Conserved overlapping gene arrangement, restricted expression, and biochemical activities of DNA polymerase ν (POLN)

Abstract

DNA polymerase ν (POLN) is one of 16 DNA polymerases encoded in vertebrate genomes. It is important to determine its gene expression patterns, biological roles, and biochemical activities. By quantitative analysis of mRNA expression, we found that POLN from the zebrafish Danio rerio is expressed predominantly in testis. POLN is not detectably expressed in zebrafish embryos or in mouse embryonic stem cells. Consistent with this, injection of POLN-specific morpholino antisense oligonucleotides did not interfere with zebrafish embryonic development. Analysis of transcripts revealed that vertebrate POLN has an unusual gene expression arrangement, sharing a first exon with HAUS3, the gene encoding augmin-like complex subunit 3. HAUS3 is broadly expressed in embryonic and adult tissues, in contrast to POLN. Differential expression of POLN and HAUS3 appears to arise by alternate splicing of transcripts in mammalian cells and zebrafish. When POLN was ectopically overexpressed in human cells, it specifically coimmunoprecipitated with the homologous recombination factors BRCA1 and FANCJ, but not with previously suggested interaction partners (HELQ and members of the Fanconi anemia core complex). Purified zebrafish POLN protein is capable of thymine glycol bypass and strand displacement, with activity dependent on a basic amino acid residue known to stabilize the primer-template. These properties are conserved with the human enzyme. Although the physiological function of pol ν remains to be clarified, this study uncovers distinctive aspects of its expression control and evolutionarily conserved properties of this DNA polymerase.

Keywords: BRCA1; DNA damage; DNA polymerase; alternative splicing; zebrafish.

FIGURE 1.

Conserved features of DNA polymerase ν (POLN). A, domains of human POLN, POLQ, and HELQ. Defined motifs are shown by vertical black stripes. B, sequence alignment of POLN-N, exoIII, motif 3, and motif 4 of POLN from three fish (zebrafish, maylandia, and tilapia), two birds (falcon and sparrow), four mammals (mouse, elephant, monkey, and human), and prokaryotic A-family DNA polymerases, E. coli DNA polymerase I (EcpolI) and Rhodococcus erythropolis DNA polymerase I (Rhodococcus). Residues are colored in similarity groups as follows: {K, R, H}, {D, E}, {I, L, V, M}, {F, Y, W}, {Q, N}, {G, A}, {S, T}, {P}, and {C}. Perfectly conserved residues are denoted by *, and highly or relatively conserved residues are denoted by colons and periods, respectively. The open arrowheads show the residues Asp-902 and Arg-957 of zebrafish POLN substituted in this study. The closed arrowhead shows an Asp residue essential for 3′–5′-exonuclease (exo) activity, which is absent in POLN. The sequence alignment was carried out using the Clustal X program.

FIGURE 2.

Syntenic relationship between POLN and HAUS3 genes in zebrafish, mouse, and human genomes. A, sequence alignment of zebrafish POLN and HAUS3 cDNA clones. Orange, blue, and gray boxes indicate the overlapping first exon of POLN and HAUS3, the second exon of POLN, and the second exon of HAUS3, respectively. The ATG start codon (bold red letters) resides within a well matched Kozak consensus sequence for translation initiation. DQ630550 is the GenBank^TM accession number for the sequence as determined by our experiments for DrPOLN; EB929239 is the GenBank^TM accession number for a cDNA clone encoding a partial DrPOLN. For HAUS3, BC124280 is the GenBank^TM accession number for cDNA encoding full-length DrHAUS3, and NM_001077171 is the NCBI reference sequence. B, zebrafish POLN and HAUS3 genes are mapped on chromosome 7. Introns are represented as black lines. Orange, gray, and blue boxes denote the shared first exon, exons of HAUS3, and the second exon of POLN, respectively. Exons/introns are drawn to scale; each length (in bp) is shown. Circled numbers show the number of each exon. The first exon of POLN is also the first exon of HAUS3 in mouse and human. Single major peaks of H3K27Ac and H3K4Me3, and a single CpG island are present near the first exon of human POLN and HAUS3 but not before the second exon of POLN encoding the start codon. Data accessed from genome.ucsc.edu was derived from the ENCODE project.

FIGURE 3.

A, sequence alignment of 5′-RACE products for the human POLN cDNA. The newly identified 5′-untranslated region is shown within the box. A well matched Kozak consensus sequence for translation initiation surrounds the ATG start codon. B, sequence alignment of the first exon of human POLN (sequence of RACE-PCR clone B7 shown in A) and HAUS3 mRNA (NCBI reference sequence, NM_024511.5), and mouse Poln (AY135562) and Haus3 (NM_146159). Perfectly aligned sequences are boxed. The precise transcription initiation site is not known for POLN or HAUS3 and could extend slightly 5′ of that shown for both mouse and human mRNAs.

FIGURE 4.

HAUS3 but not POLN expression can be detected by in situ hybridization in zebrafish. A, real time PCR analysis during zebrafish development. The y axis indicates transcripts per 40 ng of total RNA isolated from pre-mid-blastula transition (Pre-MBT) stage, 10-somite stage, 24-h post-fertilization (hpf) stage, and 48, 72, and 96-hpf, 5-day post-fertilization stage, brain, liver, testis, and oocyte (B). POLN expression in D. rerio is shown. Northern blot analysis in different adult tissues is shown. The same membrane was hybridized, stripped, and rehybridized sequentially with D. rerio POLN, HAUS3, or β-actin probes. C, embryonic expression patterns of zebrafish POLN and HAUS3 mRNA. Panels a–h, HAUS3 in situ hybridization; panels a, b, e, and g are with the antisense probe, and panels c, d, f, and h are with the sense probe. Panels i–p, POLN in situ hybridization. Panels i, j, m, and o are with the antisense probe, and panels k, l, n, and p are with the sense probe. Stages: panels a, c, i, and k are 10 somites; panels b, d, j, and l are 18 somites; panels e, f, m, and n are 24 hpf; panels g, h, o, and p are 48 hpf.

FIGURE 5.

Independent expression of POLN and HAUS3 and preferential expression of POLN in testis. A, RT-PCR analysis of POLN and HAUS3 in 833K, human testis, and 293T cDNA. PCR was performed with primers in exon 1 of POLN and HAUS3 (F1), ORF of HAUS3 (F3 and B1), and ORF of POLN (F2, F4, and B2). Positions of the primers are diagramed above the gel picture. β-Actin was used as a control. Expected product size (bp) are F1 + B2, 1474; F2 + B2, 1384; F4 + B2, 203; F1 + B1, 999; F3 + B1, 736; β-actin, 353. B, real time PCR analysis in human cultured cells and human testis. The y axis indicates the absolute quantity of transcripts for POLN, POLQ, and HAUS3 per 40 ng of total RNA isolated from 1618K, 833K, RKO, SUSA, TERA1, 293T, and testis. C, RT-PCR analysis of Poln and Haus3 in mouse testis and R1-ES cDNA. PCR was performed with primers in exon 1 of Poln and Haus3 (1), ORF of Haus3 (2 and 3), and ORF of Poln (4 and 5). Positions of the primers are diagramed above the gel picture. D, Rev3L and PolL were used as controls. Expected product size (bp) are 1 + 3, 1867; 2 + 3, 1713; 1 + 5, 2628; 4 + 5, 2595; Rev3L, 350 and 478; PolL, 1720. PCR products were separated on a 1% agarose gel and visualized with EZ-Vision DNA Dye (A) or ethidium bromide (C and D).

FIGURE 6.

DNA polymerase activity of zebrafish POLN (DrPOLN). A, constructs containing residues 276–1146 of DrPOLN were bacterially expressed and purified. Substituted residues in DrPOLN derivatives (Asp-902 and Arg-957) are shown in Fig. 1. Three hundred ng of purified DrPOLN derivatives and molecular mass markers were separated by electrophoresis in a 4–15% SDS-polyacrylamide gradient gel and stained with colloidal Coomassie Brilliant Blue G-250. B, DNA polymerase activities of DrPOLN. 23 nm DrPOLN and DrPOLN (D902A) and 10 pm RB69 gp43 were incubated with the 5′-³²P-labeled primer-templates indicated under “Experimental Procedures” in the presence of all four dNTPs at 37 °C for the indicated time. The activities were analyzed on the same gel.

FIGURE 7.

Evolutionarily conserved residue is important for translesion synthesis activity in DrPOLN. A, 23 nm DrPOLN and DrPOLN (R957A), denoted as WT and R957A, respectively, were incubated with the 5′-³²P-labeled primer-templates indicated above the panel; DNA synthesis on a DNA template containing an undamaged thymine (lanes 1–16) or a 5S-Tg (lanes 17–32) from the 14-mer primer (lanes 1–8 and 17–24) or the 15-mer primer (lanes 9–16 and 25–32) is shown. All reaction mixtures contained substrate at 100 nm in the presence of all four deoxynucleotide triphosphates. Incubation time of each reaction is shown at bottom. Locations of unreacted end-labeled primer (N₀), each template base position (from N₁ to N₁₆), full-length product (N₁₆ for the 14-mer primer and N₁₅ for the 15-mer primer), and positions of 5S-Tg are shown as Tg.

FIGURE 8.

Nucleotide selectivities of DrPOLN and HsPOLN derivatives. A, 23 nm DrPOLN and DrPOLN (R957A), denoted as wild-type and R957A, respectively, were incubated with 300 fmol of 5′-³²P-labeled 16-mer primer annealed to a 30-mer DNA template in the presence of four or one of the indicated dNTPs (100 μm) for 10 min. The first template base denoted by X was G or T. Template sequences are indicated above the panel. NE indicates no enzyme. B, as described for A, using HsPOLN and HsPOLN (K679A), denoted as wild-type and K679A, respectively. C, 23 nm DrPOLN and DrPOLN (R957A), denoted as wild-type and R957A, respectively, were incubated with 5′-³²P-labeled 16-mer primer annealed to a 30-mer DNA template, in which the first template base was G in the presence of all four dNTPs or dTTP (100 μm) for 10 min in indicated pH conditions. D, as described for C, using HsPOLN and HsPOLN (K679A), denoted as wild-type and K679A, respectively. The percentage (%) of the product extension from the primer is shown below each lane in C and D.

FIGURE 9.

POLN is associated with BRCA1, FANCJ, and other homologous recombination components. A, POLN-associated proteins were immunopurified from nuclear extracts prepared from HeLa S3 cells expressing FLAG-HA epitope-tagged POLN. The complex was sequentially purified with anti-FLAG and anti-HA antibodies, resolved by SDS-PAGE on a 4–20% gradient gel, and visualized by silver staining. Approximate migration positions of proteins identified in gel sections are shown. B, V5-tagged POLN and FLAG-HA epitope-tagged BRCA1 were transiently co-expressed in 293T cells. The whole-cell extracts prepared from the transfected cells were sonicated and incubated in the presence of benzonase. FLAG-HA epitope-tagged BRCA1 was immunoprecipitated (IP) with anti-FLAG and anti-HA antibodies. V5-tagged POLN in the immunoprecipitated samples was detected with anti-V5 antibody. C, interaction between V5-tagged POLN and FLAG-HA epitope-tagged FANCJ was examined similarly as described in B.

FIGURE 10.

A, POLN complex was immunopurified from nuclear extract prepared from HeLa S3 cells expressing FLAG-HA epitope-tagged POLN. Immunoblotting with specific antibodies confirmed the presence or absence of the proteins in the POLN complex. A single membrane was cut into sections as shown in supplemental Fig. S1 for immunoblotting with different antibodies. One of the membrane sections was first used for HELQ immunoblotting and then stripped using Thermo Restore Western Blot Stripping Buffer and reblotted for identification of FANCD2 and then again for POLN. B, POLN does not interact with the Fanconi core complex in the presence or absence of HU. The FANCL complex was immunopurified from nuclear extracts prepared from HeLa S3 cells expressing FLAG-HA epitope-tagged FANCL with or without 3 mm HU treatment for 24 h. The complex was sequentially purified with anti-FLAG and anti-HA antibodies. The complex was resolved by SDS-PAGE on a 4–20% gradient gel. Immunoblotting with specific antibodies confirmed the presence or absence of the proteins in the FANCL complex. FANCA but not POLN is present in the complex before and after HU treatment. Crude extract prepared from 293T cells transiently expressing POLN (lane labeled POLN) was used as a positive control for the POLN antibody.