Rdp1, a novel zinc finger protein, regulates the DNA damage response of rhp51(+) from Schizosaccharomyces pombe

Abstract

The Schizosaccharomyces pombe DNA repair gene rhp51(+) encodes a RecA-like protein with the DNA-dependent ATPase activity required for homologous recombination. The level of the rhp51(+) transcript is increased by a variety of DNA-damaging agents. Its promoter has two cis-acting DNA damage-responsive elements (DREs) responsible for DNA damage inducibility. Here we report identification of Rdp1, which regulates rhp51(+) expression through the DRE of rhp51(+). The protein contains a zinc finger and a polyalanine tract similar to ones previously implicated in DNA binding and transactivation or repression, respectively. In vitro footprinting and competitive binding assays indicate that the core consensus sequences (NGG/TTG/A) of DRE are crucial for the binding of Rdp1. Mutations of both DRE1 and DRE2 affected the damage-induced expression of rhp51(+), indicating that both DREs are required for transcriptional activation. In addition, mutations in the DREs significantly reduced survival rates after exposure to DNA-damaging agents, demonstrating that the damage response of rhp51(+) enhances the cellular repair capacity. Surprisingly, haploid cells containing a complete rdp1 deletion could not be recovered, indicating that rdp1(+) is essential for cell viability and implying the existence of other target genes. Furthermore, the DNA damage-dependent expression of rhp51(+) was significantly reduced in checkpoint mutants, raising the possibility that Rdp1 may mediate damage checkpoint-dependent transcription of rhp51(+).

FIG. 1

Screening for DRE_rhp51⁺-binding protein. (A) Schematic representation of the rhp51⁺ promoter. Numbering is relative to the first base in the rhp51⁺ coding sequence. Filled rectangles indicate two decamer DRE sequences (−233 to −224 and −213 to −204), and hatched diamonds indicate two MCBs (−192 to −187 and −183 to −178). (B) Identification of a positive clone by an X-Gal plate assay. Levels of expression of the lacZ reporter gene between the empty vector and putative clones were compared by X-Gal assays. A positive clone, pGAD236, became dark blue, while the parental empty vector (pGAD424) and other putative clones did not show enhanced β-galactosidase activity.

FIG. 2

Protein domains of Rdp1. (A) Alignment of Rdp1 with other transcription factors. The Rdp1 shows a limited but significant homology with RAP1 (S. cerevisiae) and many homeodomain proteins from higher eukaryotes. The amino acid sequences from RAP1 and human HOXA13 (hHox13) were aligned with Rdp1 by the CLUSTAL W program, and output was generated by Genedoc. Black and gray shadows indicate identical and homologous amino acids, respectively. (B) Schematic diagram of the Rdp1 protein domain. The gray rectangle of Rdp1 indicates a region homologous with the activation domain of RAP1 and the polyalanine tract of HOXA13. Percentages indicate sequence identities (and similarities).

FIG. 3

Rdp1 specifically binds to DRE_rhp51⁺ in vitro. (A) EMSA to test the binding specificity and affinity of Rdp1 to DRE_rhp51⁺. A ³²P-labeled DRE oligonucleotide (5 nM), either without (lane 1) or with (lanes 2 to 12) GST-Rdp1, was incubated and electrophoresed as described in Materials and Methods. In lanes 3 to 12, the indicated unlabeled competitor was added. Competitor concentrations were as follows: in lanes 3 and 8, 25 nM; in lanes 4 and 9, 50 nM; in lanes 5 and 10, 100 nM; in lanes 6 and 11, 250 nM; and in lanes 7 and 12, 500 nM. Arrows indicate the DNA-protein complexes. (B). Footprinting of Rdp1 on the upstream regulatory region of rhp51⁺ containing the two DREs. End-labeled DNA fragments containing the two DRE_rhp51⁺s were incubated without (lanes 2 and 3) or with (lanes 4 and 5) GST-Rdp1 protein and subsequently subjected to DNase I digestion as described in Materials and Methods. The region protected from DNase I digestion is indicated by asterisks.

FIG. 4

Determination of the binding consensus sequences by EMSA. (A) Nucleotide sequences of competitors used. Sites changed relative to the sequence of wild-type DRE are indicated by gray boxes. (B) Competition assay. The radiolabeled DRE was incubated with GST-Rdp1 with or without the indicated unlabeled competitor. Two concentrations are shown for each competitor, 100 nM (lanes 3, 5, 7, 9, 11, 13, and 15) and 500 nM (lanes 4, 6, 8, 10, 12, 14, and 16). Lane 1 contains the DNA substrate only, and lane 2 contains the substrate and GST-Rdp1 protein without competitor. Arrows indicate the bound DNA-protein complexes.

FIG. 5

Effect of mutated DREs on rhp51⁺ expression and survival after treatment with UV and MMS. (A) Illustration of the rhp51⁺ gene structure in a host strain harboring wild-type DRE (JAC10) or mutated DRE (JAC20). Mutated bases are indicated by dots under the bases. (B) mRNA levels of rhp51⁺ following UV irradiation or MMS treatment. Exponentially growing cells were exposed to 0.1% MMS or 180 J of UV light per m² and postincubated for 1 h. Total RNAs were extracted, and rhp51⁺ mRNA levels were assessed by Northern blotting. Symbols: C, mock treatment; M, 0.1% MMS treatment; U, 180 J of UV irradiation per m². (C) Relative rhp51⁺ mRNA levels after DNA damage. The data were obtained from five independent experiments and normalized to data with act1⁺. The error bars indicate standard deviations. Symbols: C, mock treatment; M, 0.1% MMS treatment; U, 180 J of UV irradiation per m². (D) Comparison of UV sensitivities. Cells were exposed to UV light at the indicated doses on YES plates, and the surviving colonies were counted after 4 to 5 days. The data points are averages from at least three independent experiments, and the error bars indicate standard deviations. (E) Comparison of MMS sensitivities. The MMS survival test was performed as described in Materials and Methods.

FIG. 6

Disruption of the rdp1⁺ gene and terminal phenotype of an rdp1Δ mutant. (A) Replacement of the rdp1⁺ gene by ura4⁺. The 1.1-kb BalI-BclI fragment was replaced with the 1.8-kb ura4⁺ gene as described in Materials and Methods. (B) Southern blot to confirm rdp1⁺/rdp1::ura4⁺ heterozygotes. The 3.3-, 2.17-, and 1.2-kb fragments were detected in the heterozygote when it was probed with the 3.3-kb EcoRV fragment of rdp1⁺. Lanes: 2 and 6, heterozygotes with rdp1⁺/rdp1::ura4⁺; 1, 3, 4, and 5, wild-type homozygotes. (C) Tetrad analysis of the rdp1⁺/rdp1::ura4⁺ heterozygote. The spores were microdissected onto YES plates and incubated for 4 days at 30°C. Heterozygotic tetrads produced only one or two viable spores with the Ura⁻ phenotype, while most of tetrads from the wild-type diploid showed four viable spores with uracil auxotrophy. (D) Terminal morphology of wild-type and rdp1Δ spores after germination. The rdp1⁺/rdp1⁺ and rdp1⁺/rdp1::ura4⁺ diploid strains were sporulated, and the resulting spores were inoculated into minimal medium supplemented with uracil for the wild-type spores or uracil-free medium for the rdp1::ura4⁺ spores. Germinating cells were stained with DAPI and examined by fluorescence microscopy. Left plate, germinating wild-type spores (22 h); right plate, germinating rdp1Δ spores (22 and 24 h). Scale bar, 10 μm.

FIG. 7

Defects in DNA damage checkpoints cause a significant decrease in transcriptional induction of rhp51⁺ in response to DNA damage. Total RNAs extracted from mid-log-phase cells of checkpoint mutant strains were electrophoresed on formaldehyde-agarose gels and transferred onto nitrocellulose membrane. The RNA blots were hybridized with ³²P-labeled rhp51⁺ or act1⁺ DNA probes and autoradiographed. Symbols: C, mock treatment; M, 0.1% MMS treatment; U, 180 J of UV irradiation per m².