet group, n = 12) for two weeks under a 12-h light/dark cycle in a specific, pathogen-free barrier facility. All animal experiments were approved by the Institutional Animal Care and Use Committee of Tohoku University, and performed in accordance with the Regulations for Animal Experiments and Related Activities at Tohoku University and Guidelines for Proper Conduct of Animal Experiments by the Ministry of Education, Culture, Sports, Sciences, and Technology of Japan. Implantation of the Ni wire. The Ni wire was cut into 5-mm length, sterilized by ultraviolet irradiation, and then washed with ethanol. Mice were anesthetized using isoflurane (Wako, Osaka, Japan) and then sterilized Ni wires were implanted subcutaneously in their dorsa using a 13 G implant needle (Natsume, Tokyo, Japan). In the control group, mice underwent a similar surgical procedure, but without the implantation of the Ni wire.

ELISA.
After incubation of each of the sample, IL-8 in the supernatants was assayed using an ELISA kit (eBioscience, San Diego, CA) according to the manufacturer's protocol.
Determination of Ni 2+ , Zn 2+ , Mn 2+ , Co 2+ concentrations with ICP-MS. THP-1 cells were stimulated by NiCl 2 for 24 h in Fig. 1a and b, or for the indicated time in Fig. 1c and d. The cells were stimulated by NiCl 2 and/or other metal chlorides for 24 h in Figs 2, 4 and 5. After the incubation, they were collected and washed five times with PBS (phosphate-buffered saline), and then suspended in 150 μl PBS. The cell suspension was sonicated for 30 s and the aliquot was diluted 10-fold with 5% (w/w) HNO 3 . The concentration of Ni 2+ and other metal ions in each sample was determined by Agilent 7500 Series ICP-MS (Agilent Technology, Santa Clara, CA).
To determine the metal concentrations in the mouse skin and serum, circular skin tissue sections (1 cm in diameter) from the region surrounding the Ni wire were excised and the wet weight of skin was measured. The skin tissue sample, approximately 80 mg, was boiled in 3 ml 69% (w/w) HNO 3 for 30 min, and then, 300 μl 30% (w/v) H 2 O 2 was added to the samples, on ice. The skin samples were then boiled again for approximately 30 min, and pure water was added to attain a total weight of 10 g. Mouse blood was incubated for 12 h at 4 °C and then centrifuged at 1,200 × g, 4 °C for 30 min. The supernatant was diluted 10-fold with 5% (w/w) HNO 3 , and centrifuged at 500 × g, 4 °C for 5 min. The supernatant was collected. The Ni 2+ , Zn 2+ concentration of each sample was also determined by ICP-MS. Newport green fluorescence staining of intracellular Ni ions. THP-1 cells were stimulated by NiCl 2 and/or ZnCl 2 for 24 h. After the incubation, the cells were collected and washed five times with 1 × PBS, and then treated for 30 min with 5 μM Newport Green TM DCF diacetate (Invitrogen, Carlsbad, CA) dissolved in dimethyl sulfoxide. After this treatment, the cells were washed once with 1 × PBS and placed on a Micro Slide Glass (76 × 26 mm, 0.9-1.2 mm thickness, Matsunami-glass, Osaka, Japan), cover-slipped with Fluoromount (DBS, Diagnostic BioSystems, CA). Fluorescence images (excitation at 505 nm and emission at 535 nm) were acquired using a laser scanning confocal microscope LSM 800 (Carl Zeiss, Germany).

Statistical analysis.
The statistical significance of the results was analyzed using the unpaired two-tailed Student's t-test, and the Bonferroni multiple comparison test or Student-Newman-Keuls test for multiple comparisons. For some experiments, a statistical outlier removal was performed using the Smirnov-Grubbs' rejection test and the Thompson test.