Biochemical Journal

Research article

Transcriptional regulation of the Zrg17 zinc transporter of the yeast secretory pathway

Yi-Hsuan Wu , Avery G. Frey , David J. Eide


The Msc2 and Zrg17 proteins of Saccharomyces cerevisiae are members of the cation diffusion facilitator family of zinc transporters. These proteins form heteromeric complexes that transport zinc into the ER (endoplasmic reticulum). Previous studies suggested that the ZRG17 gene is regulated in response to zinc status by the Zap1 transcription factor. Zap1 activates the expression of many genes in zinc-deficient cells. In the present study, we assessed whether ZRG17 is a direct Zap1 target gene. We showed that ZRG17 mRNA levels were elevated in zinc-limited cells in a Zap1-dependent manner and were also elevated in zinc-replete cells expressing a constitutively active allele of Zap1. Furthermore, Zrg17 protein levels correlated closely with mRNA levels. A candidate Zap1-binding site [ZRE (zinc-responsive element)] in the ZRG17 promoter was required for this induction. Using electrophoretic mobility-shift assays and chromatin immunoprecipitation, we demonstrated that Zap1 binds specifically to the ZRG17 ZRE both in vitro and in vivo. By using a chromosomal ZRG17 mutant with a non-functional ZRE, we found that Zap1 induction of ZRG17 is required for ER function as indicated by elevated ER stress under zinc-limited conditions. Together, these results establish that ZRG17 is a direct Zap1 target gene and its regulation has biological importance in maintaining ER function.

  • endoplasmic reticulum (ER)
  • homoeostasis
  • Saccharomyces cerevisiae
  • transcription
  • zinc
  • Zrg17


Zinc serves as a catalytic or structural cofactor for a large number of proteins. Many of these proteins are secreted from cells and acquire their zinc as they pass through the secretory pathway. These include enzymes such as matrix metalloproteases, alkaline phosphatases and angiotensin-converting enzymes. In addition, many other zinc-dependent proteins are resident in secretory pathway compartments such as the ER (endoplasmic reticulum) and Golgi apparatus. These include protein chaperones and co-chaperones (e.g. calreticulin, calnexin or DnaJ orthologues), ER-associated peptidases [e.g. ERAAP (ER aminopeptidase associated with antigen processing)] and GPI (glycosylphosphatidylinositol) PETs (phosphoethanolamine transferases) involved in GPI anchor synthesis. Evidence that the early secretory pathway requires zinc to function comes from studies showing that zinc deficiency causes induction of the UPR (unfolded protein response), an indicator of ER stress [1,2]. The ER has also been proposed to be the source of zinc that acts as an intracellular second messenger of IgE receptor activation in mast cells [3]. Conversely, excess zinc may also be disruptive to ER function. ER zinc overload was proposed to cause EDS-SCD (spondylocheiro dysplastic form of Ehlers–Danlos syndrome) resulting from mutations in the SLC39A13 (solute carrier 39 A13)/ZIP13 (Zrt- and Irt-like protein 13) zinc transporter gene [4,5].

Given these various roles, it is clear that cells must have efficient systems for the transport of zinc into the ER and Golgi under zinc deficiency and regulatory mechanisms to maintain zinc homoeostasis within those compartments. Several zinc transporters in the early secretory pathway have been identified. In vertebrates, the ZnT (zinc transporter)-5, ZnT-6 and ZnT-7 proteins have been found to contribute to secretory pathway zinc and the metallation of secreted proteins [68]. These three proteins are members of the CDF (cation diffusion facilitator)/ZnT/SLC30A family of zinc transporters. Whereas ZnT-7 is active as a homodimer, ZnT-5 and ZnT-6 form a heterodimer complex to be active [9]. In Saccharomyces cerevisiae, the Msc2 and Zrg17 proteins play key roles in maintaining secretory pathway zinc. These proteins are the yeast orthologues of vertebrate ZnT-5 and ZnT-6 and reside in the ER [1,2]. Previous results suggest that, like ZnT-5 and ZnT-6, Msc2 and Zrg17 are only active as heterodimeric complexes [2,9]. Specifically, it was shown that Msc2 and Zrg17 interact physically and that both proteins are required for zinc transport function. Moreover, it was shown that Msc2–Zrg17 activity is required only under zinc-limiting conditions and that other transport systems are sufficient to maintain ER zinc levels in zinc-replete cells [1,2].

Little is known about how secretory pathway zinc transporters are regulated in response to zinc status or other signals. In the present study, we have addressed the transcriptional regulation of ZRG17 in response to zinc by the Zap1 transcriptional activator. In yeast, Zap1 is the central player in the response of cells to zinc deficiency [10]. We currently estimate that Zap1 activates the transcription of ~80 genes in zinc-limited cells and the expression of several other genes are repressed directly or indirectly by Zap1 [1115]. Zap1 regulates target gene expression by binding to 11 bp sequences known as ZREs (zinc-responsive elements) in those target gene promoters [16]. The consensus sequence for the ZRE is 5′-ACCTTNAAGGT-3′ and flanking sequences may also contribute to Zap1 ZRE recognition [17]. Regulation of Zap1 activity is controlled by zinc binding directly to the protein to repress activation domain function in zinc-replete cells [18,19].

Previous analyses of the Zap1 regulon suggested that ZRG17 is a direct target of Zap1 activation [11,12,20]. Using whole-genome DNA microarrays, we found that ZRG17 expression increased in zinc-limited cells in a Zap1-dependent manner. In the present study, we tested this hypothesis and confirmed that ZRG17 is indeed a direct target of Zap1 regulation. In addition, we have also demonstrated the physiological importance of this regulation to ER function.


Yeast strains and growth conditions

Media used were YPD [1% (w/v) yeast extract/2% (w/v) peptone/2% (w/v) glucose], SD (synthetic dextrose: 0.67% yeast nitrogen base without amino acids), YPGE [1% (w/v) yeast extract/2% (w/v) peptone/2% (w/v) glycerol/2% (w/v) ethanol] and LZM (low-zinc medium) as described previously [21]. LZM contains 1 mM EDTA and 20 mM citrate to both limit and buffer available zinc levels. Yeast strains DY1457 (MATa ade6 can1 his3 leu2 trp1 ura3), DY150 (MATa ade2 can1 his3 leu2 trp1 ura3) and ZHY6 (DY1457 zap1Δ::TRP1) have also been described previously [22]. The chromosomal ZRG17 mutant, zrg171m2ZRE, was constructed by first integrating the CORE (counterselectable reporter) cassette [23] containing the Kluyveromyces lactis URA3 gene and kanMX4 into the ZRG17 promoter of DY1457. The promoter fragment from pZRG17-m2ZRE-lacZ (see below) was then amplified by PCR and transformed into the CORE cassette-containing strain and selected for loss of the URA3 gene by selection on 5-fluoro-orotic acid [24]. Correct mutation of the chromosomal ZRG17 promoter was confirmed by PCR and DNA sequencing. The reconstructed WT (wild-type) ZRG17 strain, ZRG17rZRE, was constructed in the same way with a promoter fragment amplified from pZRG17-lacZ containing the WT promoter. To generate the zrg17Δ mutant in the DY1457 background, the KanMX cassette with 500 bp flanking the ZRG17 open reading frame was amplified by PCR from the CEY9 (DY150 zrg17Δ::KanMX) [2]. The PCR fragment was then transformed into DY1457 strain to generate DY1457 zrg17Δ.

Yeast plasmids

pYef2 (Vec) [25], pAFH35 (Zap1up) [26], YEp353 (Vec) [27], pZRG17-HA [2] and pMCZ-Y [28] [UPRE (UPR element)–lacZ; provided by A. Cooper, Garvan Institute, Sydney, Victoria, Australia] were described previously. Reporter plasmid pZRG17-lacZ was constructed in YEp353 by homologous recombination. PCR products were generated from genomic DNA that contained 1000 bp of ZRG17 promoter sequence (bases −1000 to +1) flanked by homology with the vector. This fragment was gel-purified and co-transformed with EcoRI- and BamHI-digested YEp353; transformants were selected for URA3 prototrophy. The mutant alleles of ZRG17 ZRE (pZRG17-m1ZRE-lacZ and pZRG17-m2ZRE-lacZ) were constructed in a similar fashion after generation of the mutant promoter fragments by overlapping PCR. pYef2L(Vec) and pYef2L-Zap1–6x-myc (Zap1–myc) used in ChIP (chromatin immunoprecipitation) were constructed as described previously [18]. All plasmid constructs were confirmed by sequencing.

RNA and protein analyses

S1 nuclease protection assays were performed with total RNA as described previously [29]. Total RNA was extracted from cells grown to mid-exponential phase with hot acid phenol. For each reaction, 15 μg of total RNA was hybridized to 32P-end-labelled DNA oligonucleotide probes for ZRG17 (5′-CGGGGGAAATGCCTCTTACCGGTGATCTTGTTCTGGGAGGAGGCGGCACCAGCTTTGGTGCTGGTACGCGCC-3′) and CMD1 (calmodulin 1) (5′-GGGCAAAGGCTTCTTTGAATTCAGCAATTTGTTCTTCGGTGGAGCC-3′) before digestion with S1 nuclease and separation on a 10% polyacrylamide, 5 M urea polyacrylamide gel. Band intensities were quantified by phosphorimager analysis (PerkinElmer Life Sciences). Protein extracts were generated by lysis in trichloroacetic acid, and immunoblot analysis was performed as described previously [30]. The primary antibodies used were mouse anti-HA (haemagglutinin) (12CA5, Roche Applied Science) and mouse anti-Pgk1 (3-phosphoglycerate kinase 1) (Molecular Probes). The secondary antibody used was HRP (horseradish peroxidase)-conjugated goat anti-mouse IgG (Pierce Chemical Co.). Band intensities were measured using ImageJ (NIH).

β-Galactosidase assays

Cells were grown to mid-exponential phase in LZM supplemented with the indicated amount of ZnCl2. β-Galactosidase activity was measured in permeabilized cells as described previously [31], and activity was normalized to cell density. For UPRE–lacZ analysis, β-galactosidase assays were performed on protein extracts and specific activity was normalized to protein content [32]. LZM+0.3 μM ZnCl2 was used for the most zinc-limiting conditions in these assays instead of higher conditions (e.g. LZM+1 μM ZnCl2) because the induction of the UPR was highest in all strains under these conditions and allowed for clearest assessment of Zrg17's contribution to ER stress resistance.

EMSAs (electrophoretic mobility-shift assays)

The Zap1 DNA-binding domain (Zap1DBD, residues 687–880) was expressed in Escherichia coli as a fusion to glutathione transferase and purified [33]. EMSAs were performed as described previously using purified Zap1DBD [33] and radiolabelled oligonucleotides (Table 1). 32P-end-labelled (50 pmol) oligonucleotides were purified using G-50 Quick Spin columns (Roche Applied Science). Double-stranded oligonucleotides were prepared by annealing complementary single-stranded oligonucleotides (1 μM) in 10 mM Tris/HCl (pH 7.5), 100 mM NaCl and 1 mM EDTA. The annealed mixtures were incubated for 15 min at 85 °C and then 55 °C for 4 h. For EMSAs, 15 μl reactions were prepared containing 0.5 pmol of radiolabelled oligonucleotide (20000 c.p.m./pmol), 10 mM Tris/HCl (pH 8.0), 10 mM MgCl2, 50 mM KCl, 1 mM dithiothreitol, 0.02 mg/ml poly(dI-dC)·(dI-dC), 0.2 mg/ml BSA, 0.04% Igepal CA-630, 10% glycerol and the indicated concentrations of purified Zap1DBD. After incubation for 1 h at room temperature (20 °C), the samples were resolved on 6% polyacrylamide gels. Gels were dried on to blotting paper, and the signals were measured by autoradiography.

View this table:
Table 1 Oligonucleotides used for EMSAs

ZRG17 m1ZRE was mutated such that each position in the potential ZRE was altered by a transversion mutation. ZREs or regions mutated in each complementary oligonucleotide pair are indicated by the underline. The lower-case letters indicate EagI- and SalI-complementary overhangs included for cloning these fragments into a lacZ reporter plasmid for future studies.


ChIP was performed as described in [34]. WT cells transformed with either the vector (pYef2L) or a plasmid expressing a myc-tagged Zap1 protein (pYef2L-Zap1–6x-myc) [18] were grown to a D600 of ~0.5 in LZM+3 μM ZnCl2. This level of zinc was used because cells grown with lower levels of zinc are more difficult to lyse and chromatin recovery is much lower. The cells were then treated with 1% (w/v) formaldehyde to cross-link protein–DNA complexes. The cross-linking reaction was quenched by adding 2.5 ml of 2 M glycine. After two 25 ml washes with ice-cold PBS, the cells were lysed with glass beads in buffer containing Complete™ protease inhibitor cocktail (Roche), 1 mM PMSF and 2 mM benzamidine. Following centrifugation at 16000 g for 10 min, the supernatants were immunoprecipitated with anti-myc antibody at 4 °C overnight and isolated with Protein A–Sepharose. The cross-links were reversed in Tes, and co-immunoprecipitation of specific promoter fragments with Zap1–myc was assessed by PCR using primers flanking the ZRG17 ZRE by 100 bp. Primers specific to the ZRT1 ZRE and CMD1 were used as positive control and negative controls respectively. PCR products generated from 10-fold serially diluted input samples were used to confirm the semi-quantitative nature of the analysis.


Zinc regulation of ZRG17 mRNA abundance

Our previous results suggested that ZRG17 is a direct target of Zap1 regulation [11,12]. Microarray analyses indicated that ZRG17 mRNA levels were elevated in zinc-limited cells in a Zap1-dependent manner and were also elevated in zinc-replete cells expressing a constitutive allele of Zap1, Zap1up, which contains mutations that disrupt zinc sensing (Figure 1A). These effects are similar to those observed for several known Zap1 target genes including ZRT1, ZAP1, TSA1, ZRC1 and FET4. The results for ZRG17 were confirmed when we assayed mRNA levels by S1 nuclease protection assay with cells grown under the same conditions as were used for the microarray experiments (Figure 1B). ZRG17 mRNA abundance increased ~2-fold in zinc-limited WT cells, but not zap1Δ mutant cells. In addition, ZRG17 mRNA abundance increased ~3-fold in zinc-replete cells expressing the constitutive Zap1up allele. When a functional HA-epitope-tagged ZRG17 allele was expressed from its own promoter, we found that ZRG17 mRNA was induced by relatively severe zinc deficiency (LZM+≤10 μM ZnCl2) and that the levels of Zrg17–HA protein showed a similar dose-response and magnitude of regulation (Figure 2, and results not shown). These results suggest that control of ZRG17 transcription is the principal mechanism regulating Zrg17 protein accumulation in response to zinc. CMD1 mRNA and the Pgk1 protein are not zinc-regulated and these served as loading controls for the S1 nuclease protection assays and immunoblots respectively. A lower-molecular-mass Zrg17–HA band of ~50 kDa was observed in the LZM+1 μM ZnCl2-grown cells, perhaps resulting from partial proteolysis. Supporting this hypothesis, we have shown previously that expression of some vacuolar proteases is induced by zinc limitation [11].

Figure 1 S1 nuclease protection assays confirm microarray results

(A) Previous microarray results for ZRG17 and selected previously confirmed Zap1 target genes [11,12]. (B) ZRG17 mRNA levels were analysed by S1 nuclease-protection assays. Total RNA was isolated from cells grown under the same conditions as the microarray experiments. zrg17Δ mutant cells, WT cells (DY1457) or zap1Δ mutant cells (ZHY6) were grown in LZM supplemented with high zinc (+, 1000 μM ZnCl2) or low zinc (−, 1 μM ZnCl2) (left-hand panel). LZM contains 1 mM EDTA and 20 mM citrate to buffer available zinc levels. Also, WT (DY1457) cells expressing either the vector (pYef2) or a Zap1up plasmid (pAFH35) were grown in zinc-replete SD medium containing 1 μM ZnCl2 and 2% galactose as the carbon source to drive expression of the constitutively active Zap1up allele from the GAL1 promoter (right-hand panel). CMD1 was used as a loading control. The fold changes shown were quantified from the level of ZRG17 mRNA normalized to CMD1 in each lane. Results are representative of two independent experiments.

Figure 2 Zrg17 protein levels correlate with ZRG17 mRNA

zrg17Δ mutant cells or zrg17Δ cells expressing a C-terminal HA-epitope-tagged allele of Zrg17 expressed from its own promoter (pZRG17-HA) were grown in LZM supplemented with the indicated concentration of ZnCl2. (A) S1 nuclease-protection assays were performed to detect ZRG17 and CMD1 mRNA. The fold changes shown were quantified from the level of ZRG17 mRNA normalized to that of CMD1. (B) Immunoblot analysis of Zrg17–HA protein levels. Lysates were prepared from the same cells used in (A) and subjected to immunoblotting with anti-HA antibody. Pgk1 served as a loading control. The asterisk indicates a non-specific background band. The fold changes shown were quantified from the level of Zrg17 protein normalized to that of Pgk1. A lower-molecular-mass Zrg17–HA band was observed in the LZM+1 μM ZnCl2-grown pZRG17-HA transformants, perhaps resulting from partial proteolysis, and both full-length and truncated Zrg17 forms were included in the quantification. Results are representative of two independent experiments.

A potential Zap1-binding site in the ZRG17 promoter

To examine further the mechanism of ZRG17 regulation, we constructed a lacZ reporter gene in which the ZRG17 promoter, extending 1000 bp upstream of the open reading frame, was fused to the E. coli lacZ gene. As shown in Figure 3(A), the ZRG17lacZ reporter was highly regulated by zinc in WT cells, being induced approx. 6-fold in zinc-limited cells relative to zinc-replete conditions. This induction was not observed in zap1Δ mutant cells. Expression in zinc-replete zap1Δ cells was similar to that observed in WT cells, but expression decreased in zinc-limited cells to very low levels. This indicated that the basal expression of ZRG17 in zinc-replete cells is Zap1-independent and Zap1 function is required for the elevated expression observed in zinc-limited cells.

Figure 3 A candidate ZRE in the ZRG17 promoter is required for its induction

(A) β-Galactosidase activity was measured in WT cells (DY1457) or zap1Δ mutant cells (ZHY6) bearing the WT ZRG17lacZ reporter and grown in LZM over a range of zinc concentrations. (B) The ZRG17 promoter with the WT ZRE (ZRG17 ZRE) and two different promoter mutants constructed (m1ZRE and m2ZRE). Both ZRG17 m1ZRE and ZRG17 m2ZRE were mutated so that each of the 11 positions of the ZRE was altered by transversion mutations. ZRG17 m2ZRE has three additional base pairs (TGA) adjacent to the ZRE deleted. (C) β-Galactosidase activity was measured in WT cells (DY1457) bearing the vector (YEp353), the WT ZRG17lacZ reporter (WT) or the mutated reporters (ZRG17–m1ZRE and ZRG17–m2ZRE) and grown under low-zinc (LZM+1 μM ZnCl2) or high-zinc (LZM+1000 μM ZnCl2) conditions. Results are the means±S.D. for three independent cultures for each condition, representative of two independent experiments.

Motif analysis of the ZRG17 promoter indicated the presence of a candidate ZRE located at position −66 to −55 upstream of the ZRG17 transcription start site (Figure 3B) [11,35]. This sequence matched the consensus ZRE sequence compiled from experimentally verified and other candidate Zap1 target genes. To determine the importance of this potential ZRE to ZRG17 regulation, we constructed two different promoter mutant constructs. In mutant m1ZRE, transversion mutations were introduced in all positions of the ZRE. Because adjacent base pairs may also contribute to Zap1 binding [17], we constructed a second promoter mutant, m2ZRE, in which an adjacent 3 bp were also deleted. As shown in Figure 3(C), mutation of the ZRG17 ZRE, with or without deletion of the adjacent base pairs, completely eliminated induction of ZRG17 expression in zinc-limited cells. Notably, neither ZRE mutation affected zinc-replete expression, confirming that expression under these conditions is Zap1-independent.

Zap1 binds specifically to the ZRG17 ZRE in vitro and in vivo

To determine whether Zap1 binds to the ZRG17 ZRE sequence, EMSAs were performed using Zap1DBD purified from E. coli. An oligonucleotide containing a known Zap1-binding site from the TSA1 promoter was used as a positive control (Figure 4A, lanes 1 and 2). When increasing amounts of Zap1DBD were incubated with a ZRG17 ZRE oligonucleotide, increased abundance of a Zap1DBD–DNA complex was also observed (Figure 4A, lanes 3–6). However, no binding to a mutant m1ZRE probe was observed when 0.6 or 1.2 μg of the Zap1DBD was used in the reaction and only a very slight amount was detected when the highest amount of Zap1DBD protein (2.4 μg) was used (Figure 4A, lanes 7–10). In addition, a competition binding experiment confirmed that the Zap1DBD binds specifically to the ZRG17 ZRE. In this experiment, either the WT ZRG17 ZRE or the mutant m1ZRE oligonucleotides were assessed for their ability to compete for binding with the TSA1 ZRE fragment. Although the WT ZRE was an effective competitor (Figure 4B, lanes 3–6), no decrease in binding was observed with the mutant ZRE (Figure 4B, lanes 7–10). These results indicate specific binding of Zap1 to the potential ZRG17 ZRE. Slower mobility bands were observed in the presence of the WT competitor DNA (Figure 4B, lanes 3–5) suggesting that Zap1 may be able to bind to more than one oligonucleotide simultaneously when DNA concentrations are very high, perhaps through some lower-affinity zinc finger–DNA interactions.

Figure 4 Zap1 binds specifically to the ZRG17 ZRE in vitro and in vivo

(A) EMSA showing sequence-specificity of Zap1DBD binding to the ZRG17 ZRE. Radiolabelled oligonucleotides with the WT TSA1 ZRE1 (ZRE1TSA1), WT ZRG17 ZRE (ZREZRG17) or the mutated ZRG17 m1ZRE (mutant ZREZRG17) were incubated with purified Zap1DBD for 1 h before native gel electrophoresis. The experiments were performed with the indicated amounts of Zap1DBD. FP denotes the free probe, and the arrowhead indicates the Zap1DBD–ZRE complex. (B) EMSA competition assay showing that the ZRG17 ZRE competes with TSA1 ZRE1 for Zap1DBD binding. Radiolabelled WT TSA1 ZRE1 (ZRE1TSA1) oligonucleotides were incubated with 0 (−) or 2.4 μg (+) Zap1pDBD and a 25-, 50-, 100- or 200-fold excess of the non-radiolabelled WT ZRG17 ZRE (ZREZRG17) or the mutated ZRG17 m1ZRE (mutant ZREZRG17) oligonucleotides for 1 h at room temperature before native gel electrophoresis. (C) ChIP verifies Zap1 binding to the ZRG17 promoter in vivo. WT cells transformed with either the vector (pYef2L) or a plasmid expressing a myc-tagged Zap1 protein (Zap1–myc) were grown under low-zinc conditions (LZM+3 μM ZnCl2). ChIP was performed as described in the Experimental section. Enrichment of the ZRG17 ZRE was observed in the immunoprecipitates (IP) from Zap1–myc-expressing cells relative to the vector-only cells. The ZRT1 ZRE and CMD1 promoter regions were used as positive and negative controls respectively. PCR products generated from 10-fold serially diluted input samples were used to confirm the semi-quantitativeness of the analysis. Results are representative of two independent experiments.

To assess whether Zap1 binds to the ZRG17 ZRE in vivo, we performed a ChIP experiment. WT cells transformed with either the vector or a plasmid expressing a myc-tagged Zap1 protein were grown under low-zinc conditions and then treated with 1% formaldehyde to cross-link protein–DNA complexes. Chromatin was then isolated and sheared by sonication, and Zap1 was immunoprecipitated with anti-myc antibody. The cross-links were reversed and co-immunoprecipitation of specific promoter fragments with Zap1–myc was assessed by PCR using primers flanking the ZRG17 ZRE. Co-immunoprecipitation of the ZRG17 promoter was observed in the immunoprecipitates from Zap1–myc-expressing cells, but not in those from the vector control cells (Figure 4C). Co-immunoprecipitation was also observed for the ZRT1 promoter, a known Zap1 target gene, but no enrichment was detected for the non-zinc-responsive CMD1 promoter, which served as a negative control. These results indicate that Zap1 binds to the ZRG17 promoter in vivo as well as in vitro.

Biological importance of ZRG17 transcriptional control

Having established that ZRG17 is indeed a direct Zap1 target gene, we next addressed the importance of Zap1 regulation to the function of the Zrg17 protein. As shown in Figures 1 and 3, our results indicated that Zap1 regulation was required for maximal expression of ZRG17 in zinc-limited cells, but was not needed for basal expression in zinc-replete cells. Therefore, to assess the importance of Zap1 regulation for Zrg17 function, we replaced the ZRE in the chromosomal ZRG17 gene with the non-functional m2ZRE mutant sequence to generate the zrg171m2ZRE allele. The rest of the ZRG17 promoter and the complete open reading frame are unaltered in the zrg171m2ZRE allele, and thus only its regulation by Zap1 was predicted to be affected. As a control, we simultaneously generated a strain identical with zrg171m2ZRE in which we restored the WT ZRE sequence and this strain was designated ZRG17rZRE (‘rZRE’ for reconstructed ZRE, see the Experimental section). S1 nuclease-protection assays indicated that the chromosomal m2ZRE mutation disrupted induction of chromosomal ZRG17 under low-zinc conditions, but did not greatly affect expression in zinc-replete cells (Figure 5A). Expression of ZRG17 in the ZRG17rZRE strain was similar to that in the WT. These data confirmed that the m2ZRE mutation disrupts Zap1-mediated ZRG17 induction in low zinc, but does not alter basal expression.

Figure 5 Zap1 regulation of ZRG17 expression is important for ER function

(A) ZRG17 mRNA levels were analysed by S1 nuclease-protection assay of RNA isolated from WT (DY1457), a chromosomal ZRG17 ZRE mutant (zrg171m2ZRE) and reconstructed WT ZRE (ZRG17rZRE) cells grown in LZM supplemented with 1000 μM ZnCl2 (+) or 1 μM ZnCl2 (−). CMD1 was used as a normalization control. Results are representative of two independent experiments. (B) WT (DY1457), zrg17Δ mutant (DY1457 zrg17Δ), chromosomal ZRG17 ZRE mutant (zrg171m2ZRE) and reconstructed WT ZRE (ZRG17rZRE) cells were grown in SD liquid medium overnight. Cultures were then diluted in fresh medium, and 5 μl volumes containing 104, 103 and 102 cells (from left to right) were plated on to YPD medium and YPGE medium plates. The plates were incubated at 30 or 37 °C and photographed after 3 days. (C) UPRE–lacZ activity (pMCZ-Y) was assayed in the same cells as in (B) grown in LZM supplemented with 0.3 μM ZnCl2, 3 μM ZnCl2 and 100 μM ZnCl2. Results are means ± S.D. for three independent cultures for each condition, representative of two independent experiments. The letters (a, b and c) denote statistically significant differences. Values marked with the same letter were not significantly different from WT cells grown under the same zinc condition, whereas values marked with different letters were significantly different from WT and from each other (P<0.05).

Direct assays of Msc2–Zrg17 complex activity, e.g. 65Zn transport assays, have not yet been developed. However, we could assess function of this complex using phenotypic assays that serve as indirect indicators of transporter activity and ER function. For example, zrg17Δ mutants are unable to grow at 37 °C on rich yeast extract/peptone medium containing glycerol and ethanol as carbon sources (i.e. YPGE medium) [2] (Figure 5B). This defect is likely to be due to disruption of cell wall synthesis caused by perturbation of zinc levels in the ER [2]. No such defect is seen at 30 °C, on glucose-containing YPD medium, or when high levels of zinc are supplemented in the medium [2]. As shown in Figure 5(B), neither the zrg171m2ZRE mutant nor the ZRG17rZRE strain showed any growth defect on glycerol/ethanol-containing plates at 37 °C. Yeast extract/peptone medium is relatively zinc-replete, so these data indicate that basal Zap1independent expression of ZRG17, as occurs in the zrg171m2ZRE promoter mutant, produces sufficient levels of this protein for function in zinc-replete cells.

We have also shown previously that the Msc2–Zrg17 complex is required for ER function in zinc-limited cells grown in glucose-containing media such as LZM [2]. This was carried out using a reporter of ER stress that responds to the UPR. The UPR is triggered by misfolded proteins in the ER and up-regulates the expression of protein chaperones and degradation systems to refold or degrade the aberrant proteins. This control occurs at the transcriptional level and is mediated by the Hac1 transcription factor using regulatory elements (UPREs) in UPR-regulated promoters. To assess the importance of Zap1-mediated ZRG17 regulation to ER function in zinc-limited cells, we used a UPRE–lacZ reporter that responds to ER stress. As we have shown previously, expression of a UPRE–lacZ reporter increased in zinc-limited WT cells indicating the ER requirement for zinc. UPR induction was even higher in a zrg17Δ mutant, indicating the importance of this zinc transporter to ER function under low-zinc conditions [2] (Figure 5C). Higher concentrations of zinc suppressed UPR induction in both WT and zrg17Δ mutants and probably reflects the transport of zinc into the secretory pathway via other transporters [1,2]. UPRE–lacZ expression in the ZRG17rZRE strain was identical with that in the WT, which was consistent with normal ZRG17 expression in that strain bearing a reconstructed WT promoter. In the zrg171m2ZRE mutant where Zap1-dependent ZRG17 induction in low zinc is disrupted, however, UPRE–lacZ expression was induced to a higher degree than in WT cells, albeit not to as high a level as in the zrg17Δ null mutant. These results indicate that, whereas basal Zap1-independent expression of ZRG17 does provide some Zrg17 function in zinc-limited cells, the additional level of ZRG17 expression provided by Zap1 induction is required for full activity under those conditions.


Previous studies have estimated that Zap1 induces the expression of approx. 80 genes in zinc-limited yeast cells [11,12]. These include genes involved in adaptation to zinc-limiting conditions and genes involved in maintaining zinc homoeostasis within the cell and within intracellular compartments [10]. Among Zap1 targets are several genes encoding zinc transporters that play key roles in zinc homoeostasis. For example, ZRT1, ZRT2 and FET4 encode the high-affinity and low-affinity zinc transporters required for efficient zinc uptake across the plasma membrane [3638]. ZRT3 encodes a vacuolar zinc transporter responsible for mobilization of zinc stores from the vacuole for use by other compartments of the cell [39]. Conversely, ZRC1 encodes a transporter that moves zinc into the vacuole. Zrc1 activity is important for zinc storage and its induction in zinc-limited cells helps to protect those cells from the high level of zinc that can be taken in when they are resupplied with zinc [40].

To this list of Zap1-regulated genes, we can now add ZRG17 as an important target for maintaining zinc homoeostasis. Zap1 induces ZRG17 expression by as much as 5-fold (Figure 5). Zinc transport into the ER of zinc-limited cells is mediated largely by the Msc2–Zrg17 complex [1,2]. The vacuolar zinc transporters Zrc1 and Cot1 also contribute, perhaps while they pass through the early secretory pathway on their way to their final vacuolar localization. Despite the contributions of the vacuolar transporters, Msc2 and Zrg17 appear to play the major role in maintaining ER function. On the basis of our genetic studies, both Msc2 and Zrg17 are required for transporter activity, and biochemical analyses indicated that these proteins form heteromeric complexes. Results from studies of other members of the CDF/ZnT/SLC30A family [9,41] suggest that Msc2 and Zrg17 form heterodimers and formation of this complex is required for transporter function.

Some CDF/ZnT/SLC30A transporters are active as homodimers. For these proteins, both subunits probably contribute to zinc transport. In the case of the heteromeric complexes in this family, the role of the individual subunits is much less clear. One tool for assessing the zinc-transport role of the subunits in these complexes is the crystal structure model of the E. coli YiiP zinc efflux transporter [41,42]. This structural model has indicated that a zinc-binding site using ligand residues from transmembrane domains II and V is the likely site of transient zinc occupancy during transport. For the ZnT-5–ZntT-6 heterodimer, this binding site is conserved in ZnT-5, but is not present in ZnT-6. This observation suggested that the ZnT-6 subunit is not involved in transport and may play a structural role in the transporter complex [9,43]. Similarly, the zinc-binding site is conserved in Msc2, but not in Zrg17. Given that Zrg17 expression is regulated by zinc and the possibility that Zrg17 does not act directly as a transporter, we hypothesize that Zrg17 serves as a regulatory subunit of the complex.

Regardless of its role, Zrg17 appears to be the rate-limiting subunit for transporter function in zinc-limited cells. This was suggested by our observation that disrupting Zap1-mediated induction of ZRG17, but not its basal expression, by mutating the ZRE in its promoter prevented full function in zinc-limited cells when assayed using the UPRE–lacZ ER stress reporter. Levels of Zrg17 protein correlated well with Zap1-mediated mRNA regulation, indicating that transcriptional control is the primary mechanism regulating Zrg17 accumulation in response to zinc. However, we note that Zrg17 is phosphorylated in vivo, suggesting that post-translational modifications may also contribute to its regulation [44]. Moreover, the lower-molecular-mass form of Zrg17 observed in Figure 2(B) that we attributed to proteolysis may instead be the result of some specific post-translational regulation. These other levels of regulation could also influence Zrg17 activity in response to zinc or other signals.

As we learn more about how zinc transporters of the early secretory pathway are regulated, we can begin to compare the regulation of the yeast systems with those in higher eukaryotes, specifically heterodimeric ZnT-5–ZnT-6 and homodimeric ZnT-7. In investigations of transcript levels, it was noted that both ZnT-5 and ZnT-7 mRNA levels are increased in zinc-limited cells analogously to the regulation we have observed in yeast [45]. Whereas the mechanism of ZnT-7 regulation has not been explored further, it appears that ZnT-5 is regulated both transcriptionally and at the level of mRNA stability [46]. ZnT-5 promoter activity is increased in zinc-limited cells, whereas mRNA stability decreases under low zinc conditions. Given that the net effect is an increase in mRNA under low zinc conditions, the transcriptional control appears to outweigh the mRNA-stability effects. The zinc sensors and regulatory factors responsible for these changes are currently unknown. It has also been found that ZnT-5 is regulated by the UPR pathway mediated by XBP1 (X-box-binding protein 1) [7]. Given that zinc is required for ER function and the UPR is induced by zinc deficiency [1,2], it is logical that cells would induce zinc transport activity in the early pathway to fully meet the requirements of those compartments. Microarray studies of yeast suggested that ZRG17 was regulated by the yeast orthologue of XBP1, the Hac1 transcription factor [47]. However, when ZRG17 expression in response to UPR inducers such as tunicamycin and dithithreitol was tested with S1 nuclease protection assays, no such regulation was observed (C.Y. Wu and D.J. Eide, unpublished work). These results indicate that ZRG17 is probably not a target of the UPR.

Finally, we are also considering the regulation of Msc2, the other subunit of the yeast complex. MSC2 mRNA was not found to be zinc-regulated in our previous microarray experiments [11,12] and direct analysis by S1 nuclease-protection assays confirmed that MSC2 mRNA levels do not change in response to zinc status (C. Y. Wu and D. J. Eide, unpublished work). We are currently investigating whether Msc2 accumulation or activity are regulated at post-transcriptional levels. These studies will ultimately lead us to an integrated understanding of the regulation of zinc homoeostasis in the early secretory pathway of eukaryotic cells.


Yi-Hsuan Wu, Avery Frey and David Eide conceived the study, and designed and interpreted the experiments; Yi-Hsuan Wu and Avery Frey conducted the experiments; and Yi-Hsuan Wu and David Eide wrote the paper.


This work was supported by the National Institutes of Health [grant numbers RO1-GM056285 and T32-DK007665] and the U.S. Department of Agriculture [Hatch grant number WIS01323].

Abbreviations: CDF, cation diffusion facilitator; ChIP, chromatin immunoprecipitation; CMD1, calmodulin 1; CORE, counterselectable reporter; EMSA, electrophoretic mobility-shift assay; ER, endoplasmic reticulum; GPI, glycosylphosphatidylinositol; HA, haemagglutinin; LZM, low-zinc medium; Pgk1, 3-phosphoglycerate kinase 1; SD, synthetic dextrose; SLC, solute carrier; UPR, unfolded protein response; UPRE, UPR element; WT, wild-type; XBP1, X-box-binding protein 1; YPD, yeast extract/peptone/dextrose; YPGE, yeast extract/peptone/glycerol/ethanol; Zap1DBD, Zap1 DNA-binding domain; ZnT, zinc transporter; ZRE, zinc-responsive element


View Abstract