Timeless Interacts with PARP-1 to Promote Homologous Recombination Repair

•Timeless interacts with PARP-1 independent of poly(ADP-ribosyl)ation•Crystal structures of Timeless PAB in free form and in complex with PARP-1•Specific recognition of PARP-1 by Timeless does not affect its enzymatic activity•PARP-1 is required for Timeless recruitment to DSB sites to promote HR repair Human Timeless helps stabilize replication forks during normal DNA replication and plays a critical role in activation of the S phase checkpoint and proper establishment of sister chromatid cohesion. However, it remains elusive whether Timeless is involved in the repair of damaged DNA. Here, we identify that Timeless physically interacts with PARP-1 independent of poly(ADP-ribosyl)ation. We present high-resolution crystal structures of Timeless PAB (PARP-1-binding domain) in free form and in complex with PARP-1 catalytic domain. Interestingly, Timeless PAB domain specifically recognizes PARP-1, but not PARP-2 or PARP-3. Timeless-PARP-1 interaction does not interfere with PARP-1 enzymatic activity. We demonstrate that rapid and transient accumulation of Timeless at laser-induced DNA damage sites requires PARP-1, but not poly(ADP-ribosyl)ation and that Timeless is co-trapped with PARP-1 at DNA lesions upon PARP inhibition. Furthermore, we show that Timeless and PARP-1 interaction is required for efficient homologous recombination repair. Human Timeless helps stabilize replication forks during normal DNA replication and plays a critical role in activation of the S phase checkpoint and proper establishment of sister chromatid cohesion. However, it remains elusive whether Timeless is involved in the repair of damaged DNA. Here, we identify that Timeless physically interacts with PARP-1 independent of poly(ADP-ribosyl)ation. We present high-resolution crystal structures of Timeless PAB (PARP-1-binding domain) in free form and in complex with PARP-1 catalytic domain. Interestingly, Timeless PAB domain specifically recognizes PARP-1, but not PARP-2 or PARP-3. Timeless-PARP-1 interaction does not interfere with PARP-1 enzymatic activity. We demonstrate that rapid and transient accumulation of Timeless at laser-induced DNA damage sites requires PARP-1, but not poly(ADP-ribosyl)ation and that Timeless is co-trapped with PARP-1 at DNA lesions upon PARP inhibition. Furthermore, we show that Timeless and PARP-1 interaction is required for efficient homologous recombination repair. Faithful transmission of genetic information from parent to daughter cells is central for eukaryotic life. However, eukaryotic cells are constantly facing challenges from both environmental and endogenous sources and defects in DNA replication and repair pathways are major causes of cancer and premature aging. Cells have evolved elaborate surveillance mechanisms, or checkpoints, to detect and respond to DNA damage and replication stress. Under conditions of replication stress, components of the DNA replication fork protection complex including Timeless and Tipin act together with a variety of proteins such as RPA, Claspin, ATR/ATRIP, and Chk1 to activate the intra-S phase checkpoint, maintain structural stability of stalled replication forks, and eventually promote restart of the stalled forks (Branzei and Foiani, 2010Branzei D. Foiani M. Maintaining genome stability at the replication fork.Nat. Rev. Mol. Cell Biol. 2010; 11: 208-219Crossref PubMed Scopus (608) Google Scholar, Ciccia and Elledge, 2010Ciccia A. Elledge S.J. The DNA damage response: making it safe to play with knives.Mol. Cell. 2010; 40: 179-204Abstract Full Text Full Text PDF PubMed Scopus (2924) Google Scholar, Errico and Costanzo, 2012Errico A. Costanzo V. Mechanisms of replication fork protection: a safeguard for genome stability.Crit. Rev. Biochem. Mol. Biol. 2012; 47: 222-235Crossref PubMed Scopus (115) Google Scholar). Mammalian Timeless was originally identified on the basis of its homology to the Drosophila Timeless gene (Tim-1) (Sangoram et al., 1998Sangoram A.M. Saez L. Antoch M.P. Gekakis N. Staknis D. Whiteley A. Fruechte E.M. Vitaterna M.H. Shimomura K. King D.P. et al.Mammalian circadian autoregulatory loop: a timeless ortholog and mPer1 interact and negatively regulate CLOCK-BMAL1-induced transcription.Neuron. 1998; 21: 1101-1113Abstract Full Text Full Text PDF PubMed Scopus (300) Google Scholar, Zylka et al., 1998Zylka M.J. Shearman L.P. Levine J.D. Jin X. Weaver D.R. Reppert S.M. Molecular analysis of mammalian timeless.Neuron. 1998; 21: 1115-1122Abstract Full Text Full Text PDF PubMed Scopus (157) Google Scholar). Later, phylogenetic sequence analysis indicated that mammalian Timeless is more closely related to Drosophila Timeout gene (Tim-2) (Benna et al., 2000Benna C. Scannapieco P. Piccin A. Sandrelli F. Zordan M. Rosato E. Kyriacou C.P. Valle G. Costa R. A second timeless gene in Drosophila shares greater sequence similarity with mammalian tim.Curr. Biol. 2000; 10: R512-R513Abstract Full Text Full Text PDF PubMed Scopus (67) Google Scholar). Drosophila Tim-1 and Tim-2 have been demonstrated to be key players in circadian rhythm (Benna et al., 2010Benna C. Bonaccorsi S. Wülbeck C. Helfrich-Förster C. Gatti M. Kyriacou C.P. Costa R. Sandrelli F. Drosophila timeless2 is required for chromosome stability and circadian photoreception.Curr. Biol. 2010; 20: 346-352Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar, Hardin, 2005Hardin P.E. The circadian timekeeping system of Drosophila.Curr. Biol. 2005; 15: R714-R722Abstract Full Text Full Text PDF PubMed Scopus (346) Google Scholar), but it remains debatable if mammalian Timeless functions as a circadian clock factor (Barnes et al., 2003Barnes J.W. Tischkau S.A. Barnes J.A. Mitchell J.W. Burgoon P.W. Hickok J.R. Gillette M.U. Requirement of mammalian Timeless for circadian rhythmicity.Science. 2003; 302: 439-442Crossref PubMed Scopus (171) Google Scholar, Gotter, 2003Gotter A.L. Tipin, a novel timeless-interacting protein, is developmentally co-expressed with timeless and disrupts its self-association.J. Mol. Biol. 2003; 331: 167-176Crossref PubMed Scopus (54) Google Scholar, Gotter et al., 2000Gotter A.L. Manganaro T. Weaver D.R. Kolakowski Jr., L.F. Possidente B. Sriram S. MacLaughlin D.T. Reppert S.M. A time-less function for mouse timeless.Nat. Neurosci. 2000; 3: 755-756Crossref PubMed Scopus (147) Google Scholar). Nonetheless, mammalian Timeless and its paralogs (Tof1 in Saccharomyces cerevisiae, Swi1 in Schizosaccharomyces pombe, and Tim-1 in Caenorhabditis elegans) have been shown to play an essential role in proper progression of DNA replication, activation of cell-cycle checkpoints, and the establishment of sister chromatid cohesion (Chou and Elledge, 2006Chou D.M. Elledge S.J. Tipin and Timeless form a mutually protective complex required for genotoxic stress resistance and checkpoint function.Proc. Natl. Acad. Sci. USA. 2006; 103: 18143-18147Crossref PubMed Scopus (129) Google Scholar, Errico et al., 2007Errico A. Costanzo V. Hunt T. Tipin is required for stalled replication forks to resume DNA replication after removal of aphidicolin in Xenopus egg extracts.Proc. Natl. Acad. Sci. USA. 2007; 104: 14929-14934Crossref PubMed Scopus (81) Google Scholar, Gotter et al., 2007Gotter A.L. Suppa C. Emanuel B.S. Mammalian TIMELESS and Tipin are evolutionarily conserved replication fork-associated factors.J. Mol. Biol. 2007; 366: 36-52Crossref PubMed Scopus (106) Google Scholar, Leman et al., 2010Leman A.R. Noguchi C. Lee C.Y. Noguchi E. Human Timeless and Tipin stabilize replication forks and facilitate sister-chromatid cohesion.J. Cell Sci. 2010; 123: 660-670Crossref PubMed Scopus (99) Google Scholar, Smith et al., 2009Smith K.D. Fu M.A. Brown E.J. Tim-Tipin dysfunction creates an indispensible reliance on the ATR-Chk1 pathway for continued DNA synthesis.J. Cell Biol. 2009; 187: 15-23Crossref PubMed Scopus (68) Google Scholar, Unsal-Kaçmaz et al., 2005Unsal-Kaçmaz K. Mullen T.E. Kaufmann W.K. Sancar A. Coupling of human circadian and cell cycles by the timeless protein.Mol. Cell. Biol. 2005; 25: 3109-3116Crossref PubMed Scopus (249) Google Scholar, Unsal-Kaçmaz et al., 2007Unsal-Kaçmaz K. Chastain P.D. Qu P.P. Minoo P. Cordeiro-Stone M. Sancar A. Kaufmann W.K. The human Tim/Tipin complex coordinates an Intra-S checkpoint response to UV that slows replication fork displacement.Mol. Cell. Biol. 2007; 27: 3131-3142Crossref PubMed Scopus (193) Google Scholar, Urtishak et al., 2009Urtishak K.A. Smith K.D. Chanoux R.A. Greenberg R.A. Johnson F.B. Brown E.J. Timeless maintains genomic stability and suppresses sister chromatid exchange during unperturbed DNA replication.J. Biol. Chem. 2009; 284: 8777-8785Crossref PubMed Scopus (32) Google Scholar, Yoshizawa-Sugata and Masai, 2007Yoshizawa-Sugata N. Masai H. Human Tim/Timeless-interacting protein, Tipin, is required for efficient progression of S phase and DNA replication checkpoint.J. Biol. Chem. 2007; 282: 2729-2740Crossref PubMed Scopus (105) Google Scholar). Timeless forms a stable complex with its partner protein Tipin. The Timeless-Tipin complex has been reported to travel along with the replication fork during unperturbed DNA replication. Tipin physically interacts with RPA, while Timeless directly interacts with replisome components including the replicative helicase CDC45-MCM-GINS (CMG) complex and DNA polymerase ε (Aria et al., 2013Aria V. De Felice M. Di Perna R. Uno S. Masai H. Syväoja J.E. van Loon B. Hübscher U. Pisani F.M. The human Tim-Tipin complex interacts directly with DNA polymerase epsilon and stimulates its synthetic activity.J. Biol. Chem. 2013; 288: 12742-12752Crossref PubMed Scopus (13) Google Scholar, Cho et al., 2013Cho W.H. Kang Y.H. An Y.Y. Tappin I. Hurwitz J. Lee J.K. Human Tim-Tipin complex affects the biochemical properties of the replicative DNA helicase and DNA polymerases.Proc. Natl. Acad. Sci. USA. 2013; 110: 2523-2527Crossref PubMed Scopus (46) Google Scholar). Timeless appears to stimulate Polε polymerase activity, but negatively regulates CMG helicase activity. This unique function of Timeless might be essential to prevent disassembly of the replisome at paused replication forks, and re-assemble a replisome to facilitate fork restart when DNA replication is reactivated. In addition, the Timeless-Tipin complex has also been demonstrated to interact with Claspin and facilitate the accumulation of Claspin at the replication fork in response to replication stress (Yoshizawa-Sugata and Masai, 2007Yoshizawa-Sugata N. Masai H. Human Tim/Timeless-interacting protein, Tipin, is required for efficient progression of S phase and DNA replication checkpoint.J. Biol. Chem. 2007; 282: 2729-2740Crossref PubMed Scopus (105) Google Scholar). The Timeless-Tipin-Claspin complex contributes to full activation of the ATR-Chk1 signaling pathway through the recruitment of Chk1 to arrested replication forks for sufficient ATR-mediated phosphorylation. Besides Tipin, Timeless has also been demonstrated to associate with a list of other nuclear proteins including ChIR1, SMC1, Dbf4, and Plk1 and is involved in multiple cellular processes such as sister chromatid cohesion (Leman et al., 2012Leman A.R. Dheekollu J. Deng Z. Lee S.W. Das M.M. Lieberman P.M. Noguchi E. Timeless preserves telomere length by promoting efficient DNA replication through human telomeres.Cell Cycle. 2012; 11: 2337-2347Crossref PubMed Scopus (48) Google Scholar, Murakami and Keeney, 2014Murakami H. Keeney S. Temporospatial coordination of meiotic DNA replication and recombination via DDK recruitment to replisomes.Cell. 2014; 158: 861-873Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar, Serçin and Kemp, 2011Serçin O. Kemp M.G. Characterization of functional domains in human Claspin.Cell Cycle. 2011; 10: 1599-1606Crossref PubMed Scopus (21) Google Scholar), temporal control of replication and meiotic double-strand break (DSB) formation, replication termination, and mitotic entry. In this study, we report that human Timeless forms a tight complex with poly(ADP-ribose) polymerase 1 (PARP-1) that is independent of poly(ADP-ribosyl)ation. In addition, we determined crystal structures of the Timeless PARP-1-Binding domain (PAB) in free form and in complex with the PARP-1 C-terminal catalytic domain. Interestingly, Timeless specifically binds to the catalytic region of PARP-1, but not other PARP family members and this physical interaction does not seem to affect PARP-1 enzymatic activity. Finally, we demonstrate that recruitment of Timeless to DNA damage sites depends on the physical interaction with PARP-1, but is independent of PARP-1 enzymatic activity. We postulate that the interaction of Timeless and PARP-1 could facilitate DNA repair, in particular homologous recombination (HR) repair. Human Timeless is a multi-domain containing protein involved in various cellular processes. Previous studies have defined four conserved regions in mouse Timeless based on sequence alignment (Yoshizawa-Sugata and Masai, 2007Yoshizawa-Sugata N. Masai H. Human Tim/Timeless-interacting protein, Tipin, is required for efficient progression of S phase and DNA replication checkpoint.J. Biol. Chem. 2007; 282: 2729-2740Crossref PubMed Scopus (105) Google Scholar, Zylka et al., 1998Zylka M.J. Shearman L.P. Levine J.D. Jin X. Weaver D.R. Reppert S.M. Molecular analysis of mammalian timeless.Neuron. 1998; 21: 1115-1122Abstract Full Text Full Text PDF PubMed Scopus (157) Google Scholar); however, the precise molecular functions of these different domains have been elusive. To gain further insights into the structural domain organization of human Timeless, we cloned the individual domains. We first determined the high-resolution crystal structure of the Timeless_D4 domain, encompassing amino acids 1,000 to 1,098. The structural analysis suggests this Timeless_D4 domain may function as a protein-protein interacting modulator. In an effort to unravel the biological function of the Timeless_D4 domain, we performed tandem affinity purification (TAP) using cell lysates prepared from human embryonic kidney (HEK)293T cells stably expressing triple-epitope (S-protein, FLAG, and streptavidin-binding peptide)-tagged Timeless_D4. Mass spectrometry analysis revealed that one of the possible Timeless_D4-associated proteins is PARP-1. The physical interaction between full-length (fl) Timeless and PARP-1 was then confirmed in HeLa cells using co-immunoprecipitation (coIP) followed by western blot analysis (Figure 1A). We also carried out in vitro pull-down assay and confirmed that fl-Timeless associates with PARP-1 (Figure S1A). Furthermore, we demonstrated that the Timeless_D4 domain specifically interacts with the PARP-1 catalytic domain (Figure 1B). Thus, we named the Timeless_D4 domain PARP-1-binding domain (PAB). The Timeless PAB domain exists as a dimer in solution, as demonstrated by analytical gel filtration (Figure S1B). Consistently, the Timeless PAB domain forms a head-to-tail homodimer in the crystal structure (Figure 2A and crystallographic statistics are given in Table 1) solved initially with single-wavelength anomalous diffraction. Each PAB subunit shares an almost identical structure that mainly consists of a bundle of five α helices. As shown in cartoon representation, helix α2 connects to helix α3 with a 15-amino acid (aa)-long extended loop, while helices α4 and α5 are linked by a 20-aa-long U-shaped loop. The conformation of the U-shaped loop and the α2/α3 loop is further stabilized by an inter-loop hydrogen bonding network as well as salt bridges formed between side chains of Arg1070 and Glu1077 (Figure S1C). The dimeric interface is mainly composed of residues from the U-shaped loop and helix α3 of each subunit. Both electrostatic and hydrophobic interactions contribute to the formation of the stable homodimer, burying the total solvent accessible surface area of 1,220 Å2. Notably, residue Glu1052 of helix α3 and Arg1081′ (with the prime indicating the other subunit) located in the U-shaped loop, as well as the counterpart residues Glu1052′ and Arg1081, form two ion pairs (2.9 Å) that contribute significantly to the dimer formation (Figure 2B). Additionally, residues Thr1078 and Phe1079 located in the U-shaped loop have hydrophobic interactions with their equivalent counterparts from the other subunit. It is worth to note that a structural similarity search from the DALI server revealed that the architecture of the Timeless PAB shares certain similarity to that of PUB domains of PNGase and HOIP (Allen et al., 2006Allen M.D. Buchberger A. Bycroft M. The PUB domain functions as a p97 binding module in human peptide N-glycanase.J. Biol. Chem. 2006; 281: 25502-25508Crossref PubMed Scopus (71) Google Scholar, Elliott et al., 2014Elliott P.R. Nielsen S.V. Marco-Casanova P. Fiil B.K. Keusekotten K. Mailand N. Freund S.M. Gyrd-Hansen M. Komander D. Molecular basis and regulation of OTULIN-LUBAC interaction.Mol. Cell. 2014; 54: 335-348Abstract Full Text Full Text PDF PubMed Scopus (136) Google Scholar, Schaeffer et al., 2014Schaeffer V. Akutsu M. Olma M.H. Gomes L.C. Kawasaki M. Dikic I. Binding of OTULIN to the PUB domain of HOIP controls NF-κB signaling.Mol. Cell. 2014; 54: 349-361Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar), despite the low level of aa conservation at equivalent positions (Figure S1D). The PNGase PUB domain recognizes a conserved tyrosine-based motif in AAA+ ATPase VCP/p97 and modulates ER-associated protein degradation activity (Allen et al., 2006Allen M.D. Buchberger A. Bycroft M. The PUB domain functions as a p97 binding module in human peptide N-glycanase.J. Biol. Chem. 2006; 281: 25502-25508Crossref PubMed Scopus (71) Google Scholar). More recently the PUB domain of HOIP has been demonstrated to bind to OTULIN to control NF-κB signaling (Elliott et al., 2014Elliott P.R. Nielsen S.V. Marco-Casanova P. Fiil B.K. Keusekotten K. Mailand N. Freund S.M. Gyrd-Hansen M. Komander D. Molecular basis and regulation of OTULIN-LUBAC interaction.Mol. Cell. 2014; 54: 335-348Abstract Full Text Full Text PDF PubMed Scopus (136) Google Scholar, Schaeffer et al., 2014Schaeffer V. Akutsu M. Olma M.H. Gomes L.C. Kawasaki M. Dikic I. Binding of OTULIN to the PUB domain of HOIP controls NF-κB signaling.Mol. Cell. 2014; 54: 349-361Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar). However, the putative pocket in the PUB domain for the partner protein binding is not conserved in the Timeless PAB domain (Figure S1E).Table 1Summary of X-Ray Diffraction Data and Structure Refinement StatisticsCrystalTimeless_PABTimeless_PABTimeless/PARP-1SeMETNativeData CollectionBeamlineBL17UBL17UBL17UProtein Data Bank code4XHW4XHT4XHUWavelength (Å)0.979231.069970.97907Space groupC2221C2221C121Unit cell a, b, and c (Å)66.4, 100.5, and 149.566.6, 100.6, and 149.599.4, 98.4, and 116.9Resolution (Å)50.0−2.8050.0−1.6550.0−2.05(2.85−2.80)aNumbers in parentheses refer to the highest resolution shell.(1.68−1.65)(2.09−2.05)Observed reflections181,889375,510188,441Unique reflections12,777 (613)59,619 (5,562)62,689 (3,191)Completeness (%)100.0 (100.0)98.4 (100.0)99.6 (100.0)Rmerge0.134 (0.486)0.071 (0.431)0.070 (0.458)Average I/σ (I)27.1 (6.6)29.65 (4.64)15.9 (3.0)Redundancy14.2 (14.7)6.3 (6.3)3.2 (3.2)RefinementResolution (Å)24.93−2.8026.34−1.6525.1−2.09(2.89−2.80)(1.71−1.65)(2.16−2.09)No. reflections12,75159,61960,143Rwork/RfreebRfree was calculated using 5% random data omitted from the refinement. (%)19.3 / 25.318.5 / 22.218.3 / 21.1No. atoms2,8543,2436,829Protein2,8542,9216,548Ligand0033Water0317248B-factors (Å2)46.937.552RMSDBond lengths (Å)0.0110.0070.008Bond angles (°)1.401.091.12Ramachandran StatisticsFavored (%)1009999Allowed (%)011Outliners (%)000a Numbers in parentheses refer to the highest resolution shell.b Rfree was calculated using 5% random data omitted from the refinement. Open table in a new tab We went on to determine the crystal structure of the Timeless PAB domain in complex with the PARP-1 C-terminal catalytic domain, including the helical subdomain (HD) and the ADP-ribose transferase (ART) domain at the resolution of 2.09Å (Figure 2C). The Timeless PAB domain forms a 1:1 heterodimer with the PARP-1 catalytic domain, with a total buried surface area of 1,470 Å2. Similar to the PAB homodimeric interface, the binding surface of Timeless PAB in the Timeless-PARP-1 heterodimer also comprises the U-shaped loop and helix α3 that constitute the “site I” and “site II” interface of the Timeless-PARP-1 heterodimer as highlighted in Figure 2C. On the other hand, the binding interface of the PARP-1 catalytic domain is on the opposite side of its enzymatic active site. The structure arrangement of the Timeless PAB domain is highly similar to that in free form, with 0.43Å root-mean-square deviation (RMSD) for the aligned Cα atoms. Interestingly, at the site I interface, those residues in the U-shaped loop critical for PAB homodimer formation also participate in Timeless-PARP-1 complex formation (Figure 2E). For example, Timeless Arg1081 forms an ion pair (2.7 and 2.9Å) with PARP-1 Asp993. The side chain of Timeless Phe1079 is in hydrophobic contact with PARP-1 Phe851. Additionally, the Timeless Gln1076 side chain carboxamide group forms hydrogen bonds with PARP-1 Ile879 backbone amine and carbonyl groups. Another salient feature of the Timeless-PARP-1 complex is that the Timeless PAB helix α3, with a strong negative electrostatic potential surface (site II interface), tightly associates with the PARP-1 loop segment encompassing residues 932–947, that has a highly positive charged surface (Figures 2D and 2F). Specifically, side chains of Timeless Glu1049 and Glu1056 form ion pairs with PARP-1 Lys943 and Lys940, respectively. It is likely these massive networks of electrostatic interactions provide the driving force to disrupt the Timeless PAB homodimer and promote the Timeless-PARP-1 heterodimer formation when adding the PARP-1 catalytic domain sample into the Timeless PAB solution. To corroborate the interaction observed in the crystal structure, we applied site-directed mutagenesis and quantitatively measured the effect of mutation on the binding of Timeless and PARP-1 using surface plasmon resonance (SPR). The Timeless PAB domain binds to the PARP-1 C-terminal catalytic domain with a Kd of 26.0 ± 1.5 nM (Figure 3A), whereas mutation on either Timeless R1081G or PARP-1 D993G totally abolishes the binding (Figures S2A and S2B), indicating that the specific interaction of Timeless R1081 with PARP-1 D993 contributes substantially to the complex formation. CoIP and in vitro glutathione S-transferase (GST) pull-down assay further support this conclusion (Figures 3B and 3E). PARP-1 is one of the most abundant nuclear enzymes, and it is estimated to mediate over 80% of cellular poly-ADP-ribose synthesis (Shieh et al., 1998Shieh W.M. Amé J.C. Wilson M.V. Wang Z.Q. Koh D.W. Jacobson M.K. Jacobson E.L. Poly(ADP-ribose) polymerase null mouse cells synthesize ADP-ribose polymers.J. Biol. Chem. 1998; 273: 30069-30072Crossref PubMed Scopus (262) Google Scholar). PARP-1 is also the best characterized member in the poly(ADP-ribose) polymerase family, which currently comprises 17 members (Schreiber et al., 2006Schreiber V. Dantzer F. Ame J.C. de Murcia G. Poly(ADP-ribose): novel functions for an old molecule.Nat. Rev. Mol. Cell Biol. 2006; 7: 517-528Crossref PubMed Scopus (1568) Google Scholar). So far PARP-1, PARP-2, and PARP-3 have been implicated in the DNA damage response. PARP-1 contains three zinc fingers at the N terminus that are critical for sensing DNA damage, these DNA binding domains are not present in PARP-2 or PARP-3. However, all of three PARP members share a structurally similar catalytic domain. The result from our domain mapping experiment indicated only the C-terminal catalytic domain of PARP-1 is required for Timeless PAB binding, which prompted us to investigate whether Timeless binds to the catalytic domain of PARP-2 or PARP-3. The GST pull-down assay indicated Timeless PAB binds specifically to the catalytic domain of PARP-1, but not PARP-2 or PARP-3 (Figure 3C). In fact, the structural basis of this binding specificity can be explained by the crystal structure of PARP-1 and the Timeless heterodimer. Structure-based sequence alignment of the catalytic domains of human PARP-1, PARP-2, and PARP-3 is shown in Figure 3D. The residues of the PARP-1 catalytic domain directly involved in the Timeless-PARP-1 complex formation are highlighted by the light gray background. Interestingly, most of the residues in PARP-1 critical for the complex formation are not conserved in PARP-2 or PARP-3. For example, PARP-1 Asp993 forms salt bridges with Timeless Arg1081 and the corresponding residues of PARP-1 Asp993 in PARP-2 and PARP-3 are Asn563 and Gln519, respectively, neither of which could form an ion pair with Timeless Arg1081. Furthermore, PARP-1 possesses positively charged Lys940 and Lys943 that are involved in strong electrostatic interactions with the Timeless PAB domain, but neither PARP-2 nor PARP-3 has positively charged residues at equivalent positions. To further verify the importance of these residues in recognition of Timeless by PARP-1, we generated a number of mutations in PARP-1 to examine the interactions with their wild-type counterparts using both in vitro pull-down assay and SPR. In complete agreement with our structural findings, single mutation of PARP-1 Asp993 to Gly, or double mutation of Lys940 and Lys943 to Gly and Gln that mimics the corresponding region of PARP-2, or double deletion of Pro850 and Phe851, led to the abrogation of interactions with Timeless, as demonstrated by in vitro GST pull-down assay (Figure 3E). Consistently, SPR measurement indicates that PARP-1 Pro850 and Phe851 double deletion or D993G mutation disrupts the binding of the PARP-1 catalytic domain to the Timeless PAB domain, while PARP-1 K940G/K943Q double mutation resulted in an almost 20-fold reduction in Timeless PAB binding (Figures S2B–S2D). Considering that the PARP-1 catalytic domain is required for Timeless binding, we wondered whether Timeless-PARP-1 interaction affects PARP-1 enzymatic activity. To analyze this, we performed in vitro PARP-1 enzymatic assays. In presence of DNA, PARP-1 can efficiently poly(ADP-ribosyl)ate itself when adding NAD+, as demonstrated by SDS-PAGE, the auto-ADP-ribosylated PARP-1 gradually migrated as the reaction proceeded (Figure 3F, left side). Adding the Timeless PAB domain into the reaction did not block PARP-1 auto-ribosylation (Figure 3F, right side). This actually is consistent with the structural observation that Timeless PAB binding does not cause obvious conformational changes in the PARP-1 catalytic domain. A recent structure-based study showed that conformational changes of HD preceding the ART could modulate PARP-1 catalytic activity (Langelier et al., 2012Langelier M.F. Planck J.L. Roy S. Pascal J.M. Structural basis for DNA damage-dependent poly(ADP-ribosyl)ation by human PARP-1.Science. 2012; 336: 728-732Crossref PubMed Scopus (413) Google Scholar). Structural superimposition of the PARP-1-Timeless complex and PARP-1 in free form revealed that PARP-1 catalytic domains in the two structures are almost identical (Figure S3), no obvious structural distortion was observed in the HD domain, indicating Timeless binding does not induce conformational changes in PARP-1 and thus does not affect its enzymatic activity. PARP-1 functions as a sensor protein that directly recognizes DNA DSBs or SSBs and promotes the rapid recruitment of a number of proteins to DNA damage sites in a PAR-dependent manner (Ali et al., 2012Ali A.A. Timinszky G. Arribas-Bosacoma R. Kozlowski M. Hassa P.O. Hassler M. Ladurner A.G. Pearl L.H. Oliver A.W. The zinc-finger domains of PARP1 cooperate to recognize DNA strand breaks.Nat. Struct. Mol. Biol. 2012; 19: 685-692Crossref PubMed Scopus (161) Google Scholar, Langelier et al., 2012Langelier M.F. Planck J.L. Roy S. Pascal J.M. Structural basis for DNA damage-dependent poly(ADP-ribosyl)ation by human PARP-1.Science. 2012; 336: 728-732Crossref PubMed Scopus (413) Google Scholar, Gibson and Kraus, 2012Gibson B.A. Kraus W.L. New insights into the molecular and cellular functions of poly(ADP-ribose) and PARPs.Nat. Rev. Mol. Cell Biol. 2012; 13: 411-424Crossref PubMed Scopus (855) Google Scholar). Since we have demonstrated that Timeless interacts with PARP-1, we tested if PARP-1 targets Timeless to DNA damage sites. Endogenous Timeless could be detected at DNA damage tracks as early as 10 min after microirradiation in both S and non-S phase cells (Figures 4B, 4C , and S4A). No Timeless accumulation could be observed at later time points, indicating a rather transient recruitment. To analyze whether PARP-1 mediated poly(ADP-ribosyl)ation at DNA damage sites is required for recruitment of Timeless, we microirradiated U2OS cells in the presence or absence of the PARP-1 inhibitor Olaparib. Interestingly, we observed a prolonged retention of Timeless and PARP-1 at laser tracks, while recruitment of XRCC1 was abolished in the presence of Olaparib (Figures 4B and S4B). The prolonged retention of Timeless at damaged chromatin is likely due to trapping of PARP-1 at DNA damage sites. To monitor the spatio-temporal accumulation of Timeless and its trapping at DNA damage sites upon PARP-1 inhibition in real time, we performed live-cell analysis in cells expressing GFP-tagged Timeless. GFP-Timeless readily accumulated at laser-induced DNA damage sites and reached the maximum level after 10–15 s (Figures 4D and 4E), suggesting an early role for Timeless in the DNA damage response (DDR). Interestingly, PARP-1 inhibition led to stronger and prolonged recruitment of GFP-Timeless, as indicated by the Timeless accumulation reaching its plateau much later at around 75 s