Multifaceted role of the Topo IIIα-RMI1-RMI2 complex and DNA2 in the BLM-dependent pathway of DNA break end resection

Abstract

BLM, a RecQ family DNA helicase mutated in Bloom's Syndrome, participates in homologous recombination at two stages: 5' DNA end resection and double Holliday junction dissolution. BLM exists in a complex with Topo IIIα, RMI1 and RMI2. Herein, we address the role of Topo IIIα and RMI1-RMI2 in resection using a reconstituted system with purified human proteins. We show that Topo IIIα stimulates DNA unwinding by BLM in a manner that is potentiated by RMI1-RMI2, and that the processivity of resection is reliant on the Topo IIIα-RMI1-RMI2 complex. Topo IIIα localizes to the ends of double-strand breaks, thus implicating it in the recruitment of resection factors. While the single-stranded DNA binding protein RPA plays a major role in imposing the 5' to 3' polarity of resection, Topo IIIα also makes a contribution in this regard. Moreover, we show that DNA2 stimulates the helicase activity of BLM. Our results thus uncover a multifaceted role of the Topo IIIα-RMI1-RMI2 ensemble and of DNA2 in the DNA resection reaction.

Figure 1.

Topo IIIα-RMI1-RMI2 stimulates DNA unwinding by BLM-RPA. (A) The effects of Topo IIIα (1 or 5 nM) and RMI1-RMI2 (1 or 5 nM) on the unwinding of a 2 kb randomly radiolabeled dsDNA (0.5 nM ends) by BLM (1 nM) and RPA (200 nM) were examined. The reaction was incubated at 37°C for 30 min. In lane 1, the DNA was heat denatured (HD) by boiling for 2 min. The data shown are the average from three independent experiments with the error bars representing one standard deviation. (B) Topo IIIα (wild-type or Y133F) (5 nM) was incubated with BLM (2.5 nM) and RPA (200 nM) for 30 min at 37°C. (C) Yeast Top3 (1 or 5 nM) was incubated with BLM (4 nM) for 10 min at 37°C. (D) TR (15, 30 or 60 nM) was incubated with WRN (30 nM) and RPA (200 nM) for 30 min at 37°C.

Figure 2.

Recognition of DNA ends by Topo IIIα. (A) Topo IIIα (25, 50 or 100 nM) was incubated with radiolabeled 80-mer dsDNA (5 nM ends). Quantification of DNA binding is shown in the right panel. The arrow denotes nucleoprotein complexes and the asterisk indicated the location of the radiolabel in the DNA substrate. (B) DNA mobility shift was conducted as in (A), except that dsDNA substrate blocked by biotin-streptavidin (denoted by the circled S) at either one or both of the ends was used. The data shown were the average from three independent experiments and the error bars represent 1 SD. (C, D) AFM was used to image nucleoprotein complexes of Topo IIIα and Top3. Arrows indicate DNA end binding events.

Figure 3.

Stimulation of BLM helicase processivity by the TR complex and DNA2-D277A. (A) BLM-mediated DNA unwinding was examined with 5 nM BLM and the addition of DNA2-D277A or DNA2-D277A/K654E (5 or 20 nM). Reactions were incubated for 10 min at 37°C (B) Binding of an 80-mer dsDNA by BLM (7.5, 10.5 or 15 nM) without or with DNA2-D277A (10 or 20 nM). The data shown are the average from three independent experiments and the error bars represent 1 SD. (C) A randomly radiolabeled 2 kb dsDNA (0.5 nM ends) was incubated with BLM (5.5 nM) and RPA (200 nM), and 10-fold excess unlabeled dsDNA (identical to the labeled substrate in sequence) was added at the 2-min point along with the TR complex (5 nM) or DNA2-D277A (16.5 nM). Samples were taken at 2.5 min intervals and analyzed. Quantification of the results is shown below. (D) Resection of a 2 kb dsDNA (0.5 nM ends) by BLM (2 nM), DNA2 (10 nM) and RPA (200 nM) was examined without or with the TR complex (5 nM). The average data from three independent experiments are shown and the error bars represent 1 SD.

Figure 4.

Effect of Topo IIIα and RMI1-RMI2 on DNA2 endonuclease activity. (A) DNA2 (10 or 20 nM protein for the substrates with the 5′ or 3′ overhang, respectively) was incubated with RPA (5 or 10 nM) or Topo IIIα (1 or 5 nM) and RMI1-RMI2 (1 or 5 nM) for 15 min at 30°C. The average data from three independent experiments are shown and the error bars represent 1 SD. The radiolabel is denoted by the asterisk. (B) DNA substrates with either a 19- or 44-nt 3′ or 5 overhang were incubated with DNA2 at the indicated concentrations. The radiolabel is denoted by the asterisk. (C) DNA2 (5 nM), RPA (5 nM), Topo IIIα (30 nM) and RMI1-RMI2 (30 nM) were incubated with Y DNA substrates containing 19-base overhangs for 20 min at 30°C. The radiolabel is denoted by the asterisk. Degradation products are pictured adjacent to their positions on the gel. (D, B) DNA2 (1 nM), RPA (5 nM), Topo IIIα (30 nM) and RMI1-RMI2 (30 nM) were incubated with a forked DNA substrate containing 44-base overhangs. The substrate was labeled (denoted by the asterisk) on the end of the 5′ overhang (D) or 3′ overhang (E). (F) Model for the multifaceted role of the TR complex and DNA2 in end resection. Topo IIIα and DNA2 contribute to BLM recruitment to DNA ends (top), and when unwinding begins, TR and DNA2 both enhance the processivity of BLM (bottom), where the double arrow indicates dissociation of BLM from the DNA. RPA enforces the 5′ to 3′ directionality of resection by modulating the endonuclease activity of DNA2 (bottom), with Topo IIIα also contributing in this regard.