Identification and characterization of alternative promoters, transcripts and protein isoforms of zebrafish R2 gene

Abstract

Ribonucleotide reductase (RNR) is the rate-limiting enzyme in the de novo synthesis of deoxyribonucleoside triphosphates. Expression of RNR subunits is closely associated with DNA replication and repair. Mammalian RNR M2 subunit (R2) functions exclusively in DNA replication of normal cells due to its S phase-specific expression and late mitotic degradation. Herein, we demonstrate the control of R2 expression through alternative promoters, splicing and polyadenylation sites in zebrafish. Three functional R2 promoters were identified to generate six transcript variants with distinct 5' termini. The proximal promoter contains a conserved E2F binding site and two CCAAT boxes, which are crucial for the transcription of R2 gene during cell cycle. Activity of the distal promoter can be induced by DNA damage to generate four transcript variants through alternative splicing. In addition, two novel splice variants were found to encode distinct N-truncated R2 isoforms containing residues for enzymatic activity but no KEN box essential for its proteolysis. These two N-truncated R2 isoforms remained in the cytoplasm and were able to interact with RNR M1 subunit (R1). Thus, our results suggest that multilayered mechanisms control the differential expression and function of zebrafish R2 gene during cell cycle and under genotoxic stress.

Figure 1. 5′-flanking region of zebrafish R2 locus.

Nucleotides are numbered with the first nucleotide of the proximal ATG designated as +1 (indicated by solid triangle). Two putative translational initiation sites (ATGs) are shadowed. Potential binding sites for a variety of transcription factors are underlined or overlined. Three alternative transcriptional start sites (TSS1, TSS2 and TSS3) are indicated by rightwards arrows. Exons (E-1a, E-1b, E1a, E1b and E1c) are boxed.

Figure 2. Genomic organization and transcript variants of zebrafish R2 gene.

(A) Comparative analysis of R2 gene organization between human and zebrafish. Exons (E-1 to E10) are numbered and indicated by boxes. Solid boxes indicate the R2 coding region, whereas open boxes represent the 5′/3′-untranslated regions. Introns and other 5′-flanking regions are indicated by solid lines. Positions of primers used for RT-PCR are named and indicated by arrows. Alternative polyadenylation sites in exon 10 are shown as pAS1 and pAS2. (B–D) RT-PCR analysis of zebrafish R2 transcript variants. M: DNA size markers. (E) Schematic representation of zebrafish R2 transcript variants. Three distinct transcript variants named R2_v1, R2_v2 and R2_v3 are generated through alternative promoter usage. Alternative splicing of R2_v3 transcripts results in four transcript variants R2_v3a, R2_v3b, R2_v3c and R2_v3d. All six transcript variants contain E3 to E10. The three R2 forms are referred to as R2, Δ29R2 and Δ52R2.

Figure 3. Analysis of the transcriptional regulatory region of the zebrafish R2 locus.

(A) Genomic structure of the 5′ control region of R2 gene. The proximal translational initiation site (ATG) is designated as +1. Positions of two ATGs and three alternative transcriptional start sites (TSS1, TSS2 and TSS3) are shown. (B–D) Relative luciferase (LUC) activities (firefly/Renilla) of deletion constructs from three R2 promoters in zebrafish embryos and HeLa cells. Negative control pGL3-Basic, positive control pGL3-Promoter and promoter deletion constructs containing different lengths of the 5′-flanking region of R2 gene are listed in the left panel. Relative luciferase activities (firefly/Renilla) of corresponding constructs are presented in the right panel. Histograms represent means ± SD of three independent experiments.

Figure 4. Spatiotemporal expression patterns of R2 transcript variants in zebrafish.

Expression levels of R2 transcripts were detected by qPCR. (A) Expression of R2 transcripts during embryogenesis. Total RNA was isolated from thirty embryos at indicated stages. (B) Distribution of R2 transcripts in adult tissues examined. Total RNA was isolated from indicated tissues of two adult zebrafish. Expression levels of R2 transcripts were normalized to 18S rRNA expression, and the vertical bars represent the mean ± SD of three independent experiments.

Figure 5. Functional analysis of E2F site and CCAAT box in the proximal promoter of zebrafish R2 gene.

(A) Comparative analysis of 230-bp nucleotides upstream of the proximal TSS (indicated by vertical box) from human, chicken, frog and zebrafish. Predicated E2F binding site, CCAAT and TATA boxes are shown as ovals, arrows and rectangles, respectively. Accession numbers of these sequences were listed in Table S2. (B) Effects of mutations in the E2F binding site and CCAAT box on strength of proximal promoter in zebrafish R2 gene. Wild type (WT) and mutated promoters are indicated in the left panel and relative luciferase (LUC) activities of corresponding constructs are expressed as a percentage of wild type promoter activity in the right panel. ** and * indicate p<0.01 and p<0.05, respectively. (C) Effects of an E2F mutation on cell cycle-specific activation of zebrafish R2 promoter. HepG2 cells were transfected with wild type or mE2F reporter constructs plus reference vector pRL-SV40, synchronized by serum-starvation for 48 h and then stimulated by adding fresh DMEM with 20% FBS. Cells were harvested for luciferase assays at indicated time points. Values are expressed as fold induction compared with the relative luciferase activity (firefly/Renilla) at 0 h. Data represent mean ± SD from three independent experiments.

Figure 6. DNA damage-induced expression of zebrafish R2 gene.

(A) Induced expression of R2 transcripts in CPT-treated embryos. Embryos at 24 hpf were treated with 2000 or 4000 nM CPT for 6 h and total RNA was isolated for real-time PCR. The relative expression levels of R2 transcripts are normalized to β-actin expression. Values are showed as the fold induction compared with untreated samples. (B) The activity of promoter P3 for zebrafish R2 gene was induced by CPT. The P3 pGL-(-5149/-954) reporter construct was co-injected with pRL-SV40 into one-cell stage embryos and injected embryos at 24 hpf were treated with indicated concentration of CPT for another 24 h. Then, luciferase (LUC) assays were performed with embryo lysates. Values are expressed as fold induction compared with untreated samples. All data represent means ± SD of three independent experiments.

Figure 7. Multiple sequence alignments of RNR small subunits from nine species.

Gaps were introduced to maximize the alignment. Amino acids with a similarity of more than 80% are shadowed. The KEN boxes are indicated by rectangle. The N-termini of three zebrafish R2 isoforms is indicated by rounded rectangle. Residues crucial for enzyme activity are boxed and indicated by different symbols: • =  iron ligands; ▴ =  tyrosyl radical; Δ = hydrophobic pocket; ♦ = electron transport; ▪ =  R1 binding heptapeptide. Accession numbers of these sequences were listed in Table S2.

Figure 8. Subcellular localization of zebrafish R2 isoforms.

(A) Immunofluorescence staining was performed to detect the subcellular distribution of zebrafish R2 isoforms in transfected Hela cells. Three R2 isoforms were tagged with a Flag at N-terminus. At 36 h after transfection, R2 isoforms were detected with primary anti-Flag antibody and FITC-conjugated secondary antibody. Nuclei were stained with DAPI. (B) Co-localization of fluorescent protein-tagged zebrafish R1 and R2 isoforms in the cytosol of Hela cells. R2 isoforms or R1 of zebrafish were fused with GFP or RFP to their C-termini. At 36 h after transfection, images were directly acquired under fluorescence microscopy.

Figure 9. Physical interaction of zebrafish R2 isoforms with R1.

293T cells were transfected with vectors expressing HA-R1 and one of Flag-R2 isoforms. At 45 h after transfection, protein extracts were subjected to immunoprecipitation (IP) with anti-HA or anti-Flag antibodies. IP materials and total cell lysates (TCL) were detected by Western blotting (WB) analysis. Data for R1 are representative of three IP samples.