A novel class of plant-specific zinc-dependent DNA-binding protein that binds to A/T-rich DNA sequences

Abstract

Complementary DNA encoding a DNA-binding protein, designated PLATZ1 (plant AT-rich sequence- and zinc-binding protein 1), was isolated from peas. The amino acid sequence of the protein is similar to those of other uncharacterized proteins predicted from the genome sequences of higher plants. However, no paralogous sequences have been found outside the plant kingdom. Multiple alignments among these paralogous proteins show that several cysteine and histidine residues are invariant, suggesting that these proteins are a novel class of zinc-dependent DNA-binding proteins with two distantly located regions, C-x(2)-H-x(11)-C-x(2)-C-x((4-5))-C-x(2)-C-x((3-7))-H-x(2)-H and C-x(2)-C-x((10-11))-C-x(3)-C. In an electrophoretic mobility shift assay, the zinc chelator 1,10-o-phenanthroline inhibited DNA binding, and two distant zinc-binding regions were required for DNA binding. A protein blot with (65)ZnCl(2) showed that both regions are required for zinc-binding activity. The PLATZ1 protein non-specifically binds to A/T-rich sequences, including the upstream region of the pea GTPase pra2 and plastocyanin petE genes. Expression of the PLATZ1 repressed those of the reporter constructs containing the coding sequence of luciferase gene driven by the cauliflower mosaic virus (CaMV) 35S90 promoter fused to the tandem repeat of the A/T-rich sequences. These results indicate that PLATZ1 is a novel class of plant-specific zinc-dependent DNA-binding protein responsible for A/T-rich sequence-mediated transcriptional repression.

Figure 1

Schematic diagram of one-hybrid screening.

Figure 2

(A) Nucleotide and amino acid sequences of the PLATZ1 gene. Nucleotides are numbered on the right side, and amino acids are numbered on the left side. (B) Phylogenetic tree of PLATZ1 and its paralogous Arabidopsis proteins. (C) Amino acid sequence alignments of PLATZ1 and its paralogous Arabidopsis proteins. Several Arabidopsis proteins were predicted from genome sequences. The left column shows the accession number of the genome sequence. Conserved cysteine and histidine residues are shown in gray. Asterisks indicate positions which have a single, fully conserved residue. Colons indicate positions that have similar residues.

Figure 3

DNA-binding activity of the PLATZ1 protein. (A) Nucleotide sequence of the upstream region (–734 to –642) of the pea pra2 gene. The DE1 element is boxed. Arrows indicate the 68-bp DNA probe used in the EMSA experiment. The 31-bp DE1 probe is underlined. (B) The recombinant PLATZ1 protein binds to the 68-bp DNA fragment. Binding reactions were carried out with the recombinant PLATZ1 protein and the ³²P-labeled 68-bp DNA fragment (2.5 fmol) in the absence or presence of the unlabeled 68-bp DNA fragment as competitor. Lane 1, no protein; lane 2, 100 ng PLATZ1 protein; lanes 3–6, 200 ng PLATZ1 protein. As indicated, 100, 200 and 400-fold competitors were added (lanes 4, 5 and 6). Arrows indicate the free probe and the protein-bound complex. (C) Nucleotide sequences of various competitors used to define the PLATZ1-binding site. The positions of the mutated nucleotides are underlined. (D) Competition experiments using the 68-bp WT competitor and LS-mutated competitors. Binding reactions were carried out with 200 ng of the PLATZ1 protein and the ³²P-labeled 68-bp DNA fragment (2.5 fmol) in the absence or presence of the unlabeled competitor DNA. A 400-fold competitor was added in each reaction. The control does not contain added protein or competitor DNA.

Figure 4

The PLATZ1 protein binds to the enhancer element of the pea plastocyanin gene (petE). (A) The sequence similarity between the 30-bp A/T-rich sequence of the pra2 gene and the 31-bp A/T-rich enhancer of the pea petE gene. (B) EMSA using the 31-bp A/T-rich enhancer of the pea petE gene. Binding reactions were carried out with the recombinant PLATZ1 protein and the ³²P-labeled 31-bp A/T-rich enhancer fragment (100 fmol) in the absence or presence of the unlabeled 31-bp A/T-rich enhancer fragment as competitor. Lane 1, no protein; lane 2, 100 ng PLATZ1 protein; lanes 3–5, 200 ng PLATZ1 protein. As indicated, 100 and 200-fold competitors were added (lanes 4 and 5). An arrow indicates a protein-bound complex. (C) Competition experiments using oligonucleotides poly(dA–dT)–poly(dA–dT) and the labeled enhancer fragment of the pea petE gene. Lane 1, no protein; lane 2, 100 ng PLATZ1 protein; lanes 3–5, 200 ng PLATZ1 protein. As indicated, 50 ng and 100 ng of competitors were added (lanes 4 and 5). An arrow indicates a protein-bound complex.

Figure 5

Inhibition of the DNA-binding activity of the PLATZ1 protein by the zinc chelator 1,10-o-phenanthroline. Binding reactions were carried out with the PLATZ1 protein (200 ng) and the ³²P-labeled 68-bp DNA fragment (2.5 fmol) in the presence of 1,10-o-phenanthroline (1.25–5 mM). Lane 1, no protein; lane 2, no 1,10-o-phenanthroline; lanes 3–5, 1.25, 2.5 and 5.0 mM 1,10-o-phenanthroline, respectively. Arrows indicate the free probe and the protein-bound complex.

Figure 6

Zinc-binding activity of the PLATZ1 protein. The PLATZ1 protein (1 µg) was separated by SDS–PAGE and transferred to PVDF filters. The filters were probed with ⁶⁵ZnCl₂ in the metal-binding buffer. In competition experiments, divalent metal ions (ZnCl₂, CuCl_2, FeCl₂, CoCl₂ and MnCl₂) were included in the buffer for all steps. Arrowheads indicate the position of the PLATZ1 protein.

Figure 7

Two distantly located zinc-binding motifs are required both for zinc binding and DNA binding. (A) Schematic representation of three recombinant proteins. (B) Zinc-binding activity of the PLATZ1 protein and its truncated versions, PLATZ1-N and PLATZ1-C. Each protein (∼1 µg) was separated by SDS–PAGE and transferred to PVDF filters. One filter was stained with Coomassie Blue (right panel), and the other was probed with ⁶⁵ZnCl₂ (left panel). Arrowheads indicate the position of each protein. M, molecular weight markers. (C) DNA-binding activity of the N-terminal (PLATZ1-N) and central regions (PLATZ1-C) of PLATZ1 proteins. The DNA-binding activity of each protein was compared with the binding activity of the entire region of PLATZ1 (PLATZ1). Binding reactions were carried out with different amounts (100 and 200 ng) of each fusion protein and the ³²P-labeled 68-bp DNA fragment (2.5 fmol). The control contains no added protein.

Figure 8

PLATZ1 represses gene expression in a binding site-dependent manner. (A) Schematic diagrams of the PLATZ1 expression constructs and LUC reporter constructs used in the transient assay. NOS, nopaline synthase terminator. (B) The indicated reporter and expression constructs were co-bombarded into the growing region of etiolated pea epicotyls. Twelve hours after bombardment, LUC and GUS activities were measured. LUC activity was normalized with respect to GUS activity for each bombardment experiment, and the average of p35S90 bombarded with p35S was taken as 100. Values are the means of at least seven independently bombarded samples, with error bars representing S.E.

Figure 9

RNA gel blotting of PLATZ1 mRNA in various organs. Roots, root tips (terminal 5 mm) and terminal buds were collected from 7-day-old pea seedlings. Mature and young leaflets were collected from the third and fifth nodes of 7-day-old plants, respectively. Elongated and elongating stems were collected from the second and fourth internodes of 10-day-old seedlings.