Subunit sharing among high- and low-fidelity DNA polymerases

3 downloads 0 Views 524KB Size Report
Jul 31, 2012 - Lance D. Langston and Mike O'Donnell1. Howard Hughes Medical Institute, The Rockefeller University, New York, NY 10065. To preserve ...
Subunit sharing among high- and low-fidelity DNA polymerases Lance D. Langston and Mike O’Donnell1 Howard Hughes Medical Institute, The Rockefeller University, New York, NY 10065

T

o preserve genetic information through generations, DNA must be duplicated with precision, and thus DNA polymerases (Pols) that replicate chromosomes are highly accurate. However, DNA is constantly bombarded by damage and undergoing repair, and, if the replication fork encounters a damaged base before repair, highly accurate replicases stall and cannot continue fork progression. All cells contain specialized translesion (TLS) Pols that have low accuracy but can extend DNA over template lesions (1, 2). Although some TLS Pols have evolved to insert the correct nucleotide opposite particular lesions, they can also be mutagenic by inserting an incorrect base. The conservation of TLS Pols in all cells suggests that single-base mutations may offer a better outcome than fork collapse, cell death, or recombination that results in gross chromosome abnormalities. However, there has been longstanding confusion regarding the participation of the high-fidelity eukaryotic replicase Pol δ in the mutagenic TLS process, a dilemma that is largely resolved by data presented in PNAS showing sharing of subunits between Pol δ and TLS Pol ζ (3). TLS can be error-free or error-prone, and can thus prevent or contribute to cancer. Yeast have TLS Pols η, ζ, and Rev1, and mammalian cells contain yet other TLS Pols (1). TLS Pol η can be mutagenic but is error-free in extending DNA over T-T cis-syn dimers, a common UV lesion, and mutations in Pol η confer sensitivity to UV radiation and underlie cancer disposition in the variant form of Xeroderma pigmentosum syndrome. Pol ζ is not itself error-prone, but it promotes mutations by extending DNA chains from mispaired bases after a low-fidelity Pol inserts the incorrect nucleotide across from a template lesion (4). Genetic data show that the high-fidelity replicase Pol δ also participates in errorprone DNA repair, particularly that induced by DNA damaging agents (5). Further investigation in yeast demonstrated that mutations that affect the catalytic or proofreading functions of Pol δ interfere only with bulk DNA replication, not DNA repair, whereas those that leave the Pol activity intact but weaken the interaction between the catalytic subunit of Pol δ (Pol3) and its essential accessory subunit (Pol31) show defects in DNA

Fig. 1. Subunit organization of eukaryotic B-family Pols. The Pol catalytic subunit is designated the A subunit (red); a flexible linker connects it to the metal-binding C-terminal domain (CTD). The B subunit (green) binds the CTD of the A subunit. Other subunits bind the A or B subunits as indicated (the p32 subunit that binds the B subunit in Pols δ and ζ is referred to as the C subunit in the diagram). Each subunit is labeled according to the nomenclature in yeast, and the human designation is in parentheses. The figure is adapted from figure 1 of Johansson and Macneill (12).

replication and damage-induced DNA repair (5, 6). Deletion of the gene encoding a nonessential Pol δ subunit in yeast, Pol32, also confers sensitivity to DNAdamaging agents and renders the cells defective for damage-induced mutagenesis, further implicating Pol δ in error-prone DNA repair (7). This confusing situation—participation of a high-fidelity Pol in a low-fidelity repair process—receives a dose of clarity from a PNAS study showing that the accessory subunits of Pol δ in yeast, Pol31 and Pol32, are also functionally essential subunits of Pol ζ (3). These findings confirm and extend the observations from another recent publication showing interaction between the analogous subunits of human Pol δ, p50/p66, and the catalytic subunit of human Pol ζ (8). They also provide a gratifyingly straightforward explanation for why mutations in these subunits, which were previously thought only to act in the context of Pol δ, also impair an error-prone DNA repair process (7). Furthermore, these studies compel a reevaluation of numerous experiments

12268–12269 | PNAS | July 31, 2012 | vol. 109 | no. 31

using mutants of the nonessential yeast POL32 gene as a proxy for Pol δ function, as well as those using mutants that disrupt the interaction between the catalytic subunit of Pol δ and its accessory subunits without affecting its Pol activity. That Pol 32 is a subunit of Pol ζ was foreshadowed by recent genetic and structural data. Pol32 is required for Pol ζ-dependent TLS (9). The requirement of Pol ζ as well as Pol32 implied that participation of Pol δ in TLS may be indirect. The authors proposed that Pol32 facilitates a switch from Pol δ to Pol ζ when Pol δ encounters a lesion. Earlier studies also suggested that Pol δ plays only an indirect role in TLS mutagenesis, with Pol ζ acting as the primary TLS Pol (10, 11). Structurally, although it has a unique ability to extend DNA chains from mispaired ends (4), Pol ζ is a B-family Pol, and the other three B-family Pols in eukaryotes—Pol α, Pol ε, and Pol δ—are collectively responsible for chromosomal DNA replication, an extremely highfidelity process. The catalytic subunits of the three replicative Pols bind through a metal-binding C-terminal domain (CTD) to an essential accessory subunit, referred to as the B subunit, that is part of a family of closely related proteins (12) (Fig. 1). Pol ζ also has a CTD that shares close sequence homology with the CTD of Pol δ, and both appear to bind a 4Fe–4S cluster instead of Zn (8, 13). Despite the similarities between the CTDs of Pols δ and ζ, previous attempts to identify an interaction between Pol ζ and the accessory subunits of Pol δ by using the two-hybrid system were unsuccessful (6), so the identity of the missing Pol ζ B subunit, if any, remained unresolved. In the study of Johnson et al., co-overexpression in yeast of a tagged version of Rev3 protein—the catalytic subunit of Pol ζ—along with Pol31, Pol32, and Rev7, a previously identified accessory subunit important for the function of Rev3, produced a highly stable foursubunit complex (3). Directed mutations of the yeast REV3 gene that were shown to decrease Rev3 binding to Pol31 in vitro

Author contributions: L.D.L. and M.O. wrote the paper. The authors declare no conflict of interest. See companion article on page 12455. 1

To whom correspondence should be addressed. E-mail: [email protected].

www.pnas.org/cgi/doi/10.1073/pnas.1209533109

switching (15). In eukaryotes, Pols δ, ε, and ζ bind the PCNA clamp (as do other TLS Pols), and one may presume they switch by using a similar strategy to bacterial Pols. However, TLS Pol switching in eukaryotes appears to involve many other levels of control. For example, the PCNA clamp is monoubiquitinated by Rad6-Rad18 in response to DNA damage, and this modification is important for TLS (2). The report by Baranovskiy et al. proposes a unique type of switch mechanism (8). They suggest that the Pol 31/32 subunits facilitate Pol switching by remaining bound to PCNA whereas the Pol δ catalytic subunit is ejected upon stalling at a lesion, vacating Pols 31/32 for the Rev3/7 complex to bind. This “subunit exchange” model for Pol switching is proposed to be triggered by conformational change caused by Pol δ stalling, a change in the redox state of the iron in the CTD, and/or ubiquitination of PCNA (8). Although a proposal of subunit exchange between tightly associated complexes may

1. Waters LS, et al. (2009) Eukaryotic translesion polymerases and their roles and regulation in DNA damage tolerance. Microbiol Mol Biol Rev 73:134–154. 2. Pavlov YI, Shcherbakova PV, Rogozin IB (2006) Roles of DNA polymerases in replication, repair, and recombination in eukaryotes. Int Rev Cytol 255:41–132. 3. Johnson RE, Prakash L, Prakash S (2012) Pol31 and Pol32 subunits of yeast DNA polymerase δ are also essential subunits of DNA polymerase ζ. Proc Natl Acad Sci USA 109:12455–12460. 4. Prakash S, Johnson RE, Prakash L (2005) Eukaryotic translesion synthesis DNA polymerases: Specificity of structure and function. Annu Rev Biochem 74:317–353. 5. Giot L, Chanet R, Simon M, Facca C, Faye G (1997) Involvement of the yeast DNA polymerase delta in DNA repair in vivo. Genetics 146:1239–1251. 6. Sanchez Garcia J, Ciufo LF, Yang X, Kearsey SE, MacNeill SA (2004) The C-terminal zinc finger of the catalytic subunit of DNA polymerase delta is responsible

for direct interaction with the B-subunit. Nucleic Acids Res 32:3005–3016. Gerik KJ, Li X, Pautz A, Burgers PM (1998) Characterization of the two small subunits of Saccharomyces cerevisiae DNA polymerase delta. J Biol Chem 273:19747–19755. Baranovskiy AG, et al. (2012) DNA polymerase δ and ζ Switch by sharing accessory subunits of DNA polymerase δ. J Biol Chem 287:17281–17287. Hanna M, Ball LG, Tong AH, Boone C, Xiao W (2007) Pol32 is required for Pol zeta-dependent translesion synthesis and prevents double-strand breaks at the replication fork. Mutat Res 625:164–176. Gibbs PE, McDonald J, Woodgate R, Lawrence CW (2005) The relative roles in vivo of Saccharomyces cerevisiae Pol eta, Pol zeta, Rev1 protein and Pol32 in the bypass and mutation induction of an abasic site, T-T (6-4) photoadduct and T-T cis-syn cyclobutane dimer. Genetics 169:575–582. Huang ME, de Calignon A, Nicolas A, Galibert F (2000) POL32, a subunit of the Saccharomyces cerevisiae DNA

Langston and O’Donnell

The accessory subunits of Pol δ in yeast, Pol31 and Pol32, are also functionally essential subunits of Pol ζ.

7.

8.

9.

10.

11.

seem unlikely and does not occur in bacterial Pol switching, it cannot be excluded. However, the subunit exchange proposal does not explain how Pols ε/ζ switch for TLS bypass on the leading strand, as the leading strand replicase Pol ε does not contain Pols 31/32. The finding that Pols 31/32 are functional components of Pol ζ brings its subunit composition to four, the most complex of any TLS Pol to date. Pol ζ also interacts with Rev1, a scaffold for numerous TLS Pols, indicating that the situation in vivo is even more intricate and complex (1). Most importantly, these findings clarify the genetic data suggesting a role for Pol δ in damage-induced mutagenesis by showing that Pols 31/32 are critical for Pol ζ function as well. At the same time, recent data indicate that Pol δ is error-prone during double-strand break repair, and the phenotype is directly attributable to the catalytic subunit (16, 17). Pol δ acts as a leading strand Pol during double-strand break repair, and the fact that it is error-prone in that role may explain why it is mostly confined to the lagging strand during DNA replication. In any case, recent findings remind us that there is still a great deal to learn about the dynamics of DNA replication in eukaryotes. Given the fundamental role of mutation in tumorigenesis, the effort could not be more urgent or important. ACKNOWLEDGMENTS. Replication studies in the authors’ laboratory are supported by National Institutes of Health Grant GM39939 and by the Howard Hughes Medical Institute.

12.

13.

14.

15.

16.

17.

polymerase delta, defines a link between DNA replication and the mutagenic bypass repair pathway. Curr Genet 38:178–187. Johansson E, Macneill SA (2010) The eukaryotic replicative DNA polymerases take shape. Trends Biochem Sci 35:339–347. Netz DJ, et al. (2012) Eukaryotic DNA polymerases require an iron-sulfur cluster for the formation of active complexes. Nat Chem Biol 8:125–132. Baranovskiy AG, et al. (2008) X-ray structure of the complex of regulatory subunits of human DNA polymerase delta. Cell Cycle 7:3026–3036. Sutton MD (2010) Coordinating DNA polymerase traffic during high and low fidelity synthesis. Biochim Biophys Acta 1804:1167–1179. Hicks WM, Kim M, Haber JE (2010) Increased mutagenesis and unique mutation signature associated with mitotic gene conversion. Science 329:82–85. Deem A, et al. (2011) Break-induced replication is highly inaccurate. PLoS Biol 9:e1000594.

PNAS | July 31, 2012 | vol. 109 | no. 31 | 12269

COMMENTARY

until absolutely needed. In bacterial systems, the replicase and TLS Pols bind the β-processivity clamp, which organizes Pol

also showed increased sensitivity in vivo to UV damage and a reduction in damageinduced mutagenesis, establishing the functional significance of Rev3 binding to Pol31 (3). In the case of the human proteins, a similar interaction was demonstrated by coexpressing a tagged version of the human Pol ζ CTD with the orthologous subunits of human Pols α, δ, and ε in Escherichia coli. The tagged protein bound to the p50/p66 subunits of human Pol δ (equivalent to Pols 31/32 in yeast) but not to the analogous subunits of Pol α or Pol ε, indicating a direct and specific interaction (8). Mutations of conserved sites in the yeast REV3 gene were used to show that disruption of the newly identified iron-sulfur (i.e., 4Fe–4S) cluster in the Pol ζ CTD also conferred sensitivity to UV damage and impaired damageinduced mutagenesis, reinforcing the functional importance of accessory subunit interactions mediated by the Rev3 CTD (8). The precise function of the B subunits and most other Pol accessory proteins is largely unknown, although a crystal structure of the Pol31/32 subunits of yeast Pol δ revealed several well-defined domains, including an oligonucleotide/ oligosaccharide-binding fold, which may support DNA binding by the catalytic subunit Pol3 (14). Pol32 binds to numerous proteins, including the Srs2 helicase and Pol α (11), suggesting that it may act as a scaffold, like the Rev1 protein, to organize distinct DNA transactions during replication (2). Little is known about how each Pol is targeted to its site of action, and, because they are error-prone, TLS Pols like Pol ζ must be kept from DNA