† Corresponding author. E-mail:
Project supported by the National Natural Science Foundation of China (Grant Nos. 11674382, 11574381, and 11574382) and the Key Research Program of Frontier Sciences, Chinese Academy of Sciences (Grant No. QYZDJ-SSW-SYS014).
The Bloom helicase (BLM) gene product encodes a DNA helicase that functions in homologous recombination repair to prevent genomic instability. BLM is highly active in binding and unfolding G-quadruplexes (G4), which are non-canonical DNA structures formed by Hoogsteen base-pairing in guanine-rich sequences. Here we use single-molecule fluorescence resonance energy transfer (smFRET) to study the molecular mechanism of BLM-catalysed G4 unfolding and show that BLM unfolds G4 in two pathways. Our data enable us to propose a model in which the HRDC domain functions as a regulator of BLM, depending on the position of the HRDC domain of BLM in action: when HRDC binds to the G4 sequence, BLM may hold G4 in the unfolded state; otherwise, it may remain on the unfolded G4 transiently so that G4 can refold immediately.
The Bloom helicase (BLM) belongs to the RecQ helicase family.[1,2] It contains two RecA-like domains which are involved in DNA binding and hydrolysis of adenosine triphosphate(ATP) and provides the ability to translocate the BLM on single-stranded DNA (ssDNA) in the 3′ to 5′ directions.[2,3] Deficiencies in the BLM could cause a serious genetic disease named Bloom syndrome (BS), the characteristics of which include genome instability, dwarfism, immunodeficiency, reduced fertility, and elevated levels of many types of cancers.[4,5] Cells from patients with BS show chromosomal instability characterized by higher rates of chromatid gaps and breaks, and sister chromatid exchanges.[6,7] BLM can resolve many different DNA substrates, such as Watson–Crick’s duplex DNA, G-quadruplexes (G4)[8–10] and Holliday junctions.[11–13] It is generally accepted that BLM is ATP-dependent as it catalyzes G4 unfolding because it is an ATP hydrolysis-driven helicase.[10] However, a study showed that co-incubation of G4 with very a high concentration BLM (100 nM–2 M) may lead to 16%–45% unfolding of G4 in the absence of ATP or in the presence of the non-hydrolysable ATP analog ATP-γ-S.[9]
BLM contains an HRDC domain which is specific to the RecQ family. Previous studies showed the HRDC domain of the BLM works mainly as an assistant in DNA binding and translocation along ssDNA.[14,15] However, BLM mutants lacking HRDC show deficiency in strand annealing[16] and double Holliday junction dissolution[17] while they act similarly as the wild type.[16–18] The BLM binds G4 structures with high specificity.[19] BLM without the HRDC domain can also unfold G4.[18,20] It therefore seems that HRDC is not necessary for BLM to unfold G4. Yet there are reports that HRDC binds to ssDNA,[21,22] to assist the BLM to dissolve G4.[20] It was also proposed that HRDC could regulate the process of ATP hydrolysis of the BLM.[23] Here we use single-molecule fluorescence resonance energy transfer (smFRET) to study the interaction between BLM and G4. Surprisingly, we find that BLM with and without the HRDC domain unfolds G4 differently. Further studies indicate that the difference depends on the concentration of ATPs. We propose a model for explaining the results and make a suggestion that the HRDC domain is needed by BLM to hold the G4 in the unfolded state, possibly waiting for other proteins to act on the G4 sequence.
G4 is a non-canonical structure formed by Hoogsteen base-pairing in guanine-rich DNA sequences.[24,25] These regions of genome are significant for maintaining the human genome. In our experiments, G4 was composed of a DNA sequence of (GGGTTA)
BLM mutations (Figs.
All smFRET experiments were carried out with a home-built objective-type total-internal-reflection microscope at room temperature. Coverslips (Fisher Scientific) and slides were cleaned thoroughly by a mixture of sulfuric acid and hydrogen peroxide, then the surfaces of the coverslips were coated with a mixture of 99% monomethoxy-polyethylene glycol (m-PEG-5000, Laysan Bio, Inc.) and 1% of biotin-PEG (biotin-PEG-5000, Laysan Bio, Inc.). Streptavidin (1 mg/mL) in buffer containing 50-mM NaCl, 20-mM Tris-HCl, pH7.5 was added to the microfluidic chamber made of the PEG-coated coverslip and was incubated for 10 min. After washing, 20 pM–50 pM DNAs were added to the chamber and allowed to be immobilized for 10 min. Then free DNA was removed by washing with the reaction buffer. After that, the chamber was filled with the reaction buffer with an oxygen scavenging system (0.8% D-glucose, 1-mg/mL glucose oxidase, 0.4-mg/mL catalase, and 1-mM Trolox). The exposure time is 100 ms in all of our experiments. The raw fluorescence intensity trajectories were three-point averaged. Then, the FRET efficiency was calculated by using
We first measure the activity of 2-nM BLM at 200-μM ATP which is nearly the saturation concentration. About 15% of the substrates could be unfolded by BLM. Upon the addition of BLM, we observe repetitive unfolding signals (Figs.
The results above show that the BLM helicase either unfolds a G4 repetitively or holds it in the unfolded state for a long time. Now, we come to see whether such a feature of the BLM would affect its unwinding of a DNA duplex (Figs.
We lower the ATP concentrations to 20 μM and 1 μM respectively, and repeat the unfolding experiments. The phenomenon of long-time dwelling almost disappears, while BLM still repetitively unfold G4 (Fig.
We use a BLM mutation with the HRDC domain being truncated (BLM
The HRDC domain has long been a mystery for the RecQ family helicases. BLM has been found to be able to unfold G4 even in the absence of HRDC. HRDC is also reported to be not necessary for BLM to unwind DNA duplexes.[18] However, our present study indicates that HRDC is not just a redundant domain of BLM. We observe that BLM with the HRDC domain could hold G4 in an unfolded state for a very long time. Without the HRDC domain, the binding of BLM to the G4 sequence is so transient that G4 can refold immediately after it has been unfolded by BLM. A recent structural study shows that the HRDC domain is connected to the core of BLM by a long random coil so that HRDC can move freely around the core.[23] Based on our experimental results and the structure of BLM, we propose a model for BLM unfolding G4 (Fig.
Our results also show that ATP hydrolysis is needed for BLM to hold the unfolded G4 firmly. We propose that the ATP hydrolysis should cause the BLM core to be translocated on the DNA overhang. But the motion is blocked by the HRDC ahead of it so that the DNA overhang is pushed way, hence preventing G4 from refolding (Fig.
[1] | |
[2] | |
[3] | |
[4] | |
[5] | |
[6] | |
[7] | |
[8] | |
[9] | |
[10] | |
[11] | |
[12] | |
[13] | |
[14] | |
[15] | |
[16] | |
[17] | |
[18] | |
[19] | |
[20] | |
[21] | |
[22] | |
[23] | |
[24] | |
[25] | |
[26] | |
[27] | |
[28] |