Cell cycle regulation of the tobacco ribonucleotide reductase small subunit gene is mediated by E2F-like elements

Abstract

Ribonucleotide reductase (RNR) is a key enzyme involved in the DNA synthesis pathway. The RNR-encoded genes are cell cycle regulated and specifically expressed in S phase. The promoter of the RNR2 gene encoding for the small subunit was isolated from tobacco. Both in vivo and in vitro studies of the DNA-protein interactions in synchronized BY2 tobacco cells showed that two E2F-like motifs were involved in multiple specific complexes, some of which displayed cell cycle-regulated binding activities. Moreover, these two elements could specifically interact with a purified tobacco E2F protein. Involvement of the E2F elements in regulating the RNR2 promoter was checked by functional analyses in synchronized transgenic BY2 cells transformed with various RNR2 promoter constructs fused to the luciferase reporter gene. The two E2F elements were involved in upregulation of the promoter at the G1/S transition and mutation of both elements prevented any significant induction of the RNR promoter. In addition, one of the E2F elements sharing homology with the animal E2F/cell cycle-dependent element motif behaved like a repressor when outside of the S phase. These data provide evidence that E2F elements play a crucial role in cell cycle regulation of gene transcription in plants.

Figure 1.

Nucleotide Sequence of the Tobacco RNR2 Promoter. The arrows indicate the position of the primers used in inverse PCR to amplify the sequence of the RNR2 promoter (a1/a2, a1'/a2' and b1/b2, b1'/b2') and primers used to subclone the RNR2 promoter (P1 and P2). Putative cis elements are in boldface characters, and the overlapping E2Fb/CDE motifs are indicated by thin lines.

Figure 2.

In Vivo Footprinting Analysis of the Tobacco RNR2 Promoter during the Cell Cycle. (A) Tobacco BY2 cells were synchronized by treatment with aphidicolin, and cell cycle progression was monitored by measuring DNA synthesis (lozenges: cpm × 10²/μg protein) and mitotic index (crosses). Cells were harvested in G2, M, G1, or S phases (arrowheads) of the cell cycle. (B) Cells harvested in G2, M, G1, or S phases of the cell cycle were treated in vivo with DMS, and LMPCRs were performed as described in Methods. Control LMPCRs were performed on in vitro DMS-treated genomic DNA (lane C). Protected and hyperreactive residues are indicated by open or filled symbols, respectively. Signals present in all steps of the cell cycle are represented by circles, in G2 by squares, in G2 and S by stars, in G1 and S by triangles, and in S by diamonds. Putative cis elements are indicated. (C) Protected and hyperreactive G residues were reported on the RNR2 promoter sequence. Putative cis elements are in boldface characters. The primers used in LMPCR are indicated by arrows. Symbols are the same as those used in (B).

Figure 3.

Specific Binding of Nuclear Proteins to E2F-like Elements of the Tobacco RNR2 Promoter. (A) DNA fragments used in EMSAs (only the upper strand sequences are presented). (B) EMSA of nuclear extracts from mid-log-phase BY2 cells with the ³²P-labeled E2Fa and E2Fb DNA probes: free probes (lanes 1); complexes with 6 μg of nuclear extract (lanes 2); competitions with a 100-fold molar excess of competitions with a 100-fold molar excess of the unlabelled probes (lanes 3), of the E2F site-mutated oligonucleotides (E2FMU, lanes 4), of an unrelated oligonucleotide (OCTA; lanes 5), and of the CHR site-mutated oligonucleotide (CHRMU, lane 6). Nuclear complexes are indicated by arrows. A nonspecific complex is indicated (ns).

Figure 4.

Involvement of E2F Factor in Nuclear Complexes Binding to E2F-like Sites of the Tobacco RNR2 Promoter. (A) Protein gel blots incubated with the antibody directed against the DNA binding domain of human E2F5 factor: 6 μg of nuclear extract from mid-log-phase BY2 cells (lane 1); 15 ng of the purified tobacco E2F protein (lanes 2 and 3); competition with 0.1 μM purified human E2F5 protein (lane 3). Silver-staining of the purified tobacco E2F protein (lane 4). (B) EMSA of nuclear extracts from mid-log-phase BY2 cells with the ³²P-labeled DNA probes E2Fa and E2Fb: free probes (lanes 4); complexes with 6 μg of nuclear extract (lanes 1); competitions with the antibody directed against the DNA binding domain of human E2F5 factor (lanes 2); and competitions with an unrelated antibody directed against α-tubulin (lanes 3). Specific E2F complexes are indicated by arrows. (C) EMSA of the purified tobacco E2F protein with the ³²P-labeled DNA probes E2Fa and E2Fb: free probes (lanes 1); complexes with 300 ng of tobacco E2F protein (lanes 2); competitions with a 200-fold molar excess of the unlabeled probes (lanes 3) or of the E2F site-mutated oligonucleotides (lanes 4); competition with the antibody directed against the DNA binding domain of human E2F5 factor (lanes 5). Specific E2F complexes are indicated by arrows.

Figure 5.

Fluctuations in Binding Activities of E2F-like Elements during Cell Cycle Progression. (A) Tobacco BY2 cells were synchronized by treatment with aphidicolin, and cell cycle progression was monitored by measuring DNA synthesis (diamonds: cpm × 10²/μg protein) and mitotic index (crosses). Cells were harvested in G2, M, G1, or S phases (arrowheads) of the cell cycle. (B) EMSA of G2, M, G1, and S phase nuclear extracts with the ³²P-labeled DNA probes E2Fa and E2Fb. Lanes C are free probes. Specific complexes are indicated by arrows.

Figure 6.

Role of E2F Elements in the Tobacco RNR2 Promoter Activity. Various RNR2 promoter constructs were fused to the LUC reporter gene: wild-type promoter (WT, −531 to +19), promoters with E2Fa or E2Fb mutated sites (E2FaMU, E2FbMU), promoter with both mutated E2F sites (E2Fa + bMU), and the minimal RNR2 promoter (P. min, −100 to +19). LUC activities were measured in transgenic BY2 cell lines (population of 500 to 1000 independent calli) containing the promoter constructs, as described in Methods. (A) LUC activities in mid-log-phase cells. Error bars indicate sd. (B) Transgenic cell lines synchronized by treatment with aphidicolin and monitored for cell cycle progression by measurements of DNA synthesis (diamonds: cpm × 10²/μg protein) and mitotic index (crosses). Cell cycle progression was similar for all the transgenic lines. A representative experiment is shown. (C) LUC activities of cells harvested at different points of the cell cycle.