Plant DNA polymerase λ, a DNA repair enzyme that functions in plant meristematic and meiotic tissues

Authors


K. Sakaguchi, Department of Applied Biological Science, Faculty of Science and Technology, Tokyo University of Science, 2641 Yamazaki, Noda-shi, Chiba-ken 278-8510, Japan.
Fax: + 81 471 23 9767, Tel.: + 81 471 24 1501 (Ex. 3409),
E-mail: kengo@rs.noda.tus.ac.jp

Abstract

Little is known about the functions of DNA polymerase λ (Pol λ) recently identified in mammals. From the genomic sequence information of rice and Arabidopsis, we found that Pol λ may be the only member of the X-family in higher plants. We have succeeded in isolating the cDNA and recombinant protein of Pol λ in a higher plant, rice (Oryza sativa L. cv. Nipponbare) (OsPol λ). OsPol λ had activities of DNA polymerase, terminal deoxyribonucleotidyl transferase and deoxyribose phosphate lyase, a marker enzyme for base excision repair. It also interacted with rice proliferating cell nuclear antigen (OsPCNA) in a pull-down assay. OsPCNA increased the processivity of OsPol λ. Northern blot analysis showed that the level of OsPol λ expression correlated with cell proliferation in meristematic and meiotic tissues, and was induced by DNA-damaging treatments. These properties suggest that plant Pol λ is a DNA repair enzyme which functions in plant meristematic and meiotic tissues, and that it can substitute for Pol β and terminal deoxyribonucleotidyl transferase.

Abbreviations
Pol

DNA polymerase

PCNA

proliferating cell nuclear antigen

TdT

terminal deoxyribonucleotidyl transferase

dRP lyase

5′-deoxyribose-5-phosphate lyase

GST

glutathione S-transferase

BER

base excision repair

MMS

methylmethanesulfonate

Ten years ago, only five DNA polymerase species (α to ε) were known, and the in vivo roles of each enzyme were thought to have been mostly deciphered. Thanks to the progress of the genome project, however, the story has now greatly changed. To date, at least 15 classes of DNA polymerase have been identified in animals, i.e. Pol α, β, γ, δ, ε, ζ, η, θ, ι, κ, λ, µ, ν, σ and terminal deoxyribonucleotidyl transferase (TdT), and, based on their properties, the polymerases are classified into four families, A, B, X and Y [1]. Although several polymerase species from each of these can be found in all eukaryotes, some are present only in multicellular organisms, and others are distributed phylogenetically [2]. Consequently, studies on their in vivo roles have become exceedingly complicated.

We have concentrated particularly on the X-family enzymes such as Pol β, λ, µ and TdT [3–5], because their distribution in multicellular organisms tends to be limited to specific organs, for example, the nervous system including the brain, or the testis, ovary and thymus [6–8]. Pol β knockdown mutant mammalian cultured cells proved viable [9], but in vivo they were lethal because of central nervous system defects [10]. TdT is associated with the immune system in the thymus [11]. Some of the X-family enzymes are therefore thought to have roles not only in DNA repair and recombination [1], but also in some unknown system during development of multicellular organisms. How new members of the X-family, Pol λ and µ, function is of great interest. In this report, we focus on DNA polymerase λ (Pol λ), a recently identified member of the X-family in mammalian cells [7,12]. Half of the peptide sequence of Pol λ is arthologous to DNA polymerase β (Pol β) from the same mammals. As mammalian Pol λ mRNA is abundantly expressed in testis, like Pol β, it is thought to play a role in DNA repair synthesis coupled with meiotic recombination [7]. Unlike Pol β, however, Pol λ has a BRCT domain in the N-terminal region, which has been proposed to mediate protein–protein interactions involved in DNA repair and cell cycle checkpoint regulation on DNA damage [13]. However, little is known about the actual functions of Pol λ and its differences from Pol β. One of major experimental difficulties is that the distributions and properties of Pol λ and Pol β are similar, and mammalian tissues are too complicated to investigate DNA polymerase knockdown lines [14].

For this reason, we have focused on higher plant materials as a possible alternative. Formerly, the presence of the Arabidopsis ortholog of Pol λ was introduced briefly [7] and we found that higher plants had no Pol β, Pol µ or TdT, from the database of rice (Oryza sativa L. cv. Nipponbare) and Arabidopsis thaliana using the blast algorithm, with the Pol X domain conserved among family X enzymes. Therefore Pol λ may function in DNA repair and recombination which are common requirements of mammals and plants. Moreover, as cells proliferate only in the meristematic tissues in plants, and not in the other mature tissues, the rigid partitioning allows elucidation of mechanisms of DNA repair in the cells in which DNA or does or does not replicate.

We report here the characterization of Pol λ from a higher plant, rice, using cDNA and recombinant proteins. The results suggest that plant DNA Pol λ is a DNA repair enzyme which functions in plant meristematic and meiotic tissues.

Materials and methods

Plant materials

Rice plants (Oryza sativa L. cv. Nipponbare) were grown in a growth cabinet under a 16 h day/8 h night cycle at 28 °C. Suspension-cultured cells were cultured as described previously [15].

Molecular cloning of O. sativa DNA Pol λ (OsPol λ)

The rice database was searched using the blast algorithm. One sequence (GenBank accession number AP004995) was identified that encodes a predicted protein with high similarity to A. thaliana Pol λ (AJ289628). The cDNA of this gene, named OsPol λ, was isolated from rice mRNA using the RNeasy Plant Mini kit (Qiagen), and was amplified with the SuperScript One-Step RT-PCR System (Invitrogen). The nucleotide sequence data reported in this paper are now available in the DDBJ/EMBL/GenBank nucleotide sequence databases with the accession number AB099525. Site-directed mutations were introduced into OsPol λ by a PCR-based method as described previously [16].

Overexpression of OsPol λ protein in Escherichia coli

The OsPol λ protein, which contained a six-histidine C-terminal tag, was overexpressed and purified as follows. The sequence of the OsPol λ coding region was cloned into the pET21a expression vector (Novagen), which was transfected into BL21 (DE3) (Novagen) bacteria in 500 mL Luria–Bertani medium containing 50 µg·mL−1 ampicillin. Cells were grown to an absorbance of 0.8, and isopropyl β-d-thiogalactoside was added to a final concentration of 1 mm. Cells were harvested after 3 h by centrifugation at 3000 g for 10 min. Cell pellets were resuspended in 4 mL ice-cold binding buffer (20 mm Tris/HCl, pH 7.9, 0.5 m NaCl, 5 mm imidazole, 0.1% Nonidet P40) and sonicated with 20 bursts of 10 s each. Cell lysates were centrifuged at 39 000 g for 20 min, and the soluble protein fraction was collected as the crude extract and loaded on to 10 mL His-Bind Resin (Novagen). The column was washed with 50 mL binding buffer and then with 50 mL wash buffer (20 mm Tris/HCl, pH 7.9, 0.5 m NaCl, 60 mm imidazole, 0.1% Nonidet P40), and the bound protein was eluted with 30 mL elution buffer (20 mm Tris/HCl, pH 7.9, 0.5 m NaCl, 1 m imidazole). The eluted protein was dialyzed against buffer A (50 mm Tris/HCl, pH 7.5, 1 mm EDTA, 5 mm 2-mercaptoethanol, 10% glycerol, 0.1% Nonidet P40), and the dialysate was loaded on to a HiTrap Heparin column (Amersham Pharmacia Biotech) equilibrated with buffer A. After being washed, the fraction was collected with 20 mL of a linear gradient of 0–0.5 m NaCl in buffer A. The eluted OsPol λ protein was dialyzed against buffer A and used in the subsequent experiments.

Standard DNA polymerase assay

A DNA polymerase assay was performed as described in a previous report [2]. The assay mixture (20 µL) contained: 50 mm Tris/HCl (pH 7.5) with 1 mm MnCl2, 1 mm dithiothreitol, 40 µm dTTP, 10 µCi·mL−1[3H]TTP, 1 pmol OsPol λ, 20 µg·mL−1 oligo(dT)12−18, 40 µg·mL−1 poly(dA) or poly(rA) and 15% (v/v) glycerol. Incubation was carried out at 37 °C for 30 min, and then the reaction mixtures were spotted on pieces of DE81 filter paper, washed sequentially four times with 5% (w/v) Na2HPO4, twice with distilled water, and then twice with 100% ethanol. After being dried, the radioactivity of each piece was measured in a toluene-based scintillator using a Beckman liquid-scintillation counter. One unit of DNA polymerase activity was defined as the amount of enzyme catalyzing the incorporation of 1 nmol dTTP in 60 min at 37 °C.

3′→5′ exonuclease assay

OsPol λ (1 pmol) was incubated with 100 fmol DNA substrate with 5′-end 32P-labeled primer in 20 µL reaction mixture containing 50 mm Tris/HCl (pH 7.5), 5 mm MgCl2, and 1 mm dithiothreitol at 37 °C for 20 min. The reaction was stopped by the addition of gel loading buffer (95% formamide, 20 mm EDTA 0.02% bromophenol blue and xylene cyanol), and analyzed by electrophoresis in a 10% polyacrylamide gel and visualized by autoradiography. DNA substrate used in this assay is a d15:d21-mer primer/template. The sequences are 5′-ACTGGAGATCTGCAT-3′ and 5′-TGAAGCATGCAGATCTCCAGT-3′.

Misincorporation assay

The four template/primer structures used, which differ only in the first template base, are shown in Fig. 2B. A 1 mm Mn2+-activated nucleotide insertion on each 5′-labelled DNA substrate was analyzed in the presence of either the complementary nucleotide (0.1 µm) or each of the three incorrect dNTPs (100 µm). 5′-End labeling was performed by T4 polynucleotide kinase (Takara Bio Inc.) using [γ-32P]dATP (3000 Ci·mmol−1; Amersham). After incubation for 15 min at 37 °C in the presence of 50 nm OsPol λ, extension of the 5′-labeled strand was analyzed by electrophoresis in a 8 m urea/10% polyacrylamide gel and autoradiography.

Figure 2.

DNA polymerase activity of OsPol λprotein. (A) Multiple amino acid alignment of a portion of the polymerization active site with the other X-family DNA polymerases. White letters over black background represent invariant residues among X-family DNA polymerases. The asterisk indicates changed amino acids. (B) SDS/PAGE analysis of OsPol λ wild-type (lane l) and D400A D402A mutant (lane 2). The concentration of the SDS/polyacrylamide gel was 10%, and staining was with Coomassie Brilliant Blue (CBB). (C) Standard DNA polymerase activities of OsPol λ wild-type and D400A D402A mutants were measured under the conditions described in Materials and methods.

Measurement of processivity

A d27:d52-mer primer/template was assayed in a final volume of 20 µL containing 50 mm Tris/HCl (pH 7.5), 0.2 mg·mL−1 BSA, 1 mm dithiothreitol, 1 mm MnCl2, 5 µm each unlabeled dNTP, 1 pmol OsPol λ, 0.5 pmol 5′-labeled DNA template and 40 pmol unlabeled poly(dA)/oligo(dT)12−18 as the trap. First, the reaction components were mixed in the absence of dNTPs and the trap. Next, the reaction was started by adding the dNTPs and the trap. Mixtures were incubated for 5 min at 37 °C in the presence (5 pmol) or absence of rice proliferating cell nuclear antigen (OsPCNA), and reactions were stopped by the addition of gel loading buffer (95% formamide, 20 mm EDTA 0.02% bromophenol blue and xylene cyanol), and analyzed by electrophoresis in a 10% polyacrylamide gel and visualized by autoradiography. We have checked the effectiveness of the trap. The mixture in the absence of OsPCNA was also incubated for 10 min at 37 °C. In this case, the product size was not increased. OsPCNA was expressed and purified as described previously [17].

Terminal transferase assay

A single-stranded 75-mer oligonucleotide was assayed in a final volume of 20 µL containing 50 mm Tris/HCl (pH 7.5), 0.2 mg·mL−1 BSA, 1 mm dithiothreitol, 0.5 mm MnCl2, 0.2 µm single-stranded 75-mer (3′-OH ends). The sequence of 75-mer oligonucleotide is 5′-AGCTACCATGCCTGCACGAAGAGTGCGTATTATGCCTACACTGGAGTACCGGAGCATCGTCGTGACTGGGAAAAC-3′. [3H]dTTP (10 µm; 10 Ci·mmol−1) and enzymes were added as indicated in the figure legends, and incubated for 30 min at 37 °C. The reaction mixtures were spotted on pieces of DE81 filter paper. After being washed and dried, the radioactivity of each piece was measured in a toluene-based scintillator using a Beckman liquid-scintillation counter.

5′-Deoxyribose-5-phosphate (dRP lyase) activity assay

A 21-mer oligonucleotide containing an AP site at position 10 (5′-CCTGCCCTGAPGCAGCTGTGGG-3′; Trevigen, Gaithersburg, MD, USA) was labeled at the 3′-end with terminal deoxynucleotidyltransferase (Takara) using [α-32P]ddATP (3000 Ci·mmol−1; Amersham) and annealed to its complementary oligonucleotide (Trevigen). This labeled double-stranded substrate (1 pmol) was treated with human AP endonuclease (1 unit; Trevigen) for 1 h at 37 °C in 20 µL buffer containing 10 mm Hepes/KOH, pH 6.5, 100 mm KCl and 10 mm MgCl2, thus generating the substrate for dRP lyase activity (Fig. 3A). Reaction mixtures (20 µL) contained 50 mm Hepes, pH 7.4, 5 mm MgCl2, 2 mm dithiothreitol, and 1 pmol·µL−1 concentration of the labeled substrate as described in the previous section. The reaction was initiated by adding 50 nm (final) OsPol λ or rat Pol β, and incubation was for 15 min at 37 °C. NaBH4 was then added to a final concentration of 340 mm, and the reaction continued for 30 min on ice. Stabilized DNA products were ethanol-precipitated in the presence of 0.1 µg·mL−1 tRNA, resuspended in 10 µL gel loading buffer (95% formamide, 20 mm EDTA, 0.02% bromophenol blue and xylene cyanol), and analyzed by electrophoresis in a 20% polyacrylamide gel and visualized by autoradiography. Recombinant rat Pol β was expressed and purified as described previously [4].

Figure 3.

DNA synthesis catalysed by OsPol λ. (A) OsPol λ preferentially incorporates complementary nucleotides. Four different template/primer structures were used, differing in the first template base. Reactions were carried out in the presence of 50 nm OsPol λ and 1 mm MnCl2 as a source of activating metal ions. Extension of the labeled (*) primer strand in the presence of either the correct (0.1 µm) or the incorrect (100 µm) dNTP was analyzed by 8 m urea and 10% PAGE and autoradiography. (B) TdT activity was measured under the conditions described in Materials and methods. Commercial TdT (Takara) was assayed as a positive control. (C)  In vitro interaction between OsPCNA and OsPol λ. OsPol λ-His was mixed with GST-OsPCNA or GST in NaCl/Pi buffer at room temperature for 1 h, and then with a column of GST–Sepharose-4B beads. The beads were washed and eluted and the amount of protein was quantified by Western blotting. The immunoblot was probed with a polyclonal antibody to His-tag. (D) DNA polymerase activity of OsPol λ on a d27:d52-mer primer/template under processive conditions in the absence (lanes 1 and 2) or presence (lanes 3 and 4) of OsPCNA. Reactions were carried out in the presence of 1 pmol OsPol λ, 0.5 pmol 5′-labeled DNA template, and 40 pmol unlabeled poly(dA)/oligo(dT)12−18 as the trap.

Pull-down assay

His-tagged OsPol λ (1 µg) and glutathione S-transferase (GST)-fused OsPCNA (1 µg) or GST (1 µg) were incubated with GST–Sepharose-4B beads. The beads were then washed with 10 vol. of the manufacturer's buffer, and the amount of protein present in the affinity column was quantified by Western blotting. The immunoblot was probed with polyclonal antibody to His-tag (Medical and Biological Laboratories Co., Ltd).

Other methods

Phylogenetic analysis was performed based on the amino acid sequence by the UPGMA method using genetyx mac var. 10 (Software Development Co. Ltd, http://www.sdc.co.jp/genetyx) . DNA polymerase domains of various X-family DNA polymerases were aligned and used to produce the tree. Northern blotting analyses were performed as described previously [18]. Blots were quantified with an NIH imaging analyzer.

Results

Identification of OsPol λ and comparison with other eukaryotic DNA polymerases of the X-family

As described in the Introduction, there may be only one kind of the X-family polymerase in higher plants, and it is homologous to Pol λ. Molecular cloning of a Pol λ-homologous cDNA (OsPol λ) from rice cells was carried out as described. OsPol λ was found to encode a predicted product of 552 amino acid residues with a molecular mass of 60.9 kDa. OsPol λ protein contains a BRCT domain at the N-terminus and a Pol X domain at the C-terminal region, similar to the overall domain organization of mammalian Pol λ. The nucleotide sequence data reported in this paper have been lodged in the DDBJ nucleotide sequence database with the accession number AB099525. The OsPol λ gene was mapped to chromosome 6 and was shown to contain 14 exons and 13 introns in a total genomic segment of 5.1 kb. We also succeeded in cloning Pol λ cDNA from A. thaliana (AtPol λ) to assess universality. Similar to OsPol λ structurally, AtPol λ encodes a predicted product of 529 amino acid residues with a molecular mass of 59.6 kDa.

The deduced amino acid sequence of OsPol λ was compared with those of other known eukaryotic X-family DNA polymerases and found to show 60.5% sequence identity with AtPol λ, and 29.6% with human Pol λ. As shown Fig. 1A, OsPol λ protein shows a high degree of conservation in the C-terminal DNA polymerase domain (37.3% identity with DNA polymerase domain of human Pol λ), suggesting that the DNA polymerase activity of this enzyme may be important for plant cells. On the other hand, the N-terminal BRCT domain proposed to mediate protein–protein interactions involved in DNA repair and cell cycle checkpoint regulation on DNA damage [13] shows a lower degree of homology to human Pol λ. To determine the phylogenetic relationship between OsPol λ and other X-family DNA polymerases, a tree was drawn based on the amino acid sequences of the DNA polymerase domains (Fig. 1B). For this purpose, distance matrices were generated, using the unweighted pair-group method with arithmetic averages (UPGMA). The Pearson product–moment correlation coefficient was used to estimate the agreement between the original distance matrix and that obtained directly from the dendrogram. Pol λ was confirmed to be highly conserved among the plant and animal kingdoms (Fig. 1B), suggesting that they have independently evolved to occupy a particular functional niche.

Figure 1.

Comparison of predicted amino acid sequences of OsPol λ, human Pol λ, human Pol β(A) and phylogenetic tree constructed by the UPGMA method, based on the amino acid sequences of the DNA polymerase domains of OsPol λ and the other X-family DNA polymerases(B). (A)  OsPol λ has a BRCT (BRCA1 C-terminal) domain at the N-terminal region similar to human Pol λ and has the Pol domain that is conserved among X-family members. (B) Horizontal distances are proportional to evolutionary divergence expressed as substitutions per site.

DNA polymerase activity of OsPol λ protein

The OsPol λ protein was overexpressed in E. coli and purified as described in Materials and Methods. This protein was purified to near-homogeneity as shown by SDS/PAGE analysis (Fig. 2B, lane 1) and was devoid of nuclease contaminants, as tested in nuclease assays (data not shown). The specific activity of the wild-type protein was 7243 U·mg−1 under our standard conditions. To confirm that the activity was associated with OsPol λ protein, we prepared a mutant with two substitute amino acids, Asp400Ala and Asp402Ala. These amino acids are highly conserved among various organisms and therefore must be very important (Fig. 2A). This control protein was purified to near-homogeneity as shown by SDS/PAGE analysis (Fig. 2B, lane 2). In this case, the OsPol λ D400A D402A mutants possess extremely low polymerization activity (Fig. 2C). This indicates that the activities of OsPol λ were intrinsic. The biochemical properties of OsPol λ are shown in Table 1. As expected, it required activation with metal ions, Mn2+ being more efficient than Mg2+. We also investigated the effects of DNA polymerase inhibitors on the DNA polymerase activity of OsPol λ. Strong inhibition was noted with 2,3-dideoxythymidine-5-triphosphate, a known inhibitor of mammalian DNA polymerases β and γ, but the enzyme was not sensitive to aphidicolin and N-ethylmaleimide, both of which are inhibitors of DNA polymerase α, δ, and ε. OsPol λ preferentially utilized poly(dA)/oligo(dT)12−18 as a template. These results are similar to those reported for mammalian Pol λ[19–21].

Table 1. DNA polymerization properties of OsPol λ. ddTTP, 2,3-dideoxythymidine-5-triphosphate. The activity with poly(dA)/oligo(dT) under standard conditions was expressed as 100%. Specific activity of OsPol λ was 0.435 U·(pmol enzyme)−1.
 Activity (%)dTTP incorporated (pmol)
Mn2+ (1 mm)100217
–Mn2+< 1ND
–Mn2+ + Mg2+ (1 mm)3882
–Mn2+ + Mg2+ (5 mm)70152
KCl (100 mm)4189
NaCl (100 mm)3576
Aphidicolin (100 µg·mL−1)102221
N-Ethylmaleimide (20 µm)94204
ddTTP (5 µm)54117
ddTTP (10 µm)1226
Templates
Poly(dA)/oligo(dT)100217
Poly(rA)/oligo(dT)2452
Poly(dA-dT)511
Activated DNA49106

Figure 3A shows the capacity of OsPol λ to catalyze faithful DNA synthesis. Each of the four dNTPs was assayed individually as a substrate for incorporation opposite the four possible template bases, in the presence of Mn2+ ions. OsPol λ preferentially inserted the correct dNTP, but some was slightly misinserted opposite the wrong template bases.

Human Pol λ possesses terminal deoxyribonucleotidyl transferase (TdT) activity [22]. We investigated whether OsPol λ had this activity. As shown in Fig. 3B, OsPol λ has weak TdT activity: 2 pmol OsPol λ incorporated 0.3 pmol dTMP per 30 min. In this assay, commercial TdT (Takara) was assayed as a positive control.

Recently, a direct interaction of human PCNA with human Pol λ was shown which stimulates its processivity [23]. To investigate whether OsPol λ directly interacts with rice PCNA (OsPCNA), a pull-down assay was performed. We expressed GST and OsPCNA-fused GST at the N-terminus in E. coli, and then purified them using GST–Sepharose-4B bead column chromatography. Each protein was mixed with His-tagged OsPol λ and a GST–Sepharose-4B column. The fractions washed and eluted with glutathione were electrophoresed and then stained with an antibody against the His-tag. As shown in Fig. 3C, OsPol λ appeared to bind to GST-fused OsPCNA, but not to GST only. In addition, we investigated whether this interaction affected the activity of OsPol λ. Under processive conditions (Fig. 3D), i.e. in the presence of an excess of cold poly(dA)/oligo(dT)12−18 as a trapping reagent, OsPol λ showed a processivity of one or two nucleotides on a d27:d52-mer primer/template, which was increased to about four nucleotides in the presence of OsPCNA (Fig. 3D). Under standard (distributive) conditions, OsPCNA had no effect on OsPol λ activity (data not shown). These results indicate that, similar to human Pol λ, OsPol λ interacts with OsPCNA, and its processivity is thereby stimulated.

OsPol λ has dRP lyase activity

Recently, human Pol λ has been shown to have dRP lyase activity, so it may participate in base excision repair (BER) similarly to mammalian Pol β[24]. BER, considered to be an essential pathway for the elimination of single damaged bases in DNA [25], is initiated by the removal of the modified base by a DNA glycosylase. The resulting AP site is recognized and incised by an AP endonuclease [26]. Removal of the remaining dRP residue is catalyzed by a dRP lyase activity, and the gap is filled by a DNA polymerase such as Pol β[27]. To test the ability of OsPol λ to remove a dRP group, we used a 21-mer double-stranded oligonucleotide containing an AP site at position 10. As described in Materials and Methods, the AP site-containing strand was 3′-end-labeled with [α-32P]ddATP, annealed to its complementary strand and treated with human AP endonuclease to release a dRP-containing substrate (Fig. 4A). This substrate was incubated in the absence (control) or presence of either rat Pol β or OsPol λ. Figure 4B shows that both enzymes removed the dRP moiety, as detected by the reduction in size of the labeled substrate. This suggests that OsPol λ plays a central role in plant BER, because plants are thought to have no Pol β, Pol γ or Pol ι, which have dRP lyase activity in mammalian cells.

Figure 4.

OsPol λexhibits dRP lyase activity. (A) Schematic representation of a dRP lyase reaction. A 21-mer double-strand oligonucleotide containing an abasic site at position 10 in one strand is treated with human AP endonuclease to release a dRP-containing substrate. This dRP moiety is cleaved by dRP lyase activity. (B) Autoradiogram illustrating OsPol λ dRP lyase activity. Lane 1, no protein control; lane 2 and lane 3, DNA was incubated with OsPol λ and rat Pol β, respectively.

Expression levels of OsPol λ correlated with cell proliferation

To determine the expression pattern of OsPol λ in various organs, Northern blot hybridization was performed (Fig. 5A). Total RNA samples isolated from various organs of 50-day-old and 150-day-old rice plants were blotted and probed with 32P-labeled OsPol λ cDNA. The probe for OsPol λ was a sequence of 700 bp from the C-terminus. Two kinds of transcripts thought to be splicing variants were mostly detected in the region of the shoot apical meristem and panicle (lanes 1 and 5 in Fig. 5A), and showed moderate expression in the roots, root apices, flag leaves and young leaves (lanes 3, 4, 6 and 7). In the mature leaves, no expression was evident (lane 2). The young leaves have meristem to increase the leaf width, while mature leaves have no proliferating cells. The panicle contains meiotic tissues. These results suggest that transcription of OsPol λ reflects the level of cell proliferation and is perhaps related to meiosis.

Figure 5.

Expression of OsPol λ. (A) Northern blot analysis of OsPol λ in different organs of rice. Each lane contained 20 µg total RNA isolated from shoot apical meristem (SAM; lane 1), a mature leaf (lane 2), a young leaf (lane 3), a root (lane 6) or root apex (lane 7) from an ≈ 50-day-old rice plant, and Flag leaf (lane 4) or panicle (lane 5) from about 150-day-old rice plant. The blot was probed with 32P-labeled OsPol λ cDNA (upper panel). Similar amounts of RNA were loaded in each lane as confirmed by ethidium bromide staining (lower panel). (B) Effects of sucrose starvation on the expression levels of OsPol λ. Rice cells were cultured in suspension for 6 days with sucrose (lane 1) or without sucrose (lane 2), and for 10 days without sucrose (lane 3). Lane 4, cells cultured first for 6 days without sucrose, and then for a further 4 days after sucrose had been added to the medium. Aliquots of 20 µg total RNA isolated from the cultured cells were separated on a 1.2% agarose gel containing formaldehyde, and then blotted. The blot was probed with 32P-labeled OsPol λ cDNA. Similar amounts of RNA were loaded in each lane as confirmed by ethidium bromide staining (lower panel). (C) 14-day-old rice plants were exposed to 25 J·m−2 UV-B. After UV-B exposure, plants were kept for 0 h (lane 1), 1 h (lane 2), 2 h (lane 3) or 4 h (lane 4) in the dark. Total amounts of OsPol λ mRNA were determined using an imaging analyzer and expressed relative to the transcripts 0 h after irradiation. (D) 0.25 µm MMS was added to rice suspension cultures. Next, cells were kept for 0 h (lane 1), 0.5 h (lane 2), 1 h (lane 3) or 2 h (lane 4) in the dark. Total amounts of OsPol λ mRNA were determined using an imaging analyzer and expressed relative to the transcripts 0 h after treatment.

OsPol λ was actively transcribed in rice cells in suspension culture (lane 1 in Fig. 5B). When cell proliferation was temporarily halted for 6 or 10 days by removal of sucrose from the growth medium, the level of OsPol λ expression was significantly reduced (lanes 2 and 3). When the growth-halted cells began to re-grow after addition of sucrose to the medium, OsPol λ was again expressed at high levels (lane 4). These results indicated that OsPol λ expression correlates with cell proliferation.

Effects of DNA-damaging treatments on the levels of OsPol λ expression

In addition, we investigated the effects of DNA-damaging treatments, such as UV and methylmethanesulfonate (MMS), on the expression levels of OsPol λ. The 14-day-old rice seedlings were UV-irradiated at 25 J·m−2, followed by incubation for 1–4 h (Fig. 5C). The cultured rice cells were treated with 0.25 µm MMS by adding to the dishes, followed by incubation for 0.5–2 h (Fig. 5D). The levels of OsPol λ expression were increased by both treatments (Figs 5C.D). The induced expressions reached maximum levels at 4 h and 1 h after treatments with UV and MMS, respectively. These observations suggest that the gene product was involved in DNA repair.

Discussion

For the past 10 years, we have focused on elucidating the roles of each of the eukaryotic DNA polymerases [2–5,18,28–31]. In this study, we took advantage of the characterisitcs of higher plants for this purpose and first succeeded in preparing cDNA and the recombinant protein (OsPol λ). To our knowledge, this is the first report to focus on a plant homolog of the X-family polymerases, and Pol λ may be universally present in higher plants as the only member of this family. Assessment of the phylogenetic relationship between OsPol λ and other X-family DNA polymerases showed that Pol λ may have independently evolved to occupy a particular functional niche in the plant and animal kingdoms. Interestingly, Pol λ is more closely related to Pol β than Pol µ and TdT, which have a BRCT domain similar to Pol λ (Fig. 1B), suggesting that Pol λ, µ and TdT are ancestor polymerases descended from a common prototype, and Pol β has diversified in evolution. Why Pol µ and TdT should be absent from higher plants remains unclear. Either they evolved after animals arose, or the genes might have been lost during the evolution to plants.

The recombinant OsPol λ protein showed DNA polymerase and TdT activities (Figs 2C and 3B). It is very interesting that plant Pol λ possesses TdT activity. Vertebrate TdT is very closely related to the immune system in the thymus and is thought to have a special function, for example in DNA repair. The details of this activity of Pol λ remain unclear. Higher plants may be useful in research on TdT activity in DNA metabolism. OsPol λ also possesses dRP lyase activity and interacts with OsPCNA, with stimulation of its processivity. As a misincorporation assay (Fig. 3A), OsPol λ preferentially inserted the correct dNTP, but slightly misinserted. A recent study has revealed that the fidelity of human Pol λ during short gap-filling synthesis was lower than replicative DNA polymerases and Pol β[32]. Like these, most of its biochemical properties appear to be quite similar to those of mammalian Pol λ forms reported previously [19–24]. In the present study, OsPol λ expression correlated with cell proliferation and meiosis. Plant Pol λ may function in mainly plant meristematic and meiotic tissues as one of the DNA repair enzymes. OsPol λ may participate in BER in vivo, because of the dRP lyase activity (Fig. 4) and the induction of OsPol λ by MMS (Fig. 5D). In plants, Pol λ may concurrently possess the functions of Pol β, µ and TdT as the only X-family DNA polymerase.

Pol β and TdT are phylogenetically found only in the deuterostomic branches (or vertebrates) of the animal world, but Pol λ is universally present in both animals and plants. Although the structure and function of Pol β and Pol λ are very similar, a recent study showed that human Pol λ participates in alignment-based gap filling for nonhomologous DNA end joining [33]. Therefore, Pol λ may participate not only in BER but also in DNA double-strand break repair and meiosis, and Pol β and TdT may be evolved for differentiation and development of specific organs in vertebrates such as the nervous and immune systems.

Acknowledgements

This work was supported by a grant from the Ministry of Agriculture, Forestry and Fisheries of Japan (Rice Genome Project IP-5006) and by a grant from the Ministry of Education, Science, Sports and Culture of Japan (Grant-in-Aid for Young Scientists (B), 15770031). This work was also supported by a grant from Futaba Electronics Memorial Foundation (Japan), The Asahi Glass Foundation (Japan), and The Sumitomo Foundation (Japan).

Ancillary