PCNA binding proteins in Drosophila melanogaster: the analysis of a conserved PCNA binding domain

Emma Warbrick 1 Wayne Heatherington 0 1 David P. Lane 1 David M. Glover 1 0 SmithKline Beecham Pharmaceuticals , Mundells, Welwyn Garden City, Herts AL7 1EY, UK 1 CRC Laboratories, Medical Sciences Institute, University of Dundee , Dundee DD1 4HN, UK The eukaryotic polymerase processivity factor, PCNA, interacts with cell cycle regulatory proteins such as p21WAF1/Cip1 and Gadd45, as well as with proteins involved in the mechanics of DNA repair and replication. A conserved PCNA-binding motif is found in a subset of PCNA-interacting proteins, including p21, suggesting that the regulation of these interactions is important for the co-ordination of DNA replication and repair. We have identified several classes of protein which bind to Drosophila PCNA. Two of these proteins contain the consensus PCNA-binding domain: one is the Dacapo protein, a Drosophila homologue of p21WAF1/Cip1, and the second is the transposase encoded by the Pogo DNA transposon. A conserved PCNA-binding domain is also present in a human relative of Pogo, named Tigger, suggesting that this domain has a functional role in this class of transposable element. This raises interesting possibilities for a novel method of transposition in which the transposase might be targeted to replicating DNA. Finally, we have investigated the use of this conserved PCNAbinding domain as a predictor of PCNA-binding capacity. - Proliferating cell nuclear antigen (PCNA) plays an essential role in both the replication and repair of DNA. It forms a toroidal shaped trimer which interacts non-specifically with DNA by encircling it, forming a sliding clamp which tethers the polymerase complex to the DNA duplex. PCNA is highly conserved and has been identified in a very wide range of eukaryotes (1,2). Genes showing sequence homology with PCNA have been identified as far down the evolutionary ladder as the Archaea, whose DNA replication and repair machinery appears to be more typically eukaryotic then prokaryotic (3). The b subunit of Escherichia coli polymerase III is a functional prokaryotic homologue of PCNA and although this protein shows six-fold symmetry like PCNA, it exists as a dimer rather than a trimer (4). PCNA is an essential component of the DNA replication machinery, acting as the processivity factor for polymerases d and e (5,6). In addition to its role in replication, PCNA is also required for nucleotide excision repair (7,8) and plays a role in one pathway of base excision repair (9). Recent analysis of proteins which interact with PCNA have shown that it interacts not only with enzymes involved in the mechanics of DNA repair and replication, but also binds to cell cycle regulatory proteins such as p21 and Gadd45 (1014). This suggests that the interaction of PCNA with other proteins is a key regulatory target for the co-ordination of DNA replication and repair. PCNA in Drosophila is encoded by the mus209 gene: mutations in mus209 result in lethality, although several mus209 mutants are temperature sensitive and provide unique tools to analyse the functions of PCNA in a developing multicellular organism (15). PCNA is required throughout development and maternally encoded PCNA is essential for embryogenesis (15). Analysis of the 5-flanking region of mus209 indicates that its transcription is regulated by homeodomain proteins, suggesting that its expression is coordinated with cell proliferation and differentiation (16). The mus209 mutants are highly sensitive to a range of DNA damaging agents, supporting the role of PCNA in DNA repair. The suppression of position effect variation seen in these mutants points to a role for PCNA in chromatin assembly (15). In a screen for proteins which interact with Drosophila PCNA, we have identified several interacting proteins. Two of them, which we describe here, have not previously been identified as PCNA-binding proteins. One is the Dacapo protein, which is a cyclin-dependent kinase inhibitor (17,18). The second is the transposase encoded by the Pogo transposon, which belongs to the TC1/Mariner superfamily of transposons (19). We find that both proteins contain a consensus PCNA-binding domain shared by other PCNA-binding proteins such as p21, Fen1, XPG and MCMT (12,2023). We find that a human relative of Pogo, named Tigger, also potentially encodes a protein containing a conserved PCNA-binding domain (24). This raises interesting possibilities for a novel method of tranposition of such elements, possibly by targeting the transposase to replicating DNA. MATERIALS AND METHODS Plasmid expression constructs and yeast two-hybrid methods Manipulations of E.coli and DNA were by standard methods (25). Double-stranded plasmid DNA was sequenced using the Sequenase protocol (US Biochemical). The plasmids expressing human, Drosophila and Schizosaccharomyces pombe PCNA have been described (12,14). Growth and maintenance of Saccharomyces cerevisiae was according to standard methods (26). Transformation was carried out by the method of Gietz et al. (27). The S.cerevisiae strain Y190 (MATa leu2-3,112 ura3-52 trp1-901 his3-D200 ade2-101 gal4D gal80D cyhR URA3::GAL1-lacZ LYS2::GAL1-HIS3) was used for all twohybrid analysis, which expresses the reporter genes lacZ (E.coli) and HIS3 (S.cerevisiae) under the control of the GAL1 promoter. All two-hybrid screening and analysis was carried out as previously described (12,20). pACT plasmids identified in the two-hybrid screening experiment were tested against pAS plasmids encoding unrelated Gal4 fusion proteins to exclude false positive results: these included pAS-Snf1, pAS-p53, pAS-Cdk2 and pAS-lamin (20). The following 20 amino acid peptides were used (Chiron Mimotopes, Australia). These peptides were linked via residues SGSG at the N-terminus to biotin. p21, KRRQTSMTDFYHSKRRLIFS; p21-A, KRRATSMTDFYHSKRRLIFS; Dacapo, RKRQPKITEFMKERKRLAQA; Dacapo-A, RKRAPKITEFMKERKRLAQA; Z50796, TKRQQKMTDFMAVSRKKNSL; Z50796-A, TKRQQKMTDFMAVSRKKNSL; Pogo, KLFNLHINSAVLQKKITDYF; Pogo-A, KLFNLHINSAVLAKKITDYF; Tigger, LMWQTSLLSYFKKLPQPPQP; Tigger-A, LMWQTSLLSYFKKLPQPPQP; Consensus, KKRQKRLTDFFKRKRKLKEA; Consensus-A, KKRAKRLTDFFKRKRKLKEA; Unrelated, PESVELKWSEPNEEELIKFM. ELISA analysis of PCNA binding to peptides Plastic plates for enzyme-linked immunoabsorption assay (ELISA) were coated with 100 m l 5 m g/ml streptavidin by drying overnight at 37 C. They were washed with phosphate-buffered saline (PBS) containing 0.2% Tween 20 (PBST) and blocked with 5% non-fat milk powder in PBS for at least 2 h at room temperature. Each well was incubated in turn with the following reagents (the wells were extensively washed in PBST between steps): (i) 0.5 m g peptide diluted in 100 m l 0.1% milk-PBS (2 h, room temperature); (ii) up to 50 m g protein in 100 m l 0.1% milk-PBS, either total cell lysates of E.coli BL21 expressing human or S.pombe PCNA or purified human or S.pombe PCNA (1 h, room temperature); (iii) primary anti-PCNA rabbit polyclonal antiserum 3009 or PC10 monoclonal (...truncated)