Zinc transporter ZIP14 functions in hepatic zinc, iron and glucose homeostasis during the innate immune response (endotoxemia)

Abstract

ZIP14 (slc39A14) is a zinc transporter induced in response to pro-inflammatory stimuli. ZIP14 induction accompanies the reduction in serum zinc (hypozincemia) of acute inflammation. ZIP14 can transport Zn(2+) and non-transferrin-bound Fe(2+) in vitro. Using a Zip14(-/-) mouse model we demonstrated that ZIP14 was essential for control of phosphatase PTP1B activity and phosphorylation of c-Met during liver regeneration. In the current studies, a global screening of ZIP transporter gene expression in response to LPS-induced endotoxemia was conducted. Following LPS, Zip14 was the most highly up-regulated Zip transcript in liver, but also in white adipose tissue and muscle. Using ZIP14(-/-) mice we show that ZIP14 contributes to zinc absorption from the gastrointestinal tract directly or indirectly as zinc absorption was decreased in the KOs. In contrast, Zip14(-/-) mice absorbed more iron. The Zip14 KO mice did not exhibit hypozincemia following LPS, but do have hypoferremia. Livers of Zip14-/- mice had increased transcript abundance for hepcidin, divalent metal transporter-1, ferritin and transferrin receptor-1 and greater accumulation of iron. The Zip14(-/-) phenotype included greater body fat, hypoglycemia and higher insulin levels, as well as increased liver glucose and greater phosphorylation of the insulin receptor and increased GLUT2, SREBP-1c and FASN expression. The Zip14 KO mice exhibited decreased circulating IL-6 with increased hepatic SOCS-3 following LPS, suggesting SOCS-3 inhibited insulin signaling which produced the hypoglycemia in this genotype. The results are consistent with ZIP14 ablation yielding abnormal labile zinc pools which lead to increased SOCS-3 production through G-coupled receptor activation and increased cAMP production as well as signaled by increased pSTAT3 via the IL-6 receptor, which inhibits IRS 1/2 phosphorylation. Our data show the role of ZIP14 in the hepatocyte is multi-functional since zinc and iron trafficking are altered in the Zip14(-/-) mice and their phenotype shows defects in glucose homeostasis.

Figure 1. LPS differentially regulates ZIP14 expression in mice.

Young adult mice received LPS (2 mg/kg, i.p.) or the same volume (0.5 mL) of saline (control), 1–18 hr before being killed. (A) Total RNA was isolated and Zip14 mRNA was measured by qPCR and expressed relative to TBP mRNA as the normalizer. ZIP14 protein abundance was measured by western analysis of liver homogenates. Representative western blots from multiple mice (n = 3−4) were measured for ZIP14 abundance by densitometry. (B, C) Zinc concentrations in serum and liver, in µg/mL and µg/g respectively, were measured by AAS. (D, E) Comparison of Zip14 mRNA and ZIP14 protein in WT and Zip14 KO mice 18 hr after LPS, as measured by qPCR and western analysis. Values are mean ± SD, n = 3−5. (E) ZIP14 protein is increased at the plasma membrane of hepatocytes of WT mice but not Zip14^−/− mice following LPS. Localization was by confocal microscopy using ZIP14 antibody and Alexa fluor594 secondary antibody and DAPI as the nuclear marker. (F) Serum IL-6 as measured by ELISA was used as an indicator of efficiency of LPS administration. The IL6 response from LPS was attenuated in the Zip14^−/− mice. (* = P<0.05, ** = P<0.01, *** = P<0.001, **** = P<0.0001).

Figure 2. Zip14 null mice do not have depressed serum zinc after LPS administration.

WT and Zip14^−/− mice were given LPS or saline administration (2 mg/kg; 0.5 mL; i.p.) 18 hr before being killed. (A, B) Zn concentrations in serum and liver were measured by AAS. (C, D) Global analysis of liver Zip and ZnT transcripts in KO and after LPS. Total RNA was isolated from the liver and Zip14 mRNA was measured by qPCR and normalized to TBP mRNA. Values are mean ± SD, n = 3−5. (E) ZIP4 protein abundance in liver was shown by western analysis.

Figure 3. Zinc absorption and hepatic uptake is reduced in Zip14^−/− mice.

Fasted mice received 2 µCi of ⁶⁵Zn by gavage and were killed 3 hr later. (A) Percent absorption was calculated from the radioactivity administered. (B, C) Serum and liver uptake was calculated from the specific activity of the ⁶⁵Zn. Values are mean ± SE, n = 5−12.

Figure 4. Zip14^−/− mice exhibit normal iron absorption but altered iron homeostasis.

(A) WT and Zip14^−/− mice were administered LPS (2 mg/kg or saline 0.5 mL; ip), 18 hr before being killed. (A) Serum and liver iron concentrations were measured by AAS. Liver non-heme-iron was measured colorimetrically. (B) Fasted WT and Zip14^−/− mice received 2 µCi of ⁵⁹Fe by gavage and were killed 24 hr later. Percent absorption was calculated from the radioactivity administered. Serum and liver iron uptake was calculated from the specific activity of the ⁵⁹Fe. (C) Transcript abundance for liver hepcidin, TfR-1, DMT1 and ferritin was measured by qPCR and expressed relative to TBP mRNA as the normalizer. Values are mean ± SE, n = 5−10.

Figure 5. Deletion of Zip14 in mice produces altered glucose homeostasis and IR functions.

(A) Body composition of the WT and Zip14^−/− female mice was measured using a NMR Lean/Fat analyzer. (B) Serum and liver glucose from fed-mice were measured by OneTouch UltraMini and colorimetrically, respectively. (C, D) Serum insulin and liver cAMP were measured by ELISA. (E, F) Western analysis results from liver of three mice are shown for each treatment group. (G) Total RNA was isolated from livers and relative transcript abundance for GLUT2, PEPCK, SREBP-1c, FASN and SOCS-3 were measured by qPCR and expressed relative to TBP mRNA as the normalizer. Values are mean ± SE, n = 3−5.

Figure 6. Zip14^−/− mice exhibit altered zinc transporter expression and insulin and glucagon abundance in pancreas.

(A) Total RNA isolated from pancreas and Zip14, Zip10 and ZnT8 mRNAs were measured by qPCR and TBP was used as a normalizer. Immunofluoresence microscopic analysis showing ZIP14 abundance (B) and ZnT8, insulin colocalization (C) in pancreas of the WT and Zip14^−/− mice after LPS. (D) Immunohistochemical analysis of insulin and glucagon abundance in pancreatic islets of WT and Zip14^−/− mice that were administered LPS (2 mg/kg or saline; 0. 5 mL; i.p.) 18 hr before being killed. Values are mean ± SD, n = 5.

Figure 7. Zip14 expression is greatest in WAT and muscle following LPS and Zip14 deletion produces atypical metal homeostasis.

WT and Zip14^−/− mice were administered LPS (2 mg/kg or saline; 0.5 mL; i.p.) 18 hr before being killed. (A) Total RNA was isolated from 13 tissues of the mice. Zip14 mRNA abundance was measured by qPCR and expressed relative to TBP mRNA as the normalizer. A western analysis shows ZIP14 expression in WAT. (B) Multiple tissues were removed from the mice and assayed for zinc and iron concentrations by AAS. Only tissues where there were significant differences in concentration are shown. (C, D) Global screening of ZnT and Zip mRNAs expressed in WAT and muscle was conducted by qPCR as described in (A). Only those ZnT and Zip transcripts that were significantly different in WT vs. Zip14^−/− mice or following LPS treatment for WAT (C) and muscle (D) are shown. Values are mean ± SD, n = 3−5.

Figure 8. Model showing the influence of Zip14 deletion in mice on hepatic zinc and iron metabolism and signaling pathways for glucose homeostasis during endotoxemia.

The influence of the Zip14 null deletion on reduced liver zinc and increased iron uptake is shown. Up-regulation of Zip14 by LPS via toll like receptor 4 (TLR4) in liver and activation of the NFKB pathway leading to increased IL6, for an autocrine response, and from macrophage-produced IL-6. The suppression of IR activation by IL-6 and cAMP-induced SOCSs is also shown. The apparent reduction in glucagon production in pancreatic α cells and the effect on cellular cAMP in the Zip14 KO mice are proposed. We hypothesize that these signaling events produce hypoglycemia by a reduction in gluconeogenesis.