The Zap1 transcriptional activator also acts as a repressor by binding downstream of the TATA box in ZRT2

Abstract

The zinc-responsive transcriptional activator Zap1 regulates the expression of both high- and low-affinity zinc uptake permeases encoded by the ZRT1 and ZRT2 genes. Zap1 mediates this response by binding to zinc-responsive elements (ZREs) located within the promoter regions of each gene. ZRT2 has a remarkably different expression profile in response to zinc compared to ZRT1. While ZRT1 is maximally induced during zinc limitation, ZRT2 is repressed in low zinc but remains induced upon zinc supplementation. In this study, we determined the mechanism underlying this paradoxical Zap1-dependent regulation of ZRT2. We demonstrate that a nonconsensus ZRE (ZRT2 ZRE3), which overlaps with one of the ZRT2 transcriptional start sites, is essential for repression of ZRT2 in low zinc and that Zap1 binds to ZRT2 ZRE3 with a low affinity. The low-affinity ZRE is also essential for the ZRT2 expression profile. These results indicate that the unusual pattern of ZRT2 regulation among Zap1 target genes involves the antagonistic effect of Zap1 binding to a low-affinity ZRE repressor site and high-affinity ZREs required for activation.

Figure 1

Regulation of ZRT1 and ZRT2 transcription in response to zinc. Total RNA was extracted from exponential-phase cultures of the zap1 mutant strain ZHY6 grown in LZM media supplemented with 3000 μM Zn²⁺ (lane 1) and from the wild-type strain, DY1457, grown in LZM media supplemented with 3, 10, 30, 100, 300, 1000 and 3000 μM Zn²⁺ (lanes 2–8, respectively). The levels of ZRT1 and ZRT2 mRNA were compared to the loading control CMD1 mRNA using S1 nuclease protection assays.

Figure 2

Mapping the repressor binding site. (A) The indicated reporter constructs were transformed into wild-type DY1457. All cultures were grown to exponential phase in LZM media supplemented with the indicated amount of Zn²⁺. β-Galactosidase activity was measured in triplicate by standard procedures. The numbers shown indicate the internal deletion end points. All numbers are relative to the first base of the initiation codon of lacZ, which is designated as +1. ZRE elements (open box), the TATA box (filled oval) and the lacZ gene (hatched box) are shown. (B) The single-copy plasmid pmZRE3 was introduced into the zap1 zrt2 mutant strain ZHY11 and the zrt2 mutant strain ZHY2. Total RNA was extracted from exponential-phase cultures of ZHY11 pmZRE3 that had been grown in LZM media supplemented with 3000 μM Zn²⁺ (lane 1) and from ZHY2 pmZRE3 grown in LZM media supplemented with 3, 10, 30, 100, 300, 1000 and 3000 μM Zn²⁺ (lanes 2–8, respectively). The levels of ZRT2 mRNA were compared to the loading control CMD1 mRNA using S1 nuclease protection assays.

Figure 3

Identification of ZRE3 in the ZRT2 promoter. An alignment of ZRT2 ZRE3 with the consensus ZRE and other known ZRE elements from the ZRT1 and ZRT2 promoters (A). The numbers indicate the first and last nucleotides of each element. (B) The activity of the minimal promoter reporter constructs pDg2, pDg–ZRE1, pDg–ZRE2 and pZRT2 ZRE3 was examined in the wild-type strain DY1457. The constructs contain the ZRT1 ZRE1, ZRT2 ZRE1, ZRT2 ZRE2 and ZRT2 ZRE3 inserted into a minimal CYC1 promoter, respectively. All cultures were grown to exponential phase in LZM media supplemented with the indicated amount of Zn²⁺. β-Galactosidase activity was measured in triplicate by standard procedures.

Figure 4

Zap1 binds to ZRT2 ZRE3 with a low affinity in vitro and in vivo. Electrophoretic mobility shift assays were performed with increasing amounts of purified Zap1_642–880 protein ranging from 1 pM to 1 μM, a constant radiolabeled ZRE concentration and the ZRE oligonucleotide probes ZRT1 ZRE1 or ZRT2 ZRE3. (A) Binding isotherm plots were generated by quantifying phosphorimages of the ZRT1 ZRE1 and ZRT2 ZRE3 EMSA. A representative experiment for each oligonucleotide probe is shown. (B) The plasmid pYCpZRT2 or derivatives containing the indicated ZREs (C–E) were introduced into the zrt2 mutant strain ZHY2. Total RNA was extracted from exponential-phase cultures of ZHY2 grown in LZM media supplemented with 3, 10, 30, 100, 300, 1000 and 3000 μM Zn²⁺ (lanes 1–7, respectively). The levels of ZRT2 mRNA were compared to the loading control CMD1 mRNA using S1 nuclease protection assays.

Figure 5

Repression via ZRE3 is active on transcription factors other than Zap1. The plasmid pYCpZRT2 or derivatives containing the indicated deletion/substitution mutations were introduced into the zrt2 mutant strain ZHY2 or the zap1 zrt2 mutant strain ZHY11. Total RNA was extracted from exponential-phase cultures of ZHY11 grown in LZM media supplemented with 3000 μM Zn²⁺ and from ZHY2 grown in LZM media supplemented with 3 or 3000 μM Zn²⁺ (−Zn and +Zn, respectively) (A) or from cells that had been pregrown to exponential phase in SC media before transfer to LZM media supplemented with 3 μM Zn²⁺ for a further 0, 3 or 5 h (B). The levels of ZRT2 mRNA were compared to the loading control CMD1 mRNA using S1 nuclease protection assays. The ZRT2 ZREs (open boxes), ZRT2 TATA box (closed oval), Rap1 UAS (boxed ‘X') and ZRT2 open reading frame (hatched box) are shown. A closed box indicates the disruption of ZRE3 by transversion mutations. Numbers indicate ZRT2 sequence end points.

Figure 6

S1 nuclease protection analyses of the 5′ end of ZRT2 mRNA. Total RNA from wild-type DY1457 (ZRT2) cells or from zrt2 mutant cells containing pRap1^UAS that had been grown to exponential phase in LZM media supplemented with 300 or 3 μM Zn²⁺, respectively, was subject to S1 nuclease analysis using probe P1 (A). The 75 bp P1 probe is complementary to nucleotides −31 to −105 in the ZRT2 promoter and partially overlaps with ZRE3. The position of the 5′ end of ZRT2 mRNA was estimated by comparing the sizes of the S1 nuclease products (P1) to products of known size (MW). The sequence of the 5′ end of P1 is shown. The nucleotides that are complementary to ZRE3 (underlined) and the positions of the 5′ end of the S1 nuclease products of size 51, 60, 69 or 75 bp are indicated. Arrows indicate the protected S1 nuclease products that map the 5′ ends of the mRNA. (B) The plasmids pMA424 (V), pGBD–Zap1_1–880 (Zap1) or pGBD–Zap1_642–880 (DBD) were introduced into the strain ZHY2 containing either pRap1^UAS (see Figure 5A, panel b) or pRap1^UASmZRE3 (see Figure 5A, panel c). Cells were grown in LZM media supplemented with 3 or 1000 μM Zn²⁺ (−Zn and +Zn, respectively) and S1 nuclease protection assays were performed as described before.

Figure 7

Hos2 or Hos3 loss leads to constitutive ZRT2 activation. Total RNA was extracted from exponential-phase cultures of wild-type strain DY1457, ssn6 mutant strain MAP6, hos1 mutant strain DY6073, hos2 mutant strain DY4549 and hos3 mutant strain DY8363 grown in LZM media supplemented with 3 or 300 μM Zn²⁺ (−Zn and +Zn, respectively) (A) or the indicated strains grown in LZM media supplemented with 3000 μM Zn²⁺ (B, C). The hos2 and hos3 mutant strains contained the plasmids pMA424 (V) and pGBD–Zap1¹⁻⁸⁸⁰ (Zap1) in panel C. The levels of ZRT2 and ZRT1 mRNA were compared to the loading control CMD1 mRNA using S1 nuclease protection assays.

Figure 8

The low affinity of ZRE3 is required for proper ZRT2 regulation and Zrt2 function. (A) The constructs pYCpZRT2, pYCpZRT2-H (a derivative containing ZRE3 replaced with ZRT1 ZRE1) and vector pRS316 were introduced into the zrt1 zrt2 mutant strain ZHY3. In all, 5 μl of a cell suspension (OD₆₀₀ of 1.0) and 10-fold serial dilutions (left to right) were plated onto SD agar plates supplemented with 1 mM EDTA and 400 μM Zn²⁺ (−), 1 mM EDTA and 800 μM Zn²⁺ (+) or 10 μM Zn²⁺ (++). Plates were incubated for 3 days at 30°C. (B, C) Growth of the zrt1 mutant strain transformed with either pZRT1, pRS316 (vector), pYCpZRT2 (pZRT2) or pmZRE3 in LZM media supplemented with 3 μM Zn²⁺ (−Zn) or 3000 μM Zn²⁺ (+Zn). Triplicate cultures were inoculated at an initial A₆₀₀ of 0.01 and the A₆₀₀ was measured over time. Results of a representative experiment are shown.