Yeast Rad5 Protein Required for Postreplication Repair Has a DNA Helicase Activity Specific for Replication Fork Regression

Summary Lesions in the template DNA strand block the progression of the replication fork. In the yeast Saccharomyces cerevisiae, replication through DNA lesions is mediated by different Rad6-Rad18-dependent means, which include translesion synthesis and a Rad5-dependent postreplicational repair pathway that repairs the discontinuities that form in the DNA synthesized from damaged templates. Although translesion synthesis is well characterized, little is known about the mechanisms that modulate Rad5-dependent postreplicational repair. Here we show that yeast Rad5 has a DNA helicase activity that is specialized for replication fork regression. On model replication fork structures, Rad5 concertedly unwinds and anneals the nascent and the parental strands without exposing extended single-stranded regions. These observations provide insight into the mechanism of postreplicational repair in which Rad5 action promotes template switching for error-free damage bypass.


Supplemental Result: Rad5 Binding to DNA Structures
To determine the binding specificity of Rad5 to various DNA structures we carried out gel-shift assays. We found that from a mixture containing equal amounts of radioactively labelled single stranded and partial heteroduplex DNAs, only the partial heteroduplex was bound by Rad5 whereas the amount of free single-stranded DNA did not change ( Supplementary Fig. 1A). Next, we checked a variety of DNA structures, including Yfork, three-way junction, and four-way junction and found that Rad5 was able to bind all these DNAs . Since 50% binding of all these DNA substrates occurred at a very similar Rad5 concentration, Rad5 bound these various DNAs with fairly similar affinity. This also indicates that the lack of helicase activity on heteroduplexes and Y-fork is not the consequence of low affinity binding of Rad5 to these substrates.

Figure S1. Rad5 Binding to a Variety of DNA Structures
Gel-shift assays were performed in buffer B containing 20 mM Tris-HCl, pH 7.0, 2 mM MgCl 2 , 0.1 mg/ml BSA, 1mM DTT, and 10% glycerol using 0.5 nM 32 P-labeled DNA and Rad5 in two-fold increments up to 70 nM. Reactions were assembled on ice and incubated for 15 min before loading onto a 4% native polyacrylamide gel, containing acrylamide and N,N bis-acrylamide in 30:0.8 ratio, 0.5x Tris-borate, and 2.5% glycerol, followed by electrophoresis at 4°C in 0.5x Tris-borate buffer containing no EDTA. The suggestive of the binding of more than one Rad5 molecule were also seen.

Supplemental Result: Rad5 Cannot Act as a Canonical DNA Helicase
In order to test if Rad5 could act as a canonical DNA helicase we tested its unwinding activity on short heteroduplexes with either 5' or 3' protruding end. Considering the possibility that detecting a weak helicase activity might be hindered by rapid reannealing of the removed strand to its complementary sequence, we also performed these reactions in the presence of E. coli ssDNA binding protein (SSB) that would bind to the released ssDNA thereby preventing reannealing. As shown on Fig. S2A and B, Rad5 did not exhibit any unwinding activity on these substrates. Also, Rad5 was not able to unwind heterologous replication fork-like substrates where only the nascent leading or the nascent lagging strand was present ( Fig. S2C and D, respectively). These results are in agreement with the previously reported data (Johnson et al. 1994) and support the notion that Rad5 does not possess canonical DNA helicase activity.

Replication Fork-like Structures
Helicase activity of Rad5 was tested on partial heteroduplex DNAs with 5' or 3' overhangs (A and B, respectively) and partial heterologous replication fork-like structures where only the nascent leading or the nascent lagging strand was present (C and D, respectively). Where indicated E. coli SSB was used at 160nM concentration which was added as the last component to the reaction mix followed by 5 minutes incubation at 0 °C before initiating the reaction at 37 °C. The concentration of Rad5, the structure and the position of the DNA substrate, and the mobility of the ss30 DNA are indicated. The positions of the radioactive labels are marked with asterisks.
The plasmid-sized replication fork model substrate was created essentially as described (Ralf et al., 2006). Briefly, pG46 and pG68 plasmids were gapped by digestion with nicking endonucleases Nt.BbvCI and Nb.BbvCI yielding pG46B and pG68A, respectively. pG46B was then treated with shrimp alkaline phosphatase and subsequently labelled with T4 polynucleotide kinase to yield the plasmid pG46B'. pG68A was linearized with XhoI digestion to yield pG68A Xh. The resulting plasmids containing complementary single stranded gaps were then annealed together at 53 o C in 50 mM Tris-HCl pH 7.5, 10 mM MgCl 2 , 5mM DTT to form a joint molecule. Note, that the joint molecule contains a structural mimic of a stalled replication fork in which the labelled lagging strand is longer by 14 nucleotides. pG68SapI(Het) was created from pG68 by inserting a small sequence heterology by cloning O1301/O1302 heteroduplex oligonucleotides into the SapI site of pG68. To yield pG68ASapI(Het)Xh, pG68SapI(Het) was treated similar to pG68A. We note that in the original description of this experimental system the joint was formed by annealing the two gapped circular plasmids which was subsequently converted to a plectonemic joint by topoisomerase treatment (Ralf et al., 2006). Unlike the RecQ helicases, Rad5 can not dissolve model forks to parental duplexes and therefore, this conversion was not necessary for our purposes. Thus, we think that the joint configuration shown on Fig. 3A should be considered as a more stringent situation for studying fork reversal than the previously described one.