Defining the position of the switches between replicative and bypass DNA polymerases

Abstract

Cells contain specialized DNA polymerases that are able to copy past lesions with an associated risk of generating mutations, the major cause of cancer. Here, we reconstitute translesion synthesis (TLS) using the replicative (Pol III) and major bypass (Pol V) DNA polymerases from Escherichia coli in the presence of accessory factors. When the replicative polymerase disconnects from the template in the vicinity of a lesion, Pol V binds the blocked replication intermediate and forms a stable complex by means of a dual interaction with the tip of the RecA filament and the beta-clamp, the processivity factor donated by the blocked Pol III holoenzyme. Both interactions are required to confer to Pol V the processivity that will allow it synthesize, in a single binding event, a TLS patch long enough to support further extension by Pol III. In the absence of these accessory factors, the patch synthesized by Pol V is too short, being degraded by the Pol III-associated exonuclease activity that senses the distortion induced by the lesion, thus leading to an aborted bypass process.

Figure 1

Interplay between polymerases during lesion bypass. All experiments involve circular single-stranded templates (≈2.7 kb) containing a G-AAF adduct located either within the NarI sequence primed with L-6 (A, B) or within the 3G sequence primed with L-8 (C). Standard reaction mixtures containing 50 nM β, 2 μM RecA and 10 nM SSB (see Materials and methods) were preincubated with or without γ complex. When indicated, reactions were initiated by adding Pol III^* (III) for 3 min at 30°C, followed by the addition of Pol II and/or Pol V for an additional 60 min. Incubations with Pol V alone were performed for 15 min. The concentrations of enzymes used are as follows: (A) Pol III^*: 4 nM, Pol II: 4 nM, Pol V: 100 nM; (B) Pol III^*: 62.7 nM, Pol II: 4 nM, Pol V: 100 nM; (C) Pol III^*: 20 nM, Pol II: 4 nM, Pol V: 100 nM. The reaction mixtures were digested by EcoRI located 11 and 14 nt downstream of the lesion in the 3G and NarI contexts, respectively. The reaction products are analyzed by 10% denaturing PAGE.

Figure 2

Processivity of Pol V in the presence of the β-clamp determined by single turnover experiments. Experiments are performed as outlined in the top of the figure. Briefly, Pol V is added to the preincubation mixture containing SSB, RecA, MgCl₂ and (+/−) β/γ as indicated to allow formation of an initiation complex. The reaction is started by simultaneous addition of dNTP+heparin, and then incubated for 0.5–8 min before the reaction is stopped. Heparin is used to trap free Pol V molecules, as symbolized by the crossover free Pol V molecules. The lack of replication products in lanes ‘P', in which dNTP+heparin was added to the preincubation mixture before Pol V, proves the efficiency of the heparin trap (8 min of incubation). Lanes ‘-H', without heparin, show the activity of Pol V under conditions of multiple binding during an 8 min incubation period. With undamaged template in the presence of the β-clamp, low amounts of long elongation products (>40 nt) can be seen at all time points. These products result from a minor polymerase contamination, present in the γ complex preparation, that is stimulated by the β-clamp and inhibited by the G-AAF adduct.

Figure 3

Pol III HE efficiently replicates an ATP-activated RecA filament. A circular single-stranded template (≈2.7 kb) containing a G-AAF adduct located within the NarI sequence primed with L-37 is used as a substrate for Pol III HE elongation experiments in the presence or absence of RecA and SSB. The DNA template was preincubated with SSB and RecA at the indicated concentrations in the presence of β-clamp (50 nM) for 10 min. Elongation reactions were initiated by adding 8 nM Pol III^* and terminated after 13 min. All reaction products were digested by EcoRI and analyzed by a 10% denaturing PAGE. M indicates DNA size markers.

Figure 4

Pol III HE and Pol II require a minimum of 4 or 5 base pairs beyond the lesion to resume efficient DNA synthesis. Substrates are either linear oligonucleotides or circular single-stranded plasmids onto which 5′-end-labeled primers are annealed. For linear oligonucleotides (130-mer), the lesion is located about halfway from both ends of the template (details in Materials and methods). Lesions are G-AAF (D–H), AP site (A), TT cyclobutane dimer (B) and TT (6-4) photoproduct (C). Each panel shows the reaction products using a series of primers a various length ranging from L-2 (24-mer) to positions up to L6 (32-mer). Primers are named according to the position their 3′-end anneals to the template with respect to the lesion site (i.e., the 3′-end nucleotide of primer L2 ends 2 nt beyond the lesion site that is referred to as L0). In (F, H), the G-AAF adduct located within the NarI frameshift hot spot forms a −2 frameshift intermediate (Fuchs et al, 1981; Burnouf et al, 1989); ‘−2 frameshift' primers are named as follows: primer Ln(−2) designates a primer that ends at position n (with respect to the lesion site) in the absence of misalignment. When a −2 nt misalignment intermediate forms, primer Ln(−2) ends at position n+2 as shown in the top of (F, H). All Pol III reactions (A–F) were carried out under the standard conditions containing 50 nM β-clamp and 300 nM SSB (see Materials and methods). Pol II reactions (G, H) did not contain the β-clamp. Reactions were preincubated (10 min at 30°C), initiated by adding 2 nM Pol III^* (A–F) or 4 nM Pol II (G, H) and terminated after 15 min. The reaction products were analyzed by 10% denaturing PAGE. (E) G-AAF in the NarI context is located in a circular template to evaluate the potential effect of circular versus linear templates. Reaction products in (E) were digested with EcoRI before electrophoresis. P and M indicate DNA size markers. For each reaction, the efficiencies of polymerization (% pol) and degradation (% exo) (listed above each panel) were calculated as follows: % pol, the sum of band intensities above position Ln divided by the sum of all bands except Ln; % exo, the sum of band intensities below position Ln divided by the sum of all bands except Ln.

Figure 5

An integrated view of TLS. (A) Pol III^* associated with its processivity factor the β-clamp encounters a noncoding lesion in the template and stops. A RecA filament forms on the single-stranded DNA region downstream the lesion site. This RecA filament together with the β-clamp forms a structure-specific template to which the bypass polymerase Pol V binds and mediates a synthesis patch 20 nt long on average. Pol V produces a large distribution of TLS patches ranging from 1 to 60 nt. (B) Successful lesion bypass requires the TLS patch to be ⩾5 nt long. When Pol III HE binds to the intermediate bypass products generated by Pol V, if the TLS patch is ⩾5 nt, the distortion due to the lesion is outside the ‘sensor domain' of Pol III, allowing efficient elongation to complete a successful bypass process. In contrast, if the TLS patch is <5nt, the distortion triggers primer degradation by the proofreading function leading to an aborted process. The ratio of successful to aborted events is ≈3:1 under the present experimental conditions.