Functional studies of Drosophila zinc transporters reveal the mechanism for dietary zinc absorption and regulation

Abstract

Background: Zinc is key to the function of many proteins, but the process of dietary zinc absorption is not well clarified. Current knowledge about dietary zinc absorption is fragmented, and mostly derives from incomplete mammalian studies. To gain a comprehensive picture of this process, we systematically characterized all zinc transporters (that is, the Zip and ZnT family members) for their possible roles in dietary zinc absorption in a genetically amenable model organism, Drosophila melanogaster.

Results: A set of plasma membrane-resident zinc transporters was identified to be responsible for absorbing zinc from the lumen into the enterocyte and the subsequent exit of zinc to the circulation. dZip1 and dZip2, two functionally overlapping zinc importers, are responsible for absorbing zinc from the lumen into the enterocyte. Exit of zinc to the circulation is mediated through another two functionally overlapping zinc exporters, dZnT1, and its homolog CG5130 (dZnT77C). Somewhat surprisingly, it appears that the array of intracellular ZnT proteins, including the Golgi-resident dZnT7, is not directly involved in dietary zinc absorption. By modulating zinc status in different parts of the body, we found that regulation of dietary zinc absorption, in contrast to that of iron, is unresponsive to bodily needs or zinc status outside the gut. The zinc transporters that are involved in dietary zinc absorption, including the importers dZip1 and dZip2, and the exporter dZnT1, are respectively regulated at the RNA and protein levels by zinc in the enterocyte.

Conclusions: Our study using the model organism Drosophila thus starts to reveal a comprehensive sketch of dietary zinc absorption and its regulatory control, a process that is still incompletely understood in mammalian organisms. The knowledge gained will act as a reference for future mammalian studies, and also enable an appreciation of this important process from an evolutionary perspective.

Figure 1

Drosophila Zip1 (dZip1) and dZip2 are zinc-specific transporters involved in dietary zinc absorption. (A) Gut-specific knockdown of dZip2 or of dZip1 dZip2 in combination resulted in impaired larval development under zinc deficiency (0.3 mmol/l EDTA). (B) Reverse transcriptase (RT)-PCR analysis of the knockdown effect of the dZip1-RNAi and dZip2-RNAi lines. rp49 was used as the loading control. (C) Ubiquitous knockdown of either dZip1 or dZip2 produced a reduction in alkaline phosphatase (ALP) activity but did not affect aconitase activity. (D) Gut-specific knockdown of either dZip1 or dZip2 similarly led to reduced ALP activity of whole body minus gut in flies fed on 0.1 mmol/l EDTA-supplemented food. (E)dZip2-RNAi and dZip1-, dZip2-RNAi flies are specifically sensitive to zinc deficiency. The survival rate of these RNAi flies on EDTA-supplemented food could be rescued by adding back zinc but not other metals. (A,D) Genotypes of flies used were NP3084/+, NP3084/+; dZip1-RNAi/+ (dZip1-RNAi fly), NP3084/+; dZip2-RNAi/+ (dZip2-RNAi fly), NP3084/+; dZip1-, dZip2-RNAi/+ (dZip1-, dZip2-RNAi fly). (B,C,E) Genotypes of flies used were da-GAL4/+ for control and da-GAL4/dZip1-RNAi for dZip1-RNAi, da-GAL4/dZip2-RNAi for dZip2-RNAi, and da-GAL4/dZip- dZip2-RNAi for the double RNAi fly. Values are presented as means ± SEM; n ≥ 3. *P < 0.05, **P < 0.01, ***P < 0.001; one-way ANOVA.

Figure 2

Drosophila Zip1 (dZip1) and dZip2 are localized to the plasma membrane. (A-C) Specific localization of dZip1 on the apical side in the larval midgut constriction. dZip1 immunofluorescence staining is red, and nuclear DAPI staining is blue. (A) Specific distribution of dZip1 immunofluorescence (red) in the larval midgut constriction. Arrows denote the midgut constriction. Scale bars = 50 μm. (B) dZip1 (red) is located on the apical membrane of cells in the midgut constriction. Arrowheads and arrows indicate the basal and the apical membranes, respectively. Scale bars = 10 μm. (C) A surface view of the midgut constriction clearly shows the plasma membrane localization of dZip1 (red). Scale bars = 10 μm. (D) Membrane localization of dZip2 expressed in Caco-2 cells. To facilitate visualization, enhanced green fluorescent protein (eGFP)-tagged dZip2 (green) was used. Scale bars = 10 μm. (E) Immunofluorescence (red) against hemagglutinin (HA) reveals apical distribution of dZip2-HA in the midgut cells. DAPI (blue) shows the nucleus. Arrowheads and arrows indicate the basal and the apical membranes, respectively. Scale bars = 10 μm. Flies used are NP3084/UAS-dZip2-HA.

Figure 3

dZip1 or dZip2 overexpression causes zinc accumulation. (A) Fluorescence of MtnB-eYFP was enhanced in the midgut constriction of the larvae overexpressing dZip1. Genotypes of the flies are MtnB-EYFP/+; da-GAL4/+ for the control larvae, and MtnB-EYFP/+; da-GAL4/UAS-dZip1 for the larvae overexpressing dZip1. Arrowheads denote the midgut constriction. Scale bars = 100 μm. (B) Reverse transcriptase (RT)-PCR analysis showing upregulation of MtnB and MtnC in larvae overexpressing dZip1. rp49 was used as the reference gene for normalization. (C) RT-PCR analysis of MT2a transcriptional induction by overexpression of dZip2 in Caco-2 cells. GAPDH was used as the reference gene for normalization. (D) Zinpyr-1 staining reveals intracellular zinc accumulation after dZip1 or dZip2 expression in Chinese hamster ovary (CHO) cells. Scale bars = 10 μm. (E) Ubiquitous overexpression of dZip1 specifically resulted in zinc sensitivity. (F) Gut-specific overexpression of dZip2 led to zinc sensitivity. Genotypes of flies are da-GAL4/+ for the control, da-GAL4/UAS-dZip1 for dZip1 overexpression; NP3084/+ for the gut-specific control fly) and NP3084/UAS-dZip2 for the gut-specific dZip2 overexpression fly. Values are presented as means ± SEM; n ≥ 3. *P < 0.05, **P < 0.01, ***P < 0.001; one-way ANOVA.

Figure 4

CG6672 (dZnT7), the Drosophila ortholog of hZnT7, is not directly involved in dietary zinc absorption. (A)hZnT7, but not dZnT1, partially rescues dZnT7 after RNA interference (RNAi). Ubiquitous knockdown of dZnT7 caused lethality to the wandering larvae, but with hZnT7 these flies reached the late pupal or even adult stage. (B) RT-PCR analysis of the knockdown effect of dZnT7-RNAi lines at third-instar larval stage. rp49 was used as the loading control. (C) Golgi localization of dZnT7 in Caco-2 cells. Fluorescence markers; enhanced green fluorescent protein (eGFP)-tagged dZnT7 (green); red fluorescent protein (RFP)-tagged Golgi-targeted peptide (red); merged image (yellow) of dZnT7 and Golgi markers. Scale bars = 10 μm. (D) The activity of alkaline phosphatase (ALP), but not aconitase, was significantly reduced in larvae with dZnT7-RNAi driven by da-GAL4. (E) Gut-specific knockdown of dZnT7 could not reduce the ALP activity of the whole body, although the ubiquitous silencing causes significant reduction. (F) ALP activity of whole body minus gut in gut-specific RNAi of intracellular ZnTs on zinc-limited (0.3 mmol/l EDTA) food. Neither dZnT7, nor any other intracellular ZnT transporters, had any discernible effect on zinc availability for the rest of the body. NP3084 was crossed with individual RNAi lines. Values are presented as means ± SEM; n ≥ 3. *P < 0.05, **P < 0.01, ***P < 0.001; one-way ANOVA.

Figure 5

Drosophila ZnT1 (dZnT1) and its close homolog CG5130 export zinc from the gut for systemic use. (A) Reverse transcriptase (RT)-PCR analysis of the knockdown effect of CG5130-RNAi. rp49 was used as the loading control. (B) Membrane localization of CG5130 in Caco-2 cells. Enhanced green fluorescent protein (eGFP)-tagged CG5130 is shown in green. Scale bars = 10 μm. (C) Reduced alkaline phosphatase (ALP) activity of whole body minus gut in gut-specific CG5130-RNAi larvae raised on zinc-limited food (0.1 mmol/l EDTA), with little change in aconitase activity. Genotypes of the flies are NP3084/+ (gut-specific control fly) and NP3084/CG5130-RNAi (gut-specific CG5130-RNAi fly). Values are presented as means ± SEM; n ≥ 3. *P < 0.05, **P < 0.01, ***P < 0.001; one-way ANOVA. (D) Zinpyr-1 staining showing intracellular zinc reduction caused by expressing CG5130 in Chinese hamster ovary (CHO) cells. Scale bars = 10 μm. (E) Gut-specific knockdown of CG5130 plus dZnT1 further exacerbated the phenotype on EDTA-supplemented food.

Figure 6

The plasma membrane-resident Drosophila zinc transporters, dZips and dZnTs, are regulated by zinc availability. (A-B) Reverse transcriptase (RT)-PCR analysis of (A) dZips and (B) dZnTs in the gut showing transcriptional responsiveness of zinc transporters to dietary zinc. (C) Immunostaining reveals that dZip1 is upregulated by dietary zinc limitation at the midgut constriction (arrows ). Scale bars = 50 μm. (D,E) Western blotting analysis shows that (D) dZnT1 is downregulated by dietary zinc overload, but not by zinc limitation; and (E) using anti-HA antibody, shows that CG5130-HA is not responsive to dietary zinc level. NF, normal food; ZN, 2 mmol/l ZnCl₂; EDTA, 0.3 mmol/l EDTA.

Figure 7

Dietary zinc absorption is influenced by zinc levels in the enterocyte but not the rest of the body. (A) Transcriptional responses of dZip1 and dZip2 to gut-specific zinc changes as revealed by reverse transcriptase (RT)-PCR analysis. (B) Transcription of dZip1 and dZip2 in the gut does not respond noticeably to body zinc changes achieved by modulating the zinc excretion process (ZnT35C knockdown or ZnT35C overexpression in the malpighian tubules). (C) As a control, ZnT35C knockdown in the malpighian tubules effectively induced the bodily transcription of MtnB and MtnC (reflecting zinc increase), whereas overexpression of ZnT35C repressed the transcription of MtnB and MtnC. (D) Western blotting analysis shows that dZnT1 expression in the enterocyte is not altered when zinc excretion is modulated. (E) Immunofluorescence staining shows that dZip1 is induced at the midgut constriction (arrows) when dZnT1 is specifically overexpressed in the gut, in the absence of dietary zinc change. Scale bars = 50 μm. Genotypes of flies are NP3084/+ (control fly) and NP3084/dZnT1-RNAi (dZnT1-RNAi fly), NP3084/+; UAS-dZnT1/+ (dZnT1 overexpression fly), NP1093/+ (control fly), NP1093/+; ZnT35C-RNAi/+ (ZnT35C RNAi fly), and NP1093/+; UAS-ZnT35C (ZnT35C overexpression fly). RNAi, RNA interference.

Figure 8

A model for dietary zinc absorption in Drosophila enterocytes. Dietary zinc absorption takes a direct transversing path through the midgut cells, bypassing the intracellular organelles. Zinc uptake from the lumen involves the Drosophila transporters dZip1 and dZip2 on the apical membrane, and these transporters are negatively affected at the RNA level by zinc in the enterocytes. Cytosolic zinc is pumped out into circulation by dZnT1 and CG5130 on the basolateral membrane. dZnT1 is post-transcriptionally repressed by zinc overload, whereas CG5130 is unresponsive to dietary zinc changes. Zinc in the cytosol is bound by metallothioneins to prevent its toxicity when overloaded. Zinc transporters along the secretion pathway are not directly involved in zinc efflux to the circulation. The regulated expression the zinc transporters mediating dietary zinc uptake depends autonomously on the zinc level in the enterocyte per se, and is unresponsive to the zinc status in the rest of the body. The negative regulation of zinc transporters by dietary zinc occurs through altering the zinc levels in the enterocyte.