ZnT4 provides zinc to zinc-dependent proteins in the trans-Golgi network critical for cell function and Zn export in mammary epithelial cells

Abstract

Zinc (Zn) transporter 4 (ZnT4) plays a key role in mammary gland Zn metabolism. A mutation in ZnT4 (SLC30A4) that targets the protein for degradation is responsible for the "lethal milk" (lm/lm) mouse phenotype. ZnT4 protein is only detected in the secreting mammary gland, and lm/lm mice have ∼35% less Zn in milk, decreased mammary gland size, and decreased milk secretion. However, the precise contribution of ZnT4 is unknown. We used cultured mouse mammary epithelial cells (HC11) and determined that ZnT4 was localized to the trans-Golgi network (TGN) and cell membrane and transported Zn from the cytoplasm. ZnT4-mediated Zn import into the TGN directly contributed to labile Zn accumulation as ZnT4 overexpression increased FluoZin3 fluorescence. Moreover, ZnT4 provided Zn for metallation of galactosyltransferase, a Zn-dependent protein localized within the TGN that is critical for milk secretion, and carbonic anhydrase VI, a Zn-dependent protein secreted from the TGN into milk. We further noted that ZnT4 relocalized to the cell membrane in response to Zn. Together these studies demonstrated that ZnT4 transports Zn into the TGN, which is critical for key secretory functions of the mammary cell.

Fig. 1.

Zinc (Zn) transporter 4 (ZnT4) is localized to the trans-Golgi apparatus. A: confocal micrographs of HC11 cells detected with affinity purified ZnT4 antibody (1 μg/ml) and visualized with Alexa Fluor 488-conjugated anti-rabbit IgG. Confocal micrographs illustrate that spatial distribution of endogenous ZnT4 in mammary epithelial cells is perinuclear (thick arrows) and displays distinct cell membrane localization (thin arrows). DIC, differential interference constant. B: representative immunoblot (IB) of total membrane proteins (50 μg protein/lane) from HC11 cells transfected with pcDNA3.1 [lanes 1 and 2, mock-transfected (Mock) control] or cells expressing the ZnT4-hemagglutinin (HA) fusion protein (lanes 3 and 4) detected with HA (∼47 kDa) and anti-β-actin (∼38 kDa) antibodies. C: confocal micrographs display colocalization of ZnT4 (ZnT4-HA; green) with p58 (red), a trans-Golgi network (TGN) marker. Colocalization (yellow) illustrates localization in the TGN. Scale bar represents 20 μm.

Fig. 2.

ZnT4 transports Zn out of the cytoplasm and into a labile Zn pool in mammary epithelial cells. A: HC11 cells were transfected with pGL3 empty vector (EV) and thymidine kinase promoter-linked Renilla luciferase vector (pRL-TK), 4×-metal responsive element-pGL3 luciferase reporter and pRL-TK renilla (Mock), or 4×-metal responsive element-pGL3 luciferase reporter, pRL-TK renilla, and pcDNA3.1ZnT4-HA (ZnT4), and changes in luminescence were assessed after 24 h Zn (1 μM) treatment. Data represent mean ratios of firefly:renilla luciferase light units ± SD (n = 3 samples/group); *P < 0.05. B: labile Zn accumulation into intracellular vesicles was quantified by fluorometric assay using FluoZin-3 in Mock-transfected cells and cells expressing ZnT4-HA fusion protein. Data represent mean FluoZin-3 fluorescence/microgram protein ± SD (n = 8 samples/group); *P < 0.05. Experiment was repeated more than 3 times.

Fig. 3.

ZnT4 increases trans-Golgi apparatus Zn concentration in mammary epithelial cells. HC11 cells were mock-transfected (Mock) or transfected with ZnT4-HA (ZnT4). A: UDP-galactosyltransferase activity was measured as a biochemical index of Zn metallation of resident Zn-dependent proteins in the TGN. Data represent mean galactosyltransferase activity ± SD (n = 3 samples/group); *P < 0.05. Experiment was repeated more than 3 times. B: representative IB of UDP-galactosyltransferase (∼44 kDa) in total membrane proteins isolated from mock-transfected and ZnT4-HA-transfected cells. Membranes were stripped and reprobed for β-actin to normalize for sample loading.

Fig. 4.

ZnT4 attenuation decreased carbonic anhydrase (CA)-VI protein abundance. A: representative IB of ZnT4 (∼47 kDa) in total membrane protein fractions isolated from mock-transfected (Mock) cells and cells transfected with ZnT4 small interfering (si)RNA (siZnT4). Membranes were stripped and reprobed for β-actin to normalize for sample loading. B: representative IB of CA VI (∼33 kDa) in total membrane proteins isolated from mock-transfected (Mock) cells and cells transfected with siZnT4. Membranes were stripped and reprobed for β-actin to normalize for sample loading. C: data represent mean ratios of CA VI:β-actin ± SD (n = 3 samples/group); *P < 0.05. D: data represent CA-VI mRNA expression in mock-transfected (Mock) and cells transfected with siZnT4 ± SD (n = 3 samples/group).

Fig. 5.

Zinc exposure relocalizes ZnT4 in mammary epithelial cells. A: confocal micrograph of HC11 cells that were treated with 0 or 200 μM Zn for 2 h. ZnT4-HA was detected with Alexa Fluor 488-conjugated anti-rabbit IgG (green), and cell nuclei were stained with TO-PRO-3 (blue). B: cells were treated with Zn (200 μM) for 2 h and biotinylated, and cell surface proteins were isolated following precipitation with avidinated agarose beads. Proteins were separated by electrophoresis and immunoblotted with anti-HA antibody. Representative IB of cell membrane proteins detected with anti-HA antibody following cell surface biotinylation of HC11 cells treated with 0 μM Zn (0) or 200 μM Zn (200). Nonspecific binding of proteins to avidinated beads was assessed in cells not exposed to biotin (NB). Total membrane isolated proteins (TM) served as a positive control. Scale bar represents 20 μm.