The Bloom's syndrome helicase promotes the annealing of complementary single-stranded DNA

Abstract

The product of the gene mutated in Bloom's syndrome, BLM, is a 3'-5' DNA helicase belonging to the highly conserved RecQ family. In addition to a conventional DNA strand separation activity, BLM catalyzes both the disruption of non-B-form DNA, such as G-quadruplexes, and the branch migration of Holliday junctions. Here, we have characterized a new activity for BLM: the promotion of single-stranded DNA (ssDNA) annealing. This activity does not require Mg(2+), is inhibited by ssDNA binding proteins and ATP, and is dependent on DNA length. Through analysis of various truncation mutants of BLM, we show that the C-terminal domain is essential for strand annealing and identify a 60 amino acid stretch of this domain as being important for both ssDNA binding and strand annealing. We present a model in which the ssDNA annealing activity of BLM facilitates its role in the processing of DNA intermediates that arise during repair of damaged replication forks.

Figure 1

Inhibition of BLM helicase activity occurs at high protein concentrations, which is relieved by ssDNA binding proteins. (a) Unwinding of 1 nM of a 31 bp forked duplex catalyzed by different concentrations of BLM, as indicated above the lanes. (b) Unwinding of 1 nM of a 31 bp forked duplex catalyzed by BLM in the presence of 60 nM SSB or (c) in the presence of 3 nM RPA. In (a–c), the positions of the forked duplex and the ssDNA product of unwinding are indicated on the left. (d) Quantification of the helicase activity of BLM from the data in (a–c). All reactions were incubated for 30 min at 37°C.

Figure 2

BLM promotes annealing of ssDNA. (a) Effect of BLM concentration on the annealing of two ssDNA molecules to generate a forked duplex. Reactions were incubated for 30 min. (b) Time course of ssDNA annealing in reactions containing 20 nM BLM. (c) Effect of increasing concentrations of RPA on ssDNA annealing catalyzed by 10 nM BLM. In (a–c), the percentage of ssDNA annealed was quantified and the data are presented graphically on the right of the corresponding dataset.

Figure 3

Effect of Mg²⁺ and adenine nucleotide co-factors on ssDNA annealing. (a) Strand annealing by BLM (10 nM) as a function of increasing Mg²⁺ concentration. The reaction shown in the left panel was performed in the absence of BLM. (b) Strand annealing by BLM (10 nM) as a function of increasing concentration of ATP, ATPγS or ADP, as indicated. Graph below shows quantification of the data.

Figure 4

Strand annealing of different ssDNA structures by BLM. (a) In each case, different concentrations of BLM (indicated above the lanes) were incubated with the ssDNA species indicated on the left of each autoradiogram. The positions of the unannealed ssDNA and the annealed fully duplex or partial duplex products are indicated on the left. (b) Quantification of the data in (a). (c) Gel retardation assays with increasing quantities of BLM and the ssDNA oligonucleotides indicated above the wells. The positions of the ssDNA and the retarded BLM–DNA complexes are shown on the right. (d) Quantification of the data from (c).

Figure 5

Schematic representation of the full-length BLM (1–1417) and truncated BLM variants used in this study. Amino acid residue numbers of each protein are indicated on the left. The positions of the helicase (red), RQC (yellow), HRDC (gray) domains are indicated. Green boxes denote poorly conserved regions.

Figure 6

The C-terminal domain of BLM, between residues 1290 and 1350, is required for ssDNA annealing. (a) Strand annealing as a function of increasing protein concentration for BLM213–1417 and BLM213–1267. Reactions were incubated for 30 min. The graph below shows quantification of the data. (b) Time course of ssDNA annealing by 20 nM BLM213–1417 or BLM213–1267. The graph below shows quantification of the data. (c) Comparison of the helicase activity of BLM213–1417 and BLM213–1267. Assays were as described in Figure 1. Graph below shows quantification of the data. (d) Strand annealing as a function of increasing protein concentration for BLM642–1290 and BLM642–1350. Graph below shows quantification of the data.

Figure 7

The C-terminal domain of BLM is required for the formation of higher-order protein–DNA complexes. (a) Gel retardation assays of BLM1–1417, BLM213–1417, BLM213–1267 (upper panel), BLM642–1417, BLM642–1350 and BLM642–1290 (lower panel) using a 50mer ssDNA oligonucleotide. The positions of the gel wells and of BLM–DNA complexes that could be resolved in the gel are indicated on the right. (b) Quantification of total ssDNA bound by BLM and its truncated derivatives. (c) Quantification of the ssDNA that was retarded in the gel wells. For clarity, the data are represented on two graphs in each case.

Figure 8

Models for the possible role of the strand annealing function of BLM in replication fork maintenance. (a) Branch migration model. Schematic representation of how torsional stress and steric hindrance may impede the convergence of two Holliday junctions by branch migration. The strand annealing activity of BLM acting on strands behind each junction may act to overcome these constraints and thereby facilitate the juxtaposition of two Holliday junctions. See text for details. (b) Fork regression model. Template strands are shown in black and nascent strands in red. The yellow triangle depicts a fork-blocking adduct on the leading strand template. See text for details.