sign in sign up
<<
Home , DNA clamp

bb10a07.qxd 02/05/2000 10:57 Page R25

bb10a07

Dispatch R25

Sliding clamps: A (tail)ored fit Manju M. Hingorani and Mike O’Donnell

New structural information on the architecture of a available on the structure of DNA , not many DNA provides insights into a number of structural details are known about how these DNA metabolic processes and their modulation by interact with their accessory proteins, and how all the pro- circular ‘sliding clamps’, which form rings around DNA teins assemble together in a replisome at the DNA replica- that play an important role in processive processes tion fork. Shamoo and Steitz [6] have taken the first step in such as replication. solving the crystal structure of a replisome, by determining how the RB69 DNA binds to its Address: The Rockefeller University, 1230 York Avenue, New York, New York 10021, USA. factor. Furthermore, they have found that the poly- E-mail: [email protected]; [email protected] merase–clamp interaction appears remarkably similar to the connection between the cell-cycle inhibitor p21CIP1 Current Biology 2000, 10:R25–R29 and PCNA [4], suggesting a general mode by which clamps 0960-9822/00/$ – see front matter tether proteins to DNA. © 2000 Elsevier Science Ltd. All rights reserved. The RB69 polymerase interacts with its clamp through the DNA sliding clamps are ring-shaped proteins that bind last ten or so amino acids at the carboxyl terminus. This DNA, not through sequence-specific contacts, but rather carboxy-terminal peptide projects straight out from behind by encircling the DNA double helix and forming a the polymerase [7], and appears to literally hook the poly- topological link with it. The rings have an inner diameter merase to the clamp (Figure 2a) [6]. Highly conserved of 30–35 Å, easily large enough to accommodate DNA with residues at the carboxyl terminus of the polymerase bind to no steric repulsion, allowing the clamps to slide freely a hydrophobic pocket on one of three identical subunits of along the duplex (Figure 1). Sliding clamps are well known the RB69 clamp (Figure 2b). The final five residues have a primarily as DNA replication accessory proteins that stable helical structure, but the adjoining five residues may increase the processivity of DNA polymerases [1]. assume different conformations, resulting in fewer con- straints on the position of the polymerase on DNA with Polymerases replicating genomic DNA typically function respect to its clamp. This apparently flexible connection along with several accessory proteins in a complex known as and rather small area of contact between the polymerase a replisome. These proteins fine-tune the polymerase activ- and its processivity factor has mechanistic and biological ity for rapid and efficient DNA synthesis. For example, the significance as discussed below. polymerase itself catalyzes continuous synthesis of only a few before falling off the primer–template Consider, for example, the problem of torsional stress junction. The sliding clamp binds the polymerase and generated when a DNA replisome containing two maintains a topological link with DNA, functioning as a polymerases coordinates leading-strand and lagging-strand mobile tether to keep the polymerase attached to DNA as it DNA replication (Figure 3). Each polymerase must follow replicates several thousand nucleotides. This strategy for the turn of the double helix as it extends DNA. As DNA is processive DNA replication is widely used among organ- synthesized at a rate of ~1 kilobase per second, the isms ranging from bacteriophage to humans. The similar polymerases must go through one complete turn every toroidal structures of the (β), Saccharomyces 10 milliseconds. Given that the two polymerases are firmly cerevisiae (yPCNA), human (hPCNA) and bacteriophage T4 attached to each other, the result could be a rapid build-up (gp45) sliding clamps [2–5] all attest to the remarkable of tangled DNA and proteins at the replication fork. The utility and ubiquity of the mobile tether mechanism for pile-up could be avoided if the newly synthesized DNA polymerase processivity (Figure 1). turns behind the polymerase, as shown in Figure 3. The leading strand remains constrained, however, as the The recently determined structure of the bacteriophage duplex DNA behind the polymerase cannot swivel around RB69 DNA replicase has revealed another circular clamp, freely to relieve the torsional stress — as can occur on the similar to the ones listed above (Figure 1) [6]. More signifi- lagging DNA strand because of the single-stranded region cantly, the work has provided an image of the interaction behind the Okazaki fragment. between DNA polymerase and its sliding clamp (Figure 2a). The authors solved the crystal structures of the Consider, then, the possibility that the polymerase might RB69 DNA polymerase as well as the RB69 clamp, both transiently release DNA, while remaining bound to the free and in a complex with a carboxy-terminal peptide clamp on DNA via the peptide connector, and allow the from the polymerase. While substantial information is DNA to swivel (within the clamp), before rapidly rebinding bb10a07.qxd 02/05/2000 10:57 Page R26

R26 Current Biology Vol 10 No 1

Figure 1

Clamps, clamps everywhere… Front (left) and side (right) views of crystal structures of sliding clamps from bacteriophage T4 (gp45), bacteriophage RB69, E. coli (β), S. cerevisiae (yPCNA) and humans (hPCNA).

the primer–template junction and continuing replication as well. For example, this property might aid the poly- (Figure 3). Interestingly, a replicating E. coli DNA poly- merase by allowing it to bypass lesions — sliding over merase III holoenzyme can release the 3′ primer terminus them on the clamp — and restart synthesis at new primed and pass over an intervening double-stranded region on the sites downstream of the blockage. The potential advan- template without dissociating from the DNA [8]. In this tages of a flexible connector are also noted by Shamoo and reaction, the polymerase releases the 3′ end of DNA and Steitz [6], who suggest that it may facilitate small move- likely traverses the duplex region riding piggy-back on the ments of the polymerase during transfer of DNA between sliding clamp, then restarts DNA synthesis at the down- the polymerase and editing active sites. stream 3′ end. The presence of a flexible tether between the RB69 polymerase and clamp leaves open the possibility The RB69 bacteriophage DNA replisome shares that DNA could move transiently in and out of the active common features with from other organisms, site on a polymerase and yet remain firmly associated with and is therefore a good model system for understanding the replisome through the circular clamp. in detail how DNA polymerases might interact with and use their sliding clamps. In addition to the RB69 poly- The ability to make small and rapid movements on/off the merase, the bacteriophage T4 and the HSV1 DNA poly- 3′ terminus without complete dissociation from template merases also bind their processivity factors through DNA may serve the polymerase under other circumstances hydrophobic residues at their carboxyl termini [9,10]. In bb10a07.qxd 02/05/2000 10:57 Page R27

Dispatch R27

Figure 2

RB69 DNA polymerase, the RB69 clamp and PCNA. (a) A model and human PCNA are shown complexed with the carboxy-terminal structure of DNA polymerase with primer–template DNA in the peptides (red) of (b) RB69 DNA polymerase and (c) p21CIP1. editing site, docked with the sliding clamp. (b,c) The RB69 clamp

fact, the RB69 and T4 DNA polymerases share a consen- fork might also use the strategy of hooking up with each sus sequence that is essential for interaction with their other through flexible peptide connector domains. respective circular clamps [6]. Another well known function of circular sliding clamps The work of Shamoo and Steitz [6] has also provided involves their ability to serve as mobile, DNA-tracking important clues regarding the architecture of eukaryotic scaffolds for other enzymes besides the DNA polymerase. DNA replisomes. This study has revealed striking struc- Elegant research by Geiduschek et al. [16] has shown that tural similarities between the carboxy-terminal peptide of the bacteriophage T4 clamp also functions as a transcrip- RB69 polymerase and the carboxyl terminus of p21CIP1, a tional activator of RNA polymerase. The RNA polymerase cell-cycle regulator that binds the human clamp, PCNA, binding proteins, gp33 (a co-activator of the RNA poly- and inhibits DNA replication [11,12]. The carboxy-termi- merase) and gp55 (a sigma factor for late genes) also bind nal peptide of p21CIP1 binds PCNA at a hydrophobic the T4 clamp, which diffuses along DNA and thus facili- pocket that is analogous, in structure and position, to the tates rapid localization of the transcription complex at pro- hydrophobic DNA-polymerase-binding pocket on the moter sites [16]. Incidentally, gp33 and gp55 also bind the RB69 clamp (Figure 2c) [4]. The p21CIP1 peptide is T4 clamp through the consensus sequence of hydrophobic known to compete directly with DNA polymerase δ for amino acids at their carboxyl termini [17]. binding to PCNA [13], implying that the eukaryotic poly- merase binds its clamp at the same position and perhaps The eukaryotic sliding clamp, PCNA, also appears to be in the same manner as the phage RB69 DNA polymerase. used extensively as a DNA-tracking protein by other Thus it would appear that such an interaction between enzymes. We mentioned earlier that the cell-cycle regula- DNA polymerase and its processivity factor, tailored to fit tory protein p21CIP1 binds PCNA through a carboxy- a concise site via a flexible connector, is typical of both terminal peptide, and inhibits DNA replication by prokaryotic and eukaryotic DNA replicating enzymes. competing with DNA polymerase for the sliding clamp [13]. In recent years, numerous proteins — including Intriguingly, the recently determined structure of DNA ligase I, the Fen1 and XPG endonucleases, and another DNA replisomal , the bacteriophage T7 DNA (cytosine-5) methyl transferase — have been found DNA , has revealed a carboxy-terminal tail pro- to interact with PCNA in a manner that competes with truding from the surface of each subunit of the ring- p21CIP1 binding [18]. All these proteins contain a small, shaped hexamer [14]. Earlier studies showed that the conserved sequence of hydrophobic amino acids [18], the carboxyl terminus of the helicase is essential for its inter- PCNA-binding motif, that matches part of the p21CIP1 action with the T7 DNA polymerase [15]. These find- carboxy-terminal peptide sequence and likely binds the ings imply that other proteins at the DNA replication hydrophobic patch on the clamp shown in Figure 2b. bb10a07.qxd 02/05/2000 10:57 Page R28

R28 Current Biology Vol 10 No 1

Figure 3 subunit of the trimeric ring — it would be in place to swiftly snip 5′ flap structures before strand displacement proceeds too far. There is no experimental evidence yet for simultaneous binding of multiple proteins to PCNA, and the model structure of the RB69 replisome indicates that, although the polymerase interacts with only one subunit of the trimeric clamp, it virtually occludes the binding sites on the remaining two subunits. But it is tempting to speculate that interaction through flexible connector domains might allow more than one protein to bind the sliding clamp at the same time.

The ability of p21CIP1 to compete with multiple enzymes for the sliding clamp also suggests possible pathways for the cell-cycle control of DNA metabolism. Changing levels of p21CIP1 during the cell cycle may alternately inhibit or facilitate interactions between various enzymes and the clamp, and thereby modulate their activity. Although the fine details of p21CIP1-mediated control of DNA metabolism are not clear yet, this subject is the focus of intense ongoing research in several laboratories.

The presence of the sliding-clamp-binding site on so many functionally unrelated enzymes indicates that this compact motif transfers easily between genes, and that a protein might quickly pick up the ability to associate with a clamp (and with DNA) without significant perturbation of its structure or function. The evolution of secondary clamp-binding sites could further refine the mechanism by which each enzyme uses and, in turn, is modulated by the sliding clamp. In summary, an important consequence of the consensus clamp-binding motif is that the same protein can be used in several different pathways to mod- ulate the activity of several different enzymes. The Oh what a tangled web we create, when we begin to replicate. During common link likely aids communication and coordination coordinated leading-strand and lagging-strand replication, the leading between various metabolic pathways and may serve as a DNA strand undergoes torsional stress as it cannot swivel freely behind the polymerase (1→2). If the leading polymerase transiently releases the focal point for the regulation of these pathways. primer–template, while remaining bound to the clamp (2→3), the DNA can swivel back, within the clamp, to its original configuration (3→1). Acknowledgements We are grateful for Nicholas Cozzarelli’s insight into resolution of the tor- sional stress problem at the replication fork, and thank Yousif Shamoo and T.A. Steitz for providing the images for Figure 2, as well as David Jeruzalmi Thus, all these enzymes can apparently compete with for preparing Figure 1. This work was supported by NIH grant GM38839. each other, and presumably with DNA polymerase, for binding to the sliding clamp. References 1. Kuriyan J, O’Donnell M: Sliding clamps of DNA polymerases. J Mol Biol 1993, 234:915-925. At the very least, the DNA metabolizing enzymes might 2. Kong XP, Onrust R, O’Donnell M, Kuriyan J: Three-dimensional β use PCNA as a mobile scaffold to scan the duplex and structure of the subunit of E. coli DNA polymerase III holoenzyme: a sliding DNA clamp. Cell 1992, 69:425-437. rapidly locate their sites of action. In a more complex sce- 3. Krishna TS, Kong XP, Gary S, Burgers PM, Kuriyan J: Crystal nario, PCNA could modulate the catalytic activity of structure of the eukaryotic DNA polymerase processivity factor PCNA. Cell 1994, 79:1233-1243. enzymes by coordinating their timely assembly at target 4. Gulbis J, Kelman Z, Hurwitz J, O’Donnell M, Kuriyan J: Structure of sites on DNA. For example, the 1, the C-terminal region of p21WAF1/CIP1 complexed with human Fen1, processes the generated during PCNA. Cell 1996, 87:297-306. 5. Moarefi I, Jeruzalmi D, Turner J, O’Donnell M, Kuriyan J: Crystal lagging-strand DNA replication by cleaving the 5′ flap structure of the DNA polymerase processivity factor of T4 formed if DNA polymerase catalyzes strand displacement bacteriophage. J Mol Biol, in press. 6. Shamoo Y, Steitz TA: Building a replisome from interacting pieces: synthesis [19]. If Fen1 associates with PCNA in the DNA sliding clamp complexed to a peptide from DNA polymerase and replisome — perhaps through a flexible connection with a a polymerase editing complex. Cell 1999, 99:155-166. bb10a07.qxd 02/05/2000 10:57 Page R29

Dispatch R29

7. Wang J, Sattar AKM, Wang CC, Karam JD, Konigsberg WH, Steitz TA: Crystal structure of a pol α family replication DNA polymerase from bacteriophage RB69. Cell 1995, 89:1087-1099. 8. O’Donnell M, Kornberg A: Dynamics of DNA polymerase III If you found this dispatch interesting, you might also want holoenzyme of E. coli in replication of a multiprimed template. to read the February 1999 issue of J Biol Chem 1985, 260:12875-12883. 9. Berdis AJ, Soumillion P, Benkovic SJ: The carboxyl terminus of the bacteriophage T4 DNA polymerase is required for holoenzyme Current Opinion in complex formation. Proc Natl Acad Sci USA 1996, 93:12822-12827. Structural Biology 10. Digard P, Bebrin WR, Weisshart K, Coen DM: The extreme C terminus of herpes simplex DNA polymerase is crucial for which included the following reviews, edited functional interaction with processivity factor UL42 and for viral replication. J Virol 1993, 67:398-406. by Simon EV Phillips and Dino Moras, on 11. Flores-Rozas H, Kelman Z, Dean FB, Pan Z-Q, Harper JW, Elledge SJ, Protein–nucleic acid interactions: O’Donnell M, Hurwitz J: Cdk-interacting protein 1 directly binds with proliferating cell nuclear antigen and inhibits DNA replication Structure and mechanism in site-specific recombination catalyzed by the DNA polymerase δ holoenzyme. Proc Natl Acad Sci USA 1994, 91:8655-8659. Deshmukh N Gopaul and Gregory D Van Duyne 12. Waga S, Hannon GJ, Beach D, Stillman B: The p21 inhibitor of Getting a grip: polymerases and their cyclin-dependent kinases controls DNA replication by interaction with PCNA. Nature 1994, 369:574-578. substrate complexes 13. Gibbs E, Kelman Z, Gulbis JM, O’Donnell M, Kuriyan J, Burgers PMJ, Joachim Jäger and Janice D Pata Hurwitz J: The influence of proliferating cell nuclear antigen- interacting domain of p21CIP1 on DNA synthesis catalyzed by the Structural insights into the function of type human and S. cerevisiae polymerase δ holoenzymes. J Biol Chem IB 1997, 272:2373-2381. Matthew R Redinbo, James J Champoux and Wim GJ Hol 14. Sawaya MR, Guo S, Tabor S, Richardson CC, Ellenberger T: Crystal structure of the helicase domain from the replicative helicase- Envisioning the molecular choreography of DNA base of bacteriophage T7. Cell 1999, 99:167-177. excision repair 15. Notarnicola SM, Mulcahy HL, Lee J, Richardson CC: The acidic Sudip S Parikh, Clifford D Mol, David J Hosfield carboxyl terminus of the bacteriophage T7 gene 4 and John A Tainer helicase/primase interacts with T7 DNA polymerase. J Biol Chem 1997, 272:18425-18433. Combinatorial gene regulation by eukaryotic 16. Geiduschek EP, Fu T-J, Kassavetis GA, Sanders GM, Tinker-Kulberg transcription factors RL: Transcriptional activation by a topologically linkable protein: forging a connection between replication and gene activity. In Lin Chen Nucleic Acids and Molecular Biology, Vol 11. Edited by Eckstein F, Lilley DMJ. Heidelberg: Springer-Verlag; 1997:135-150. 17. Wong K, Geiduschek EP: Activator-sigma interaction: a Marc O'Reilly, Sarah A Teichmann and Daniela Rhodes hydrophobic segment mediates the interaction of a sigma family RNA–protein complexes promoter recognition protein with a sliding clamp transcription activator. J Mol Biol 1998, 284:195-203. Stephen Cusack 18. Warbrick E: PCNA binding through a conserved motif. BioEssays 1998, 20:195-199. 19. Lieber MR: The FEN-1 family of structure specific nucleases in the same issue also included the following eukaryotic DNA replication, recombination, and repair. BioEssays reviews, edited by Christopher M Dobson 1997, 19:233-240. and Oleg B Ptitsyn, on Folding and binding:

The fundamentals of : bringing together theory and experiment Christopher M Dobson and Martin Karplus Principles of protein folding in the cellular environment R John Ellis and F Ulrich Hartl Co-translational folding Boyd Hardesty, Tamara Tsalkova and Gisela Kramer Membrane protein folding Paula J Booth and A Rachael Curran Folding of peptide models of collagen and misfolding in disease Jean Baum and Barbara Brodsky Virus assembly Lars Liljas The full text of Current Opinion in Structural Biology is in the BioMedNet library at http://BioMedNet.com/cbiology/stb