An updated perspective on the polymerase division of labor during eukaryotic DNA replication

Abstract

In eukaryotes three DNA polymerases (Pols), α, δ, and ε, are tasked with bulk DNA synthesis of nascent strands during genome duplication. Most evidence supports a model where Pol α initiates DNA synthesis before Pol ε and Pol δ replicate the leading and lagging strands, respectively. However, a number of recent reports, enabled by advances in biochemical and genetic techniques, have highlighted emerging roles for Pol δ in all stages of leading-strand synthesis; initiation, elongation, and termination, as well as fork restart. By focusing on these studies, this review provides an updated perspective on the division of labor between the replicative polymerases during DNA replication.     

The high fidelity and unique error signature of human DNA polymerase epsilon

Bulk replicative DNA synthesis in eukaryotes is highly accurate and efficient, primarily because of two DNA polymerases (Pols): Pols δ and ε. The high fidelity of these enzymes is due to their intrinsic base selectivity and proofreading exonuclease activity which, when coupled with post-replication mismatch repair, helps to maintain human mutation rates at less than one mutation per genome duplication. Conditions that reduce polymerase fidelity result in increased mutagenesis and can lead to cancer in mice. Whereas yeast Pol ε has been well characterized, human Pol ε remains poorly understood. Here, we present the first report on the fidelity of human Pol ε. We find that human Pol ε carries out DNA synthesis with high fidelity, even in the absence of its 3'→5' exonucleolytic proofreading and is significantly more accurate than yeast Pol ε. Though its spectrum of errors is similar to that of yeast Pol ε, there are several notable exceptions. These include a preference of the human enzyme for T→A over A→T transversions. As compared with other replicative DNA polymerases, human Pol ε is particularly accurate when copying homonucleotide runs of 4-5 bases. The base pair substitution specificity and high fidelity for frameshift errors observed for human Pol ε are distinct from the errors made by human Pol δ.     

Comparison of the kinetic parameters of the truncated catalytic subunit and holoenzyme of human DNA polymerase ɛ

Numerous genetic studies have provided compelling evidence to establish DNA polymerase ɛ (Polɛ) as the primary DNA polymerase responsible for leading strand synthesis during eukaryotic nuclear genome replication. Polɛ is a heterotetramer consisting of a large catalytic subunit that contains the conserved polymerase core domain as well as a 3′ → 5′ exonuclease domain common to many replicative polymerases. In addition, Polɛ possesses three small subunits that lack a known catalytic activity but associate with components involved in a variety of DNA replication and maintenance processes. Previous enzymatic characterization of the Polɛ heterotetramer from budding yeast suggested that the small subunits slightly enhance DNA synthesis by Polɛ in vitro. However, similar studies of the human Polɛ heterotetramer (hPolɛ) have been limited by the difficulty of obtaining hPolɛ in quantities suitable for thorough investigation of its catalytic activity. Utilization of a baculovirus expression system for overexpression and purification of hPolɛ from insect host cells has allowed for isolation of greater amounts of active hPolɛ, thus enabling a more detailed kinetic comparison between hPolɛ and an active N-terminal fragment of the hPolɛ catalytic subunit (p261N), which is readily overexpressed in Escherichia coli. Here, we report the first pre-steady-state studies of fully-assembled hPolɛ. We observe that the small subunits increase DNA binding by hPolɛ relative to p261N, but do not increase processivity during DNA synthesis on a single-stranded M13 template. Interestingly, the 3′ → 5′ exonuclease activity of hPolɛ is reduced relative to p261N on matched and mismatched DNA substrates, indicating that the presence of the small subunits may regulate the proofreading activity of hPolɛ and sway hPolɛ toward DNA synthesis rather than proofreading.     

DNA polymerases eta and kappa exchange with the polymerase delta holoenzyme to complete common fragile site synthesis

Common fragile sites (CFSs) are inherently unstable genomic loci that are recurrently altered in human tumor cells. Despite their instability, CFS are ubiquitous throughout the human genome and associated with large tumor suppressor genes or oncogenes. CFSs are enriched with repetitive DNA sequences, one feature postulated to explain why these loci are inherently difficult to replicate, and sensitive to replication stress. We have shown that specialized DNA polymerases (Pols) η and κ replicate CFS-derived sequences more efficiently than the replicative Pol δ. However, we lacked an understanding of how these enzymes cooperate to ensure efficient CFS replication. Here, we designed a model of lagging strand replication with RFC loaded PCNA that allows for maximal activity of the four-subunit human Pol δ holoenzyme, Pol η, and Pol κ in polymerase mixing assays. We discovered that Pol η and κ are both able to exchange with Pol δ stalled at repetitive CFS sequences, enhancing Normalized Replication Efficiency. We used this model to test the impact of PCNA mono-ubiquitination on polymerase exchange, and found no change in polymerase cooperativity in CFS replication compared with unmodified PCNA. Finally, we modeled replication stress in vitro using aphidicolin and found that Pol δ holoenzyme synthesis was significantly inhibited in a dose-dependent manner, preventing any replication past the CFS. Importantly, Pol η and κ were still proficient in rescuing this stalled Pol δ synthesis, which may explain, in part, the CFS instability phenotype of aphidicolin-treated Pol η and Pol κ-deficient cells. In total, our data support a model wherein Pol δ stalling at CFSs allows for free exchange with a specialized polymerase that is not driven by PCNA.     

The tail that wags the dog: p12, the smallest subunit of DNA polymerase δ, is degraded by ubiquitin ligases in response to DNA damage and during cell cycle progression

DNA polymerase δ (Pol δ) is a key enzyme in eukaryotic DNA replication. Human Pol δ is a heterotetramer whose p12 subunit is degraded in response to DNA damage, leading to the in vivo conversion of Pol δ4 to Pol δ3. Two E3 ubiquitin ligases, RNF8 and CRL4^Cdt2, participate in the DNA damage-induced degradation of p12. We discuss how these E3 ligases integrate the formation of Pol δ3 and ubiquitinated PCNA for DNA repair processes. CRL4^Cdt2 partially degrades p12 during normal cell cycle progression, thereby generating Pol δ3 during S phase. This novel finding extends the current view of the role of Pol δ3 in DNA repair and leads to the hypothesis that it participates in DNA replication. The coordinated regulation of licensing factors and Pol δ3 by CRL4^Cdt2 now opens new avenues for control of DNA replication. A parallel study of Pol δ4 and Pol δ3 in Okazaki fragment processing provides evidence for a role of Pol δ3 in DNA replication. We discuss several new perspectives of the role of the 2 forms of Pol δ in DNA replication and repair, as well the significance of the integration of p12 regulation in DNA repair and cell cycle progression.     

Translesion synthesis across abasic lesions by human B-family and Y-family DNA polymerases α, δ, η, ι, κ, and REV1

Abasic (apurinic/apyrimidinic, AP) sites are the most common DNA lesions formed in cells, induce severe blocks to DNA replication, and are highly mutagenic. Human Y-family translesion DNA polymerases (pols) such as pols η, ι, κ, and REV1 have been suggested to play roles in replicative bypass across many DNA lesions where B-family replicative pols stall, but their individual catalytic functions in AP site bypass are not well understood. In this study, oligonucleotides containing a synthetic abasic lesion (tetrahydrofuran analogue) were compared for catalytic efficiency and base selectivity with human Y-family pols η, ι, κ, and REV1 and B-family pols α and δ. Pol η and pol δ/proliferating cell nuclear antigen (PCNA) copied past AP sites quite effectively and generated products ranging from one-base to full-length extension. Pol ι and REV1 readily incorporated one base opposite AP sites but then stopped. Pols κ and α were severely blocked at AP sites. Pol η preferentially inserted T and A; pol ι inserted T, G, and A; pol κ inserted C and A; REV1 preferentially inserted C opposite AP sites. The B-family pols α and δ/PCNA preferentially inserted A (85% and 58%, respectively) consonant with the A-rule hypothesis. Pols η and δ/PCNA were much more efficient in next-base extension, preferably from A positioned opposite an AP site, than pol κ. These results suggest that AP sites might be bypassed with moderate efficiency by single B- and Y-family pols or combinations, possibly by REV1 and pols ι, η, and δ/PCNA at the insertion step opposite the lesion and by pols η and δ/PCNA at the subsequent extension step. The patterns of the base preferences of human B-family and Y-family pols in both insertion and extension are pertinent to some of the mutagenesis events induced by AP lesions in human cells.     

Differential arrival of leading and lagging strand DNA polymerases at fission yeast telomeres

To maintain genomic integrity, telomeres must undergo switches from a protected state to an accessible state that allows telomerase recruitment. To better understand how telomere accessibility is regulated in fission yeast, we analysed cell cycle‐dependent recruitment of telomere‐specific proteins (telomerase Trt1, Taz1, Rap1, Pot1 and Stn1), DNA replication proteins (DNA polymerases, MCM, RPA), checkpoint protein Rad26 and DNA repair protein Nbs1 to telomeres. Quantitative chromatin immunoprecipitation studies revealed that MCM, Nbs1 and Stn1 could be recruited to telomeres in the absence of telomere replication in S‐phase. In contrast, Trt1, Pot1, RPA and Rad26 failed to efficiently associate with telomeres unless telomeres are actively replicated. Unexpectedly, the leading strand DNA polymerase ε (Polε) arrived at telomeres earlier than the lagging strand DNA polymerases α (Polα) and δ (Polδ). Recruitment of RPA and Rad26 to telomeres matched arrival of DNA Polε, whereas S‐phase specific recruitment of Trt1, Pot1 and Stn1 matched arrival of DNA Polα. Thus, the conversion of telomere states involves an unanticipated intermediate step where lagging strand synthesis is delayed until telomerase is recruited.     

Replication of ribonucleotide-containing DNA templates by yeast replicative polymerases

The major replicative DNA polymerases of S. cerevisiae (Pols α, δ, and ?) incorporate substantial numbers of ribonucleotides into DNA during DNA synthesis. When these ribonucleotides are not removed in vivo, they reside in the template strand used for the next round of replication and could potentially reduce replication efficiency and fidelity. To examine if the presence of ribonucleotides in a DNA template impede DNA synthesis, we determined the efficiency with which Pols α, δ, and ? copy DNA templates containing a single ribonucleotide. All three polymerases can replicate past ribonucleotides. Relative to all-DNA templates, bypass of ribo-containing templates is slightly reduced, to extents that depend on the identity of the ribo and the sequence context in which it resides. Bypass efficiencies for Pols δ and ? were increased by increasing the dNTP concentrations to those induced by cellular stress, and in the case of Pol ?, by inactivating the 3'-exonuclease activity. Overall, ribonucleotide bypass efficiencies are comparable to, and usually exceed, those for the common oxidative stress-induced lesion 8-oxo-guanine.     

Two forms of human DNA polymerase δ: Who does what and why?

DNA polymerase δ (Pol δ) plays a central role in lagging strand DNA synthesis in eukaryotic cells, as well as an important role in DNA repair processes. Human Pol δ4 is a heterotetramer of four subunits, the smallest of which is p12. Pol δ3 is a trimeric form that is generated in vivo by the degradation of the p12 subunit in response to DNA damage, and during entry into S-phase. The biochemical properties of the two forms of Pol δ, as well as the changes in their distribution during the cell cycle, are reviewed from the perspective of understanding their respective cellular functions. Biochemical and cellular studies support a role for Pol δ3 in gap filling during DNA repair, and in Okazaki fragment synthesis during DNA replication. Recent studies of cells in which p12 expression is ablated, and are therefore null for Pol δ4, show that Pol δ4 is not required for cell viability. These cells have a defect in homologous recombination, revealing a specific role for Pol δ4 that cannot be performed by Pol δ3. Pol δ4 activity is required for D-loop displacement synthesis in HR. The reasons why Pol δ4 but not Pol δ3 can perform this function are discussed, as well as the question of whether helicase action is needed for efficient D-loop displacement synthesis. Pol δ4 is largely present in the G1 and G2/M phases of the cell cycle and is low in S phase. This is discussed in relation to the availability of Pol δ4 as an additional layer of regulation for HR activity during cell cycle progression.     

Ribonucleotide incorporation,proofreading and bypass by human DNA polymerase δ

In both budding and fission yeast, a large number of ribonucleotides are incorporated into DNA during replication by the major replicative polymerases (Pols α, δ and ?). They are subsequently removed by RNase H2-dependent repair, which if defective leads to replication stress and genome instability. To extend these studies to humans, where an RNase H2 defect results in an autoimmune disease, here we compare the ability of human and yeast Pol δ to incorporate, proofread, and bypass ribonucleotides during DNA synthesis. In reactions containing nucleotide concentrations estimated to be present in mammalian cells, human Pol δ stably incorporates one rNTP for approximately 2000 dNTPs, a ratio similar to that for yeast Pol δ. This result predicts that human Pol δ may introduce more than a million ribonucleotides into the nuclear genome per replication cycle, an amount recently reported to be present in the genome of RNase H2-defective mouse cells. Consistent with such abundant stable incorporation, we show that the 3′-exonuclease activity of yeast and human Pol δ largely fails to edit ribonucleotides during polymerization. We also show that, like yeast Pol δ, human Pol δ pauses as it bypasses ribonucleotides in DNA templates, with four consecutive ribonucleotides in a DNA template being more problematic than single ribonucleotides. In conjunction with recent studies in yeast and mice, this ribonucleotide incorporation may be relevant to impaired development and disease when RNase H2 is defective in mammals. As one tool to investigate ribonucleotide incorporation by Pol δ in human cells, we show that human Pol δ containing a Leu606Met substitution in the polymerase active site incorporates 7-fold more ribonucleotides into DNA than does wild type Pol δ.     

Spatiotemporal recruitment of human DNA polymerase delta to sites of UV damage

Human DNA polymerase δ (Pol δ) is involved in various DNA damage responses in addition to its central role in DNA replication. The Pol δ4 holoenzyme consists of four subunits, p125, p50, p68 and p12. It has been established that the p12 subunit is rapidly degraded in response to DNA damage by UV leading to the in vivo conversion of Pol δ4 to Pol δ3, a trimeric form lacking the p12 subunit. We provide the first analysis of the time-dependent recruitment of the individual Pol δ subunits to sites of DNA damage produced by UV irradiation through 5 μm polycarbonate filters by immunofluorescence microscopy and laser scanning cytometry (LSC). Quantitative analysis demonstrates that the recruitments of the three large subunits was near complete by 2 h and did not change significantly up to 4 h after UV exposure. However, the recruitment of p12 was incomplete even at 4 h, with about 70% of the Pol δ lacking the p12 subunit. ChIP analysis of Pol δ after global UV irradiation further demonstrates that only p125, p50 and p68 were present. Thus, Pol δ3 is the predominant form of Pol δ at sites of UV damage as a result of p12 degradation. Using LSC, we have further confirmed that Pol δ was recruited to CPD damage sites in all phases of the cell cycle. Collectively, our results show that Pol δ at the DNA damage site is the Pol δ trimer lacking p12 regardless of the cell cycle phase.     

The Werner syndrome exonuclease facilitates DNA degradation and high fidelity DNA polymerization by human DNA polymerase δ

DNA Polymerase δ (Pol δ) and the Werner syndrome protein, WRN, are involved in maintaining cellular genomic stability. Pol δ synthesizes the lagging strand during replication of genomic DNA and also functions in the synthesis steps of DNA repair and recombination. WRN is a member of the RecQ helicase family, loss of which results in the premature aging and cancer-prone disorder, Werner syndrome. Both Pol δ and WRN encode 3' → 5' DNA exonuclease activities. Pol δ exonuclease removes 3'-terminal mismatched nucleotides incorporated during replication to ensure high fidelity DNA synthesis. WRN exonuclease degrades DNA containing alternate secondary structures to prevent formation and enable resolution of stalled replication forks. We now observe that similarly to WRN, Pol δ degrades alternate DNA structures including bubbles, four-way junctions, and D-loops. Moreover, WRN and Pol δ form a complex with enhanced ability to hydrolyze these structures. We also present evidence that WRN can proofread for Pol δ; WRN excises 3'-terminal mismatches to enable primer extension by Pol δ. Consistent with our in vitro observations, we show that WRN contributes to the maintenance of DNA synthesis fidelity in vivo. Cells expressing limiting amounts (～10% of normal) of WRN have elevated mutation frequencies compared with wild-type cells. Together, our data highlight the importance of WRN exonuclease activity and its cooperativity with Pol δ in preserving genome stability, which is compromised by the loss of WRN in Werner syndrome.     

DNA Polymerases that Propagate the Eukaryotic DNA Replication Fork

Abstract

Three DNA polymerases are thought to function at the eukaryotic DNA replication fork. Currently, a coherent model has been derived for the composition and activities of the lagging strand machinery. RNA-DNA primers are initiated by DNA polymerase α -primase. Loading of the proliferating cell nuclear antigen, PCNA, dissociates DNA polymerase α and recruits DNA polymerase δ and the flap endonuclease FEN1 for elongation and in preparation for its requirement during maturation, respectively. Nick translation by the strand displacement action of DNA polymerase δ, coupled with the nuclease action of FEN1, results in processive RNA degradation until a proper DNA nick is reached for closure by DNA ligase I. In the event of excessive strand displacement synthesis, other factors, such as the Dna2 nuclease/helicase, are required to trim excess flaps. Paradoxically, the composition and activity of the much simpler leading strand machinery has not been clearly established. The burden of evidence suggests that DNA polymerase ε normally replicates this strand, but under conditions of dysfunction, DNA polymerase δ may substitute.     

Regulation of oxidative DNA damage repair: The adenine:8-oxo-guanine problem

Reactive oxygen species (ROS) constantly attack DNA. One of the best-characterized oxidative DNA lesions is 7,8-dihydro-8-oxoguanine (8-oxo-G). Many human diseases, such as cancer and neurodegenerative disorders, have been correlated with oxidative DNA damage. In the last few years, DNA polymerase (Pol) λ, one of the 15 cellular Pols, has been identified to play an important role in performing accurate translesion synthesis over 8-oxo-G. This is eminently important, since normally faithful replicative Pols α, δ and ε, with their tight active center, often wrongly incorporate adenine (A) opposite the 8-oxo-G lesion. A:8- oxo-G mispairs are accurately repaired by the pathway identified in our laboratory involving MutY DNA glycosylase homolog (MutYH) and Pol λ. Until now, very little was known about the spatial and temporal regulation of Pol λ and MutYH in active repair complexes. We now showed in our latest publication that the E3 ligase Mule can ubiquitinate and degrade Pol λ, and that the control of Pol λ levels by Mule has functional consequences for the ability of mammalian cells to deal with 8-oxo-G lesions. In contrast, phosphorylation of Pol λ by Cdk2/cyclinA counteracts this degradation by recruiting it to MutYH on chromatin to form active 8-oxo-G repair complexes.     

Fidelity consequences of the impaired interaction between DNA polymerase epsilon and the GINS complex

DNA polymerase epsilon interacts with the CMG (Cdc45-MCM-GINS) complex by Dpb2p, the non-catalytic subunit of DNA polymerase epsilon. It is postulated that CMG is responsible for targeting of Pol ɛ to the leading strand. We isolated a mutator dpb2-100 allele which encodes the mutant form of Dpb2p. We showed previously that Dpb2-100p has impaired interactions with Pol2p, the catalytic subunit of Pol ɛ. Here, we present that Dpb2-100p has strongly impaired interaction with the Psf1 and Psf3 subunits of the GINS complex. Our in vitro results suggest that while dpb2-100 does not alter Pol ɛ’s biochemical properties including catalytic efficiency, processivity or proofreading activity – it moderately decreases the fidelity of DNA synthesis. As the in vitro results did not explain the strong in vivo mutator effect of the dpb2-100 allele we analyzed the mutation spectrum in vivo. The analysis of the mutation rates in the dpb2-100 mutant indicated an increased participation of the error-prone DNA polymerase zeta in replication. However, even in the absence of Pol ζ activity the presence of the dpb2-100 allele was mutagenic, indicating that a significant part of mutagenesis is Pol ζ-independent. A strong synergistic mutator effect observed for transversions in the triple mutant dpb2-100 pol2-4 rev3Δ as compared to pol2-4 rev3Δ and dpb2-100 rev3Δ suggests that in the presence of the dpb2-100 allele the number of replication errors is enhanced. We hypothesize that in the dpb2-100 strain, where the interaction between Pol ɛ and GINS is weakened, the access of Pol δ to the leading strand may be increased. The increased participation of Pol δ on the leading strand in the dpb2-100 mutant may explain the synergistic mutator effect observed in the dpb2-100 pol3-5DV double mutant.     

Mechanism of Replicative DNA Polymerase Delta Pausing and a Potential Role for DNA Polymerase Kappa in Common Fragile Site Replication

Common fragile sites (CFSs) are hot spots of chromosomal breakage, and CFS breakage models involve perturbations of DNA replication. Here, we analyzed the contribution of specific repetitive DNA sequence elements within CFSs to the inhibition of DNA synthesis by replicative and specialized DNA polymerases (Pols). The efficiency of in vitro DNA synthesis was quantitated using templates corresponding to regions within FRA16D and FRA3B harboring AT-rich microsatellite and quasi-palindrome (QP) sequences. QPs were predicted to form stems of ~ 75–100% self-homology, separated by 3–9 bases of intervening sequences. Analysis of DNA synthesis progression by human Pol δ demonstrated significant synthesis perturbation both at [A]_n and [TA]_n repeats in a length-dependent manner and at short (< 40 base pairs) QP sequences. DNA synthesis by the Y-family polymerase κ was significantly more efficient than Pol δ through both types of repetitive elements. Using DNA trap experiments, we show that Pol δ pauses within CFS sequences are sites of enzyme dissociation, and dissociation was observed in the presence of RFC-loaded PCNA. We propose that enrichment of microsatellite and QP elements at CFS regions contributes to fragility by perturbing replication through multiple mechanisms, including replicative Pol pausing and dissociation. Our finding that Pol δ dissociates at specific CFS sequences is significant, since dissociation of the replication machinery and inability to efficiently recover the replication fork can lead to fork collapse and/or formation of double-strand breaks in vivo. Our biochemical studies also extend the potential involvement of Y-family polymerases in CFS maintenance to include polymerase κ.     

GINS is a DNA polymerase epsilon accessory factor during chromosomal DNA replication in budding yeast

GINS is a protein complex found in eukaryotic cells that is composed of Sld5p, Psf1p, Psf2p, and Psf3p. GINS polypeptides are highly conserved in eukaryotes, and the GINS complex is required for chromosomal DNA replication in yeasts and Xenopus egg. This study reports purification and biochemical characterization of GINS from Saccharomyces cerevisiae. The results presented here demonstrate that GINS forms a 1:1 complex with DNA polymerase epsilon (Pol epsilon) holoenzyme and greatly stimulates its catalytic activity in vitro. In the presence of GINS, Pol epsilon is more processive and dissociates more readily from replicated DNA, while under identical conditions, proliferating cell nuclear antigen slightly stimulates Pol epsilon in vitro. These results strongly suggest that GINS is a Pol epsilon accessory protein during chromosomal DNA replication in budding yeast. Based on these results, we propose a model for molecular dynamics at eukaryotic chromosomal replication fork.     

Beyond translesion synthesis: polymerase κ fidelity as a potential determinant of microsatellite stability

Microsatellite DNA synthesis represents a significant component of human genome replication that must occur faithfully. However, yeast replicative DNA polymerases do not possess high fidelity for microsatellite synthesis. We hypothesized that the structural features of Y-family polymerases that facilitate accurate translesion synthesis may promote accurate microsatellite synthesis. We compared human polymerases κ (Pol κ) and η (Pol η) fidelities to that of replicative human polymerase δ holoenzyme (Pol δ4), using the in vitro HSV-tk assay. Relative polymerase accuracy for insertion/deletion (indel) errors within 2-3 unit repeats internal to the HSV-tk gene concurred with the literature: Pol δ4 > Pol κ or Pol η. In contrast, relative polymerase accuracy for unit-based indel errors within [GT](10) and [TC](11) microsatellites was: Pol κ ≥ Pol δ4 > Pol η. The magnitude of difference was greatest between Pols κ and δ4 with the [GT] template. Biochemically, Pol κ displayed less synthesis termination within the [GT] allele than did Pol δ4. In dual polymerase reactions, Pol κ competed with either a stalled or moving Pol δ4, thereby reducing termination. Our results challenge the ideology that pol κ is error prone, and suggest that DNA polymerases with complementary biochemical properties can function cooperatively at repetitive sequences.     

DNA polymerase δ and ζ switch by sharing accessory subunits of DNA polymerase δ

Translesion DNA synthesis is an important branch of the DNA damage tolerance pathway that assures genomic integrity of living organisms. The mechanisms of DNA polymerase (Pol) switches during lesion bypass are not known. Here, we show that the C-terminal domain of the Pol ζ catalytic subunit interacts with accessory subunits of replicative DNA Pol δ. We also show that, unlike other members of the human B-family of DNA polymerases, the highly conserved and similar C-terminal domains of Pol δ and Pol ζ contain a [4Fe-4S] cluster coordinated by four cysteines. Amino acid changes in Pol ζ that prevent the assembly of the [4Fe-4S] cluster abrogate Pol ζ function in UV mutagenesis. On the basis of these data, we propose that Pol switches at replication-blocking lesions occur by the exchange of the Pol δ and Pol ζ catalytic subunits on a preassembled complex of accessory proteins retained on DNA during translesion DNA synthesis.