Yeast Rev1 protein promotes complex formation of DNA polymerase ζ with Pol32 subunit of DNA polymerase δ

Edited by Jerard Hurwitz, Memorial Sloan–Kettering Cancer Center, New York, NY, and approved April 29, 2009
June 16, 2009
106 (24) 9631-9636

Abstract

Yeast DNA polymerase (Pol) δ, essential for DNA replication, is comprised of 3 subunits, Pol3, Pol31, and Pol32. Of these, the catalytic subunit Pol3 and the second subunit Pol31 are essential, whereas the Pol32 subunit is not essential for DNA replication. Although Pol32 is an integral component of Polδ, it is also required for translesion synthesis (TLS) by Polζ. To begin to decipher the bases of Pol32 involvement in Polζ-mediated TLS, here we examine whether Pol32 physically interacts with Polζ or its associated proteins and provide evidence for the physical interaction of Pol32 with Rev1. Rev1 plays an indispensable structural role in Polζ-mediated TLS and it binds the Rev3 catalytic subunit of Polζ. Here, we show that although Pol32 does not directly bind Polζ, Pol32 can bind the Rev1–Polζ complex through its interaction with Rev1. We find that Pol32 binding has no stimulatory effect on DNA synthesis either by Rev1 in the Rev1–Pol32 complex or by Polζ in the Polζ–Rev1–Pol32 complex, irrespective of whether proliferating cell nuclear antigen has been loaded onto DNA or not. We discuss evidence for the biological significance of Rev1 binding to Pol32 for Polζ function in TLS and suggest a structural role for Rev1 in modulating the binding of Polζ with Pol32 in Polδ stalled at a lesion site.
DNA lesions in the template strand block synthesis by replicative polymerases (Pols). In the yeast Saccharomyces cerevisiae replication through DNA lesions can be restored by translesion synthesis (TLS) by DNA Pols η, ζ, and Rev1 (1). Polη, a Y family Pol, is unique among eukaryotic Pols in its proficient and relatively accurate ability to replicate through UV-induced cyclobutane pyrimidine dimers (CPDs) (25). Inactivation of Polη in yeast and humans confers an increase in the frequency of UV-induced mutations and causes in humans the cancer-prone syndrome, the variant form of xeroderma pigmentosum (XPV) (6, 7). Polη can also replicate through other DNA lesions, such as a GG intrastrand cross-link and an 8-oxoguanine (8, 9).
DNA Polζ is comprised of the Rev3 catalytic subunit that shares homology with B family Pols and the Rev7 accessory subunit (10). By contrast to Polη, which can replicate through DNA lesions such as a CPD, GG intrastrand cross-link, and 8-oxoguanine, replication through a large variety of DNA lesions requires the sequential action of 2 TLS Pols, in which one inserts the nucleotide opposite the lesion site and the other carries out the subsequent extension step (1). Extensive biochemical studies with Polζ have shown that on undamaged DNAs it is a proficient extender of mispaired primer termini, and on damaged DNAs it performs the extension step of lesion bypass (1, 11). Opposite a number of DNA lesions, as for example, cis-syn TT dimer, (6–4) TT photoproduct, abasic site, the ring-closed form of γ-HOPdG (PdG), and 1,N6-ethenodeoxyadenosine (εdA), Polζ is considerably inhibited at inserting a nucleotide opposite the lesion site but it can proficiently extend from the nucleotide inserted opposite the DNA lesion by another Pol (1115).
Like Polη, Rev1 is a member of Y family Pols. Rev1, however, is highly specialized for incorporating a C opposite template G (16, 17), and unlike any other Pol, in Rev1, the templating G and the incoming dCTP do not pair with each other. Instead, template G is evicted from the DNA helix into a large solvent-filled cavity where its Hoogsteen edge hydrogen-bonds with residues in Rev1, and the incoming dCTP pairs with an arginine in another segment of Rev1 (18). A large number of DNA adducts can form at the highly reactive N2 group of guanine. By evicting the N2-dG adduct into a large solvent-filled cavity and pairing the incoming dCTP with a surrogate arginine, Rev1 Pol activity can mediate TLS opposite such adducts by incorporating a C opposite the N2-dG adduct, from which Polζ can then carry out the subsequent extension reaction (15).
Although Rev1's Pol activity can promote TLS opposite N2-dG adducts, Rev1 plays an additional indispensable structural role for Polζ function in TLS, and depending on the DNA lesion, its Pol activity may or may not be required for TLS. For example, opposite UV-induced DNA lesions, Polζ functions in an error-prone mode of TLS, which derives from its ability to carry out extension from the correct and the incorrect nucleotide inserted opposite the 3′ pyrimidine of a CPD or a (6–4) photoproduct by another Pol (11, 13). As a consequence, inactivation of REV3 or REV7 confers a large reduction in the incidence of UV mutagenesis (1921). Although Rev1's Pol activity plays no role in TLS opposite UV lesions, the Rev1 protein is required for UV mutagenesis and the rev1Δ mutation generates the same high level of defect in UV mutagenesis as the rev3Δ or rev7Δ mutation (20, 22). Rev1 forms a physical complex with Polζ in which the Rev1 C terminus binds Rev3. Deletion of the C-terminal region of Rev1 that is involved in Rev3 binding produces the same high level of defect in UV mutagenesis as that conferred by the rev1Δ mutation, suggesting that Rev1 binding is important for Polζ's ability to function in TLS (23).
Yeast Polδ consists of 3 subunits, Pol3, Pol31, and Pol32 (24). The Pol3 catalytic subunit and the Pol31 subunit are highly conserved in eukaryotes, and they are both essential. The third subunit, Pol32, is much less conserved and it is not essential. Even though Pol32 is a subunit of Polδ, it is also required for Polζ function in TLS (2427), and it has remained unclear how Pol32 might affect Polζ function. To begin an understanding of the role of Pol32 in Polζ-mediated TLS, we examined whether Pol32 was involved in direct physical interactions with Polζ or with Rev1. Here, we provide evidence for complex formation of Pol32 with Rev1, and we find that even though Polζ does not directly bind Pol32, the Rev1–Polζ complex can bind Pol32. From these observations we infer that Rev1 can promote the association of Polζ with Pol32. We discuss the possible implications of this and other observations for the targeting of Polζ to the replication fork stalled at a DNA lesion site.

Results

Physical Interaction of Pol32 with Rev1.

We examined the physical interaction of Pol32 with Rev1 by using the GST pull-down assay. In this assay, the GST-fused protein is incubated with the putative interacting protein, the proteins are then bound to glutathione Sepharose affinity beads, and the interacting protein is pulled down if it forms a stable complex with the GST fusion protein. The proteins were purified to near homogeneity, and GST–Pol32 was incubated with Rev1, and conversely, GST–Rev1 was incubated with Pol32. The protein mixtures were then incubated with glutathione-Sepharose beads for an additional period, and the beads were spun down and washed extensively with buffer containing 150 mM NaCl. The proteins bound to beads were eluted with SDS-containing buffer and the load, flow through, and eluate fractions were analyzed by gel electrophoresis. As shown in Fig. 1A, Pol32 bound to Rev1 regardless of whether the GST–Rev1 protein was incubated with Pol32 (lanes 1–4) or the GST–Pol32 protein was incubated with Rev1 (lanes 5–8). The binding of Pol32 to Rev1 is not caused by nonspecific binding of these proteins to GST or the Sepharose beads because neither Pol32 or Rev1 could bind the GST beads in the absence of interacting GST–fusion protein (Fig. 1A, lanes 9–16).
Fig. 1.
Complex formation of Pol32 with Rev1. (A) Physical interaction of Pol32 with Rev1. Pol32 or Rev1 was mixed with GST–Rev1 (lanes 1–4), or Rev1 was mixed with GST–Pol32 (lanes 5–8). Approximately 3 μg of each protein was used for this study. After incubation, samples were bound to glutathione-Sepharose beads, followed by washing and elution of the bound proteins by SDS-sample buffer. Aliquots of each sample before addition to the beads (L), the flow-through fraction (F), last washing fraction (W), and the eluted proteins (E) were analyzed on a SDS-12% polyacrylamide gel developed with Coomassie blue. A control experiment was also done for GST binding with Pol32 (lanes 9–12) and Rev1 (lanes 13–16). (B) Association of Pol32 with Rev1 in yeast cells. Pol32–FLAG and Rev1–HA proteins in crude cell extract (lanes 1 and 2); in the absence of tagged Pol32, Rev1-HA is not immunoprecipitated (lane 3); when Pol32-FLAG is pulled down by the anti-FLAG affinity gel, Rev1-HA is coimmunoprecipitated (lane 4).
To examine the association of Pol32 with Rev1 in yeast cells, we used a rev1Δ yeast strain in which a FLAG tag has been inserted at the C terminus of genomic wild-type Pol32 gene and that harbors a C-terminal HA-tagged REV1 gene on a plasmid. FLAG–Pol32 and its associated proteins were immunoprecipitated from cell lysates by using anti-FLAG beads, and the precipitated proteins were analyzed for the presence of HA-tagged Rev1 with anti-HA antibody. As shown in Fig. 1B (lane 4), association of Rev1 with Pol32 could be detected in yeast cells.

Mapping the Regions of Rev1 and Pol32 Involved in Physical Interaction.

To map the region of Rev1 involved in binding to Pol32, GST–Rev1 proteins from which different Rev1 portions had been deleted were incubated with Pol32, and alternatively, GST–Pol32 protein was incubated with Rev1 proteins deleted for different portions (Fig. 2Ai). The proteins were bound to glutathione-Sepharose affinity beads and bound proteins were eluted after a series of washings with 150 mM NaCl. As shown in Fig. 2B, whereas Rev1–2 protein containing the residues 329–746, from which the N-terminal BRCT domain and the C-terminal 239 residues have been deleted, retains the ability to bind Pol32 (lanes 1–8), the removal of an additional 140 C-terminal residues as in the Rev1–3 protein, which deletes the Pol-associated domain (PAD), shows no binding to Pol32 (lanes 9–16). We next examined whether the Rev1–4 peptide that contains the 201 residues that encompass the PAD region could bind Pol32 and found that this Rev1 segment was sufficient for Pol32 binding (Fig. 2B, lanes 17–24). The Rev1–5 peptide that contains only the C-terminal 200 residues showed no binding to Pol32 (Fig. 2B, lanes 25–32). From these observations, we conclude that the PAD region of Rev1 interacts with Pol32.
Fig. 2.
Mapping of regions in Rev1 and Pol32 involved in physical interaction. (A) (i) Schematic representation of wild-type Rev1 and its truncated forms. Yeast Rev1 protein (985 aa) has a BRCT domain toward its amino terminus and has the 5 conserved motifs (I–V) characteristic of Y family Pols, and motif V is followed by the PAD. All 5 motifs, including the PAD, are indispensable for Rev1 DNA synthetic activity. The region at the carboxyl terminus of Rev1 is referred to as carboxyl-terminal domain (CTD). The residues that remain in the truncated protein are indicated in parentheses. (ii) Schematic of 350-residue Pol32 protein and its truncated forms. The interaction data shown in B and C are summarized. (B) The PAD of Rev1 is necessary and sufficient for interaction with Pol32. Pol32 was mixed with GST–Rev1–2 (lanes 1–4), GST–Rev1–3 (lanes 9–12), GST–Rev1–4 (lanes 17–20), or GST–Rev1–5 (lanes 25–28). In reciprocal experiments, GST–Pol32 was mixed with Rev1–2 (lanes 5–8), Rev1–3 (lanes 13–16), Rev1–4 (lanes 21–24), or Rev1–5 (lanes 29–32). (C) Pol32 region involved in interaction with Rev1. Rev1 was mixed with GST–Pol32–1 (lanes 1–4), GST–Pol32 -2 (lanes 5–8), and GST–Pol32–3 (lanes 9–12). For both B and C, ≈3 μg of each protein was used for the study. After incubation, samples were bound to glutathione-Sepharose beads, followed by washing and elution of the bound proteins by SDS-sample buffer. Aliquots of each sample before additions to the beads (L), the flow-through fraction (F), last washing fraction (W), and the eluted proteins (E) were analyzed on a SDS-12% polyacrylamide gel stained with Coomassie blue.
Using a similar set of Rev1 proteins deleted for different portions, we have previously shown that different Rev1 regions affect its interaction with different proteins. For example, whereas the C-terminal Rev1 (Rev1–5) peptide that contains residues 786–985 does not bind Pol32 (Fig. 2 A and B), it can bind Rev3, the catalytic subunit of Polζ (23). By contrast, whereas the deletion of the Rev1 C terminus beyond the PAD has no adverse effect on Pol32 binding (Fig. 2 A and B), removal of even the C-terminal 72 residues of Rev1 inactivates its binding to Rev3 (23). Hence, Rev1 would carry out its function in Polζ-mediated TLS by binding to both the Rev3 and Pol32 proteins via its C terminus and PAD regions, respectively.
To map the region in Pol32 involved in the binding to Rev1, we used C-terminally-truncated Pol32 proteins for pull-down experiments with Rev1 (Fig. 2Aii). Pol32 has a conserved proliferating cell nuclear antigen (PCNA) binding PIP motif at the extreme C terminus, and we found that deletion of the C-terminal 8 residues that deletes the conserved FF residues of PIP motif (Pol32–1) has no adverse effect on interaction with Rev1 (Fig. 2C, lanes 1–4). Moreover, removal of the C-terminal 169 residues in the Pol32–2 protein also does not affect interaction with Rev1 (Fig. 2C, lanes 5–8), indicating that the site for binding Rev1 is contained within the N-terminal 181 residues of Pol32. Because the ability to bind Rev1 is abolished in the Pol32–3 protein that retains only the amino terminal 104 residues (Fig. 2C, lanes 9–12), the Rev1 interaction site in Pol32 lies in the region encompassing approximately residues 104–181.
In a previous study, using 2-hybrid analyses, Johansson et al. (28) examined the interactions of Pol32 protein deleted for different regions with the Pol31 subunit of Polδ. They found that the Pol32 protein deleted beyond residue 92 can bind Pol31, and in GST pull-down assays we have observed physical binding of the Pol32–3 peptide that contains only the residues 1–104 with purified Pol31. Thus, although this N-terminal Rev1 segment retains the ability to bind Pol31, it loses its capacity for binding Rev1 (Fig. 2). As we elaborate in Discussion, the previous deletion studies of Pol32 (28), when combined with our analysis of the Rev1 binding site in Pol32, have led us to conclude that Polζ-mediated TLS is governed by the ability of Rev1 to bind Pol32, which in turn has to be bound to Pol31.

Pol32 Can Also Bind the Rev1–Polζ Complex.

To determine whether Pol32 and Polζ can form a physical complex, we carried out reciprocal experiments in which we examined the binding of Pol32 protein with the GST–Rev3–Rev7 complex and the binding of Polζ (Rev3–Rev7) to the GST–Pol32 protein. As shown in Fig. 3A, however, we saw no evidence of direct physical interaction of Pol32 with Polζ. Because yeast Rev1 can bind to both Polζ and Pol32 individually, next we determined whether a Polζ–Rev1–Pol32 complex could be formed. When we added the preformed Rev1–Pol32 complex to GST–Polζ, we saw no evidence of the binding of Rev1–Pol32 with Polζ (Fig. 3B, lanes 1–4). However, when Pol32 was added to the preformed GST–Polζ–Rev1 complex, then a complex of Rev1 bound to Polζ and Pol32 could be obtained (Fig. 3B, lanes 5–8). From these observations we infer that Rev1 can bring about the association of Polζ with the Pol32 subunit of Polδ. Our observation of complex formation of Rev1–Polζ with Pol32 but not of Rev1–Pol32 with Polζ raises the possibility that only Polζ-bound Rev1 can bring about the association of Polζ with Pol32. Furthermore, it suggests that the prior binding of Pol32 to Rev1 imposes a steric hindrance to the subsequent binding of this protein complex to Polζ.
Fig. 3.
Formation of a Polζ–Rev1–Pol32 complex. (A) Polζ does not directly bind Pol32. GST–Polζ was mixed with Pol32 (lanes 1–4) or GST–Pol32 was mixed with Polζ (lanes 5–8) and examined for the binding of Polζ with Pol32. (B) The Rev1–Polζ complex binds Pol32. The Rev1–Pol32 complex was mixed with GST–Polζ (lanes 1–4) or Rev1 was first mixed with GST–Polζ for 2 h at 4 °C and the preformed GST–Polζ–Rev1 complex was then mixed with Pol32 for 1 h at 25 °C (lanes 5–8). Approximately 2 μg of each protein was used for this study. After incubation, samples were bound to glutathione-Sepharose beads, followed by washing and elution of the bound proteins by SDS-sample buffer. Aliquots of each sample before additions to the beads (L), the flow-through fraction (F), last washing fraction (W), and the eluted proteins (E) were analyzed on a SDS-12% polyacrylamide gel developed with Coomassie blue.

Pol32 Binding Does Not Affect Rev1 DNA Synthetic Activity.

To examine whether complex formation with Pol32 affects the DNA synthetic activity of Rev1, DNA substrates containing a C or an abasic residue at the template site were incubated with Rev1 alone or with the Rev1–Pol32 complex in the presence of just 1 of the 4 dNTPs (Fig. S1A). As is the case for the Rev1 protein (16, 17), the Rev1–Pol32 complex incorporates only the C nucleotide opposite the C or the abasic templating residue, indicating that complex formation with Pol32 does not affect the specificity of Rev1 action for C incorporation on the undamaged C template or damaged abasic template. Next, we analyzed by steady-state kinetics whether complex formation with Pol32 affected the catalytic efficiency of Rev1 (Fig. S1B). The kinetics of C insertion by Rev1 and the Rev1–Pol32 complex opposite a template G or an abasic site were determined as a function of deoxynucleotide concentration under steady-state conditions. From the kinetics of deoxynucleotide incorporation, the steady-state apparent kcat and Km values for C incorporation were obtained from the curve fitted to the Michaelis–Menten equation, and the efficiencies (kcat/Km) of C incorporation were calculated. Rev1 and the Rev1–Pol32 complex incorporated a C opposite the template G or the abasic site with about the same catalytic efficiency (Table 1), indicating that interaction with Pol32 has no significant effect on the DNA synthetic activity of Rev1.
Table 1.
Steady-state kinetic parameters for dCTP incorporation opposite template G and an abasic site (AP) by Rev1 and the Rev1–Pol32 complex
Template nucleotidedNTP addedEnzymekcat, nM/minKm, μMkcat/KmFold difference
GdCTPRev14.07 ± 0.110.03 ± 0.004135.71
  Rev1–Pol324.4 ± 0.150.028 ± 0.005157.11.15
APdCTPRev12.81 ± 0.170.93 ± 0.223.021
  Rev1–Pol323.2 ± 0.160.8 ± 0.2541.32

PCNA Does Not Stimulate DNA Synthesis by Rev1 in the Rev1–Pol32 Complex or by Polζ in the Polζ–Rev1–Pol32 Complex.

The carboxyl terminus of the 350-residue Pol32 protein contains the PCNA binding motif Q G T L E S F F from residues 338–345, and deletions of Pol32 that remove this motif are defective for interaction with PCNA in 2-hybrid analyses (28). Although, previously we have provided evidence for the physical and functional interaction of yeast Polη and human Pols η, ι, and κ, with PCNA (2932), we have been unable to substantiate such evidence for the interaction of Rev1 or Polζ with PCNA. Thus, whereas the binding of PCNA by yeast Polη and human Pols η, ι, and κ is stimulatory to their DNA synthetic activity, we have found no such stimulatory effect of PCNA on the Rev1 or Polζ activity (33). Because of the ability of Rev1 to bind Pol32 and simultaneously bind Polζ we considered the possibility that in the Pol32–Rev1 complex or the Polζ–Rev1–Pol32 complex perhaps some evidence could be adduced for their binding to PCNA.
To examine the effect of PCNA on the DNA synthetic activity of Rev1 in the Rev1–Pol32 complex and Polζ in the Polζ–Rev1 complex and the Polζ–Rev1–Pol32 complex, we carried out DNA synthesis reactions by using a DNA substrate in which PCNA stays bound to DNA. The DNA substrate contains a 5′ 32P-labeled 45-nt primer annealed to a linear 75-nt template DNA onto which a biotin molecule was attached at each end, and the subsequent binding of streptavidin molecules to the ends of the DNA substrate prevents sliding of PCNA from the DNA after it has been loaded by replication factor C (RFC). Because the first template nucleotide after the 3′ end of the primer is a G, we compared the ability of Rev1 and the Rev1–Pol32 complex for C incorporation opposite template G in the presence of dCTP. As shown in Fig. S2 (lanes 2–5), the addition of PCNA, RFC, and replication protein A (RPA) has no stimulatory effect on the DNA synthetic activity of either Rev1 or the Rev1–Pol32 complex. To examine the effect of PCNA on the DNA synthetic activity of Polζ in the Polζ–Rev1–Pol32 complex, we compared the activity of Polζ with the Polζ–Rev1* and Polζ-Rev1*–Pol32 complexes in the presence of 4 dNTPs. To avoid the confounding effects of DNA synthesis by Rev1, we used the catalytically-inactive Rev1 protein (Rev1*) in which the catalytic residues Asp-467 and Glu-468 required for Mg2+ binding (18) both were changed to alanine. The addition of PCNA along with RFC and RPA had no stimulatory effect on the DNA synthetic activity of Polζ in the Rev1*–Polζ complex or the Polζ–Rev1*–Pol32 complex (Fig. S2, lanes 6–11), suggesting that complex formation with Pol32 has no stimulatory effect on DNA synthesis by Rev1 or Polζ in the presence of PCNA.

Complex Formation with Pol32 Is Inhibitory to Rev1's Ability to Associate with Rev7 or Polη.

Because yeast Rev1 can form a complex with Rev7 (34), Polη (35), or Pol32 via interaction with its PAD region encompassing residues 567–767, and because Rev1 binding to Polη is inhibited in the Rev1–Rev7 complex (35), we expected that Rev1's binding to any one of these proteins would preclude its binding to the other protein. To verify this possibility, we examined whether interaction of Rev1 with Rev7 or Rad30 (Polη) is inhibited in the Rev1–Pol32 complex. For this purpose, we incubated preformed Rev1–Pol32 complex with GST–Rev7 or GST–Rad30 and examined whether a complex of Rev1–Pol32 with Rev7 or Rad30 could be isolated. As shown in Fig. 4, we found no evidence for the binding of Rev1–Pol32 complex with Rev7 or Polη. Thus, the Rev1–Pol32 complex precludes the binding of Rev1 with Rev7 or Polη. These observations taken together with previous observations (34, 35) indicate the involvement of the Rev1 PAD region in forming independent complexes with Rev7, Polη, and Pol32.
Fig. 4.
Rev1–Pol32 complex does not bind Rev7 or Rad30. Rev1–Pol32 complex was mixed with GST-Rev7 (lanes 1–4) or GST-Rad30 (lanes 5–8). Approximately 2 μg of each protein was used for this study. After incubation, samples were bound to glutathione-Sepharose beads, followed by washings and elution of the bound proteins by SDS-sample buffer. Aliquots of each sample before additions to the beads (L), the flow-through fraction (F), last washing fraction (W), and the eluted proteins (E) were analyzed on a SDS-12% polyacrylamide gel developed with Coomassie blue.

Discussion

Rev1 Promotes Complex Formation of Polζ with Pol32.

Rev1 binds Rev3 in Polζ via its C terminus, and deletion of this region in Rev1 causes the same high degree of defect in UV mutagenesis as that conferred by the inactivation of Polζ (23). Previously, we have suggested that binding of Polζ to Rev1 provides a means for targeting Polζ to the replication fork stalled at a DNA lesion (23). Here, we show that Rev1 participates additionally in complex formation with the Pol32 subunit of Polδ, which also is indispensable for Polζ function in TLS (2427). Further, we provide evidence that the Rev1–Polζ complex can also bind Pol32. By its ability to bind Rev3 in Polζ via its C terminus and to bind Pol32 via its PAD region, Rev1 thus can promote the association of Polζ with Pol32.

Biological Significance of Rev1 Interaction with Pol32 for Polζ-Mediated TLS.

How might Pol32 effect its role in Polζ-mediated TLS? A clue to this quandary is provided from studies of various Pol32 deletion mutants that have been assayed for effects on UV sensitivity and UV mutagenesis and for their ability to interact with the Pol31 subunit of Polδ by 2-hybrid analyses (28). These studies have indicated that the N-terminal 92 aa of Pol32 are required for interaction with Pol31 and that deletions of Pol32 that encompass the N-terminal region of 92 residues cause the same high degree of UV sensitivity and defective UV mutagenesis as that conferred by the pol32Δ mutation (28). These observations imply that binding to Pol31 is crucial for Pol32 to carry out its role in Polζ-mediated TLS.
Because Pol31 binding is critical for Pol32's ability to modulate Polζ-mediated TLS, and we find that Rev1 either by itself or as part of the Rev1–Polζ complex can bind Pol32, we wanted to know whether Rev1 binding to Pol32 was a prerequisite for Polζ function. The data that have been previously obtained from the deletion analyses of Pol32 (28), when combined with our observation that the Rev1 interaction site in Pol32 lies in the region encompassing approximately residues 100–180, allow us to deduce whether Rev1 interaction with Pol32 performs a crucial task in Polζ-mediated TLS. The results of Pol32 deletion analyses have indicated that even though Pol32 protein from which residues 93–309 have been deleted can interact with Pol31, it still remains defective for UV mutagenesis and confers UV sensitivity, whereas Pol32 protein that retains the N-terminal 142 aa is proficient for UV mutagenesis and exhibits wild-type UV sensitivity (28). We infer from these observations that the region of Pol32 spanning approximately residues 93–142 is also critical for Polζ function in TLS. Because the Rev1 interaction site in Pol32 is located just after the Pol31 interaction site, and deletion of this adjoining region confers UV sensitivity and defective UV mutagenesis (28), we infer that Rev1 interaction with Pol32 modulates a necessary step in Polζ-mediated TLS.

Implications of Rev1 Interaction with Pol32 and Polζ.

The results of our study, when taken together with previous observations indicating that the ability of Pol32 to bind to the Pol31 subunit of Polδ is necessary for UV mutagenesis (28) and hence for Polζ function in TLS, raise the possibility that via its ability to simultaneously bind Polζ and Pol32, Rev1 potentiates the targeting of Polζ to Polδ stalled at a DNA lesion site. This suggestion, however, is countered by recently reported observations that implicate the involvement of Polε in the synthesis of leading strand and Polδ in the synthesis of lagging strand during replication (3638), because that would then require the consignment of TLS to only the lagging strand. That clearly is not the case, because studies with a plasmid system where bidirectional replication ensues from a replication origin and where the DNA lesion is carried either on the template for the synthesis of leading strand or the template for the synthesis of lagging strand, have shown that TLS operates nearly equally frequently on the 2 DNA strands in yeast cells (27).
How could we reconcile the involvement of Pols ε and δ for the synthesis of leading and lagging strands, respectively, with the suggested role of Rev1 in linking Polζ with Polδ? One possibility is that Polδ plays a major role in the replication of both DNA strands and that Polε contributes to DNA synthesis of the leading strand only under special circumstances. A number of observations would seem to support such a notion. For example, the observations that the pol3–13 mutation in the Pol3 catalytic subunit of Polδ renders yeast cells defective in UV mutagenesis (39), and that the temperature-sensitive (ts) conditional lethal mutation, pol3–3 of Polδ confers a defect in postreplication repair (PRR) of UV-damaged DNA, whereas the ts pol2–18 mutation of Polε has no adverse effect on PRR (40), would suggest that these defects resulting from mutations in the Pol3 subunit of Polδ reflect the involvement of Polδ in the synthesis of both DNA strands. Also, the observations that yeast cells harboring deletions of the Pol domain of Polε are viable (41), whereas the DNA Pol function of Polδ is essential for viability, and that the inactivation of the proofreading exonuclease activity of Polδ causes a much greater increase in spontaneous mutability than does the inactivation of Polε proofreading exonuclease (42, 43), are all consistent with the possibility that Polδ plays a major role in the replication of both the leading and lagging DNA strands.

Yeast Rev1 Participates in Mutually-Exclusive Complex Formations with Different Proteins.

Whereas Rev1 binds Rev3 in Polζ via its C terminus, it binds Rev7, Polη, and Pol32 via its PAD region. Hence Rev1 can participate in independent complex formations with a number of proteins by binding each one of them through the same PAD region. It has remained unclear what the role of the Rev1–Rev7 complex might be (34). For the Rev1–Polη complex, we have previously suggested that this complex might provide for a more efficient means of TLS opposite certain types of DNA lesions where Rev1 inserts a nucleotide opposite the lesion site and Polη then extends. For example, opposite N2-guanine adducts such as N2-dG butadiene adducts, after the insertion of a C by Rev1, the extension step could be performed by Polη (44). And, as we have discussed above, the ability of Rev1 to simultaneously bind Polζ and Pol32 provides a way by which Rev1 could bring about the association of Polζ with Polδ. Hence, by promoting the targeting of Polζ to a stalled replication fork, Rev1 could play an indispensable structural role in Polζ-mediated TLS.

Materials and Methods

The yeast strains used and detailed methods for immunoprecipitations, pull-down assays for examining physical interactions, and DNA polymerase assays in the presence and absence of PCNA, RFC, and RPA are provided as SI Text.

Acknowledgments.

We thank Anshu Bhatnagar for participating in the initial studies of Rev1 interaction with Pol32. This work was supported by National Institutes of Health Grant CA107650.

Supporting Information

Supporting Information (PDF)
Supporting Information

References

1
S Prakash, RE Johnson, L Prakash, Eukaryotic translesion synthesis DNA polymerases: Specificity of structure and function. Annu Rev Biochem 74, 317–353 (2005).
2
RE Johnson, S Prakash, L Prakash, Efficient bypass of a thymine-thymine dimer by yeast DNA polymerase, Polη. Science 283, 1001–1004 (1999).
3
RE Johnson, MT Washington, S Prakash, L Prakash, Fidelity of human DNA polymerase η. J Biol Chem 275, 7447–7450 (2000).
4
MT Washington, RE Johnson, S Prakash, L Prakash, Accuracy of thymine-thymine dimer bypass by Saccharomyces cerevisiae DNA polymerase η. Proc Natl Acad Sci USA 97, 3094–3099 (2000).
5
MT Washington, L Prakash, S Prakash, Mechanism of nucleotide incorporation opposite a thymine–thymine dimer by yeast DNA polymerase η. Proc Natl Acad Sci USA 100, 12093–12098 (2003).
6
RE Johnson, CM Kondratick, S Prakash, L Prakash, hRAD30 mutations in the variant form of xeroderma pigmentosum. Science 285, 263–265 (1999).
7
C Masutani, et al., The XPV (xeroderma pigmentosum variant) gene encodes human DNA polymerase η. Nature 399, 700–704 (1999).
8
L Haracska, S-L Yu, RE Johnson, L Prakash, S Prakash, Efficient and accurate replication in the presence of 7,8-dihydro-8-oxoguanine by DNA polymerase η. Nat Genet 25, 458–461 (2000).
9
A Vaisman, C Masutani, F Hanaoka, SG Chaney, Efficient translesion replication past oxaliplatin and cisplatin GpG adducts by human DNA polymerase η. Biochemistry 39, 4574–4580 (2000).
10
JR Nelson, CW Lawrence, DC Hinkle, Thymine–thymine dimer bypass by yeast DNA polymerase ζ. Science 272, 1646–1649 (1996).
11
RE Johnson, MT Washington, L Haracska, S Prakash, L Prakash, Eukaryotic polymerases ι and ζ act sequentially to bypass DNA lesions. Nature 406, 1015–1019 (2000).
12
L Haracska, et al., Roles of yeast DNA polymerases δ and ζ and of Rev1 in the bypass of abasic sites. Genes Dev 15, 945–954 (2001).
13
RE Johnson, L Haracska, S Prakash, L Prakash, Role of DNA polymerase η in the bypass of a (6–4) TT photoproduct. Mol Cell Biol 21, 3558–3563 (2001).
14
DT Nair, RE Johnson, L Prakash, S Prakash, AK Aggarwal, Hoogsteen base pair formation promotes synthesis opposite the 1,N6-ethenodeoxyadenosine lesion by human DNA polymerase ι. Nat Struct Mol Biol 13, 619–625 (2006).
15
DT Nair, RE Johnson, L Prakash, S Prakash, AK Aggarwal, Protein-template directed synthesis across an acrolein-derived DNA adduct by yeast Rev1 DNA polymerase. Structure (London) 16, 239–245 (2008).
16
L Haracska, S Prakash, L Prakash, Yeast Rev1 protein is a G template-specific DNA polymerase. J Biol Chem 277, 15546–15551 (2002).
17
JR Nelson, CW Lawrence, DC Hinkle, Deoxycytidyl transferase activity of yeast REV1 protein. Nature 382, 729–731 (1996).
18
DT Nair, RE Johnson, L Prakash, S Prakash, AK Aggarwal, Rev1 employs a novel mechanism of DNA synthesis using a protein template. Science 309, 2219–2222 (2005).
19
CW Lawrence, RB Christensen, Ultraviolet-induced reversion of cyc1 alleles in radiation-sensitive strains of yeast. III. rev3 mutant strains. Genetics 92, 397–408 (1979).
20
CW Lawrence, T O'Brien, J Bond, UV-induced reversion of his4 frameshift mutations in rad6, rev1, and rev3 mutants of yeast. Mol Gen Genet 195, 487–490 (1984).
21
CW Lawrence, PE Nisson, RB Christensen, UV and chemical mutagenesis in rev7 mutants of yeast. Mol Gen Genet 200, 86–91 (1985).
22
CW Lawrence, RB Christensen, Ultraviolet-induced reversion of cyc1 alleles in radiation-sensitive strains of yeast. I. rev1 mutant strains. J Mol Biol 122, 1–21 (1978).
23
N Acharya, RE Johnson, S Prakash, L Prakash, Complex formation with Rev1 enhances the proficiency of yeast DNA polymerase ζ for mismatch extension and for extension opposite from DNA lesions. Mol Cell Biol 26, 9555–9563 (2006).
24
KJ Gerik, X Li, A Pautz, PMJ Burgers, Characterization of the two small subunits of Sacharomyces cerevisiae DNA polymerase δ. J Biol Chem 273, 19747–19755 (1998).
25
PEM Gibbs, J McDonald, R Woodgate, CW Lawrence, The relative roles in vivo of Saccharomyces cerevisiae Polη, Polζ, 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 (2005).
26
RE Johnson, et al., Identification of APN2, the Saccharomyces cerevisiae homolog of the major human AP endonuclease HAP1, and its role in the repair of abasic sites. Genes Dev 12, 3137–3143 (1998).
27
V Pages, RE Johnson, L Prakash, S Prakash, Mutational specificity and genetic control of replicative bypass of an abasic site in yeast. Proc Natl Acad Sci USA 105, 1170–1175 (2008).
28
E Johansson, P Garg, PMJ Burgers, The Pol32 subunit of DNA polymerase δ contains separable domains for processive replication and proliferating cell nuclear antigen (PCNA) binding. J Biol Chem 279, 1907–1915 (2004).
29
L Haracska, CM Kondratick, I Unk, S Prakash, L Prakash, Interaction with PCNA is essential for yeast DNA polymerase η function. Mol Cell 8, 407–415 (2001).
30
L Haracska, et al., Physical and functional interactions of human DNA polymerase η with PCNA. Mol Cell Biol 21, 7199–7206 (2001).
31
L Haracska, et al., Targeting of human DNA polymerase ι to the replication machinery via interaction with PCNA. Proc Natl Acad Sci USA 98, 14256–14261 (2001).
32
L Haracska, et al., Stimulation of DNA synthesis activity of human DNA polymerase κ by PCNA. Mol Cell Biol 22, 784–791 (2002).
33
L Haracska, I Unk, L Prakash, S Prakash, Ubiquitylation of yeast proliferating cell nuclear antigen and its implications for translesion DNA synthesis. Proc Natl Acad Sci USA 103, 6477–6482 (2006).
34
N Acharya, et al., Complex formation of yeast Rev1 and Rev7 proteins: A novel role for the polymerase-associated domain. Mol Cell Biol 25, 9734–9740 (2005).
35
N Acharya, L Haracska, S Prakash, L Prakash, Complex formation of yeast Rev1 with DNA polymerase η. Mol Cell Biol 27, 8401–8408 (2007).
36
TA Kunkel, PM Burgers, Dividing the workload at a eukaryotic repilcation fork. Trends Cell Biol 18, 521–527 (2008).
37
SA Nick McElhinney, DA Gordenin, CM Stith, PMJ Burgers, TA Kunkel, Division of labor at the eukaryotic replication fork. Mol Cell 30, 137–144 (2008).
38
ZF Pursell, I Isoz, E-B Lundstrom, E Johansoon, TA Kunkel, Yeast DNA polymerase ε participates in leading-strand DNA replication. Science 317, 127–130 (2007).
39
L Giot, R Chanet, M Simon, C Facca, G Faye, Involvement of the yeast DNA polymerase δ in DNA repair in vivo. Genetics 146, 1239–1251 (1997).
40
CA Torres-Ramos, S Praksah, L Prakash, Requirement of yeast DNA polymerase δ in postreplicational repair of UV-damaged DNA. J Biol Chem 272, 25445–25448 (1997).
41
T Kesti, K Flick, S Keranen, JE Syvaoja, C Wittenberg, DNA polymerase ε catalytic domains are dispensable for DNA replication, DNA repair, and cell viability. Mol Cell 3, 679–685 (1999).
42
A Morrison, AL Johnson, LH Johnston, A Sugino, Pathway correcting DNA replication errors in Saccharomyces cerevisiae. EMBO J 12, 1467–1473 (1993).
43
A Morrison, A Sugino, The 3′ → 5′ exonucleases of both DNA polymerases δ and ε participate in correcting errors of DNA replication in Saccharomyces cerevisiae. Mol Gen Genet 242, 289–296 (1994).
44
IG Minko, MT Washington, L Prakash, S Prakash, RS Lloyd, Translesion DNA synthesis by yeast DNA polymerase η on templates containing N2 -guanine adducts of 1,3-butadiene metabolites. J Biol Chem 276, 2517–2522 (2001).

Information & Authors

Information

Published in

Go to Proceedings of the National Academy of Sciences
Go to Proceedings of the National Academy of Sciences
Proceedings of the National Academy of Sciences
Vol. 106 | No. 24
June 16, 2009
PubMed: 19487673

Classifications

Submission history

Received: February 27, 2009
Published online: June 16, 2009
Published in issue: June 16, 2009

Keywords

  1. DNA polymerase ζ–Rev1–Pol32 complex
  2. Rev1 interaction with Pol32
  3. translesion synthesis

Acknowledgments

We thank Anshu Bhatnagar for participating in the initial studies of Rev1 interaction with Pol32. This work was supported by National Institutes of Health Grant CA107650.

Notes

This article is a PNAS Direct Submission.
This article contains supporting information online at www.pnas.org/cgi/content/full/0902175106/DCSupplemental.

Authors

Affiliations

Narottam Acharya
Department of Biochemistry and Molecular Biology, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-1061
Robert E. Johnson
Department of Biochemistry and Molecular Biology, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-1061
Vincent Pagès
Department of Biochemistry and Molecular Biology, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-1061
Louise Prakash
Department of Biochemistry and Molecular Biology, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-1061
Satya Prakash1 [email protected]
Department of Biochemistry and Molecular Biology, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-1061

Notes

1
To whom correspondence should be addressed. E-mail: [email protected]
Author contributions: N.A., L.P., and S.P. designed research; N.A. and V.P. performed research; R.E.J., V.P., and L.P. contributed new reagents/analytic tools; N.A., R.E.J., and S.P. analyzed data; and N.A., R.E.J., L.P., and S.P. wrote the paper.

Competing Interests

The authors declare no conflict of interest.

Metrics & Citations

Metrics

Note: The article usage is presented with a three- to four-day delay and will update daily once available. Due to ths delay, usage data will not appear immediately following publication. Citation information is sourced from Crossref Cited-by service.


Citation statements




Altmetrics

Citations

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

Cited by

    Loading...

    View Options

    View options

    PDF format

    Download this article as a PDF file

    DOWNLOAD PDF

    Get Access

    Login options

    Check if you have access through your login credentials or your institution to get full access on this article.

    Personal login Institutional Login

    Recommend to a librarian

    Recommend PNAS to a Librarian

    Purchase options

    Purchase this article to get full access to it.

    Single Article Purchase

    Yeast Rev1 protein promotes complex formation of DNA polymerase ζ with Pol32 subunit of DNA polymerase δ
    Proceedings of the National Academy of Sciences
    • Vol. 106
    • No. 24
    • pp. 9537-9932

    Media

    Figures

    Tables

    Other

    Share

    Share

    Share article link

    Share on social media