Search

Single molecule biophysics experiments for the study of DNA-protein interactions usually require production of a homogeneous population of long DNA molecules with controlled sequence content and/or internal tertiary structures. Traditionally, Lambda phage DNA has been used for this purpose, but it is difficult to customize. In this article, we provide a detailed and simple protocol for cloning large (~ 25 kbp) plasmids with bespoke sequence content, which can be used to generate custom DNA constructs for a range of single-molecule experiments. In particular, we focus on a procedure for making long single-stranded DNA (ssDNA) molecules, ssDNA-dsDNA hybrids and long DNA constructs with flaps, which are especially relevant for studying the activity of DNA helicases and translocases. Additionally, we describe how the modification of the free ends of such substrates can facilitate their binding to functionalized surfaces allowing immobilization and imaging using dual optical tweezers and confocal microscopy. Finally, we provide examples of how these DNA constructs have been applied to study the activity of human DNA helicase B (HELB). The techniques described herein are simple, versatile, adaptable, and accessible to any laboratory with access to standard molecular biology methods.

DNA2 is an essential enzyme involved in DNA replication and repair in eukaryotes. In a search for homologues of this protein, we identified and characterised Geobacillus stearothermophilus Bad, a bacterial DNA helicase-nuclease with similarity to human DNA2. We show that Bad contains an Fe-S cluster and identify four cysteine residues that are likely to co-ordinate the cluster by analogy to DNA2. The purified enzyme specifically recognises ss-dsDNA junctions and possesses ssDNA-dependent ATPase, ssDNA binding, ssDNA endonuclease, 5' to 3' ssDNA translocase and 5' to 3' helicase activity. Single molecule analysis reveals that Bad is a processive DNA motor capable of moving along DNA for distances of >4 kb at a rate of ∼200 bp per second at room temperature. Interestingly, as reported for the homologous human and yeast DNA2 proteins, the DNA unwinding activity of Bad is cryptic and can be unmasked by inactivating the intrinsic nuclease activity. Strikingly, our experiments show that the enzyme loops DNA while translocating, which is an emerging feature of processive DNA unwinding enzymes. The bacterial Bad enzymes will provide an excellent model system for understanding the biochemical properties of DNA2-like helicase-nucleases and DNA looping motor proteins in general.

CtIP is involved in the resection of broken DNA during the S and G2 phases of the cell cycle for repair by recombination. Acting with the MRN complex, it plays a particularly important role in handling complex DNA end structures by localised nucleolytic processing of DNA termini in preparation for longer range resection. Here we show that human CtIP is a tetrameric protein adopting a dumbbell architecture in which DNA binding domains are connected by long coiled-coils. The protein complex binds two short DNA duplexes with high affinity and bridges DNA molecules in trans. DNA binding is potentiated by dephosphorylation and is not specific for DNA end structures per se. However, the affinity for linear DNA molecules is increased if the DNA terminates with complex structures including forked ssDNA overhangs and nucleoprotein conjugates. This work provides a biochemical and structural basis for the function of CtIP at complex DNA breaks.

Bacillus subtilis ParB forms multimeric networks involving non-specific DNA binding leading to DNA condensation. Previously, we found that an excess of the free C-terminal domain (CTD) of ParB impeded DNA condensation or promoted decondensation of pre-assembled networks (Fisher et al., 2017). However, interpretation of the molecular basis for this phenomenon was complicated by our inability to uncouple protein binding from DNA condensation. Here, we have combined lateral magnetic tweezers with TIRF microscopy to simultaneously control the restrictive force against condensation and to visualise ParB protein binding by fluorescence. At non-permissive forces for condensation, ParB binds non-specifically and highly dynamically to DNA. Our new approach concluded that the free CTD blocks the formation of ParB networks by heterodimerisation with full length DNA-bound ParB. This strongly supports a model in which the CTD acts as a key bridging interface between distal DNA binding loci within ParB networks.

Combining single-molecule techniques with fluorescence microscopy has attracted much interest because it allows the correlation of mechanical measurements with directly visualized DNA:protein interactions. In particular, combination with total internal reflection fluorescence microscopy (TIRF) is advantageous because of the high signal-to-noise ratio this technique achieves. This, however, requires stretching long DNA molecules across the surface of the flow cell to maximize polymer exposure to the excitation light. In this work, we develop a module to laterally stretch DNA molecules at a constant force, which can be easily implemented in regular or combined magnetic tweezers (MT)-TIRF setups. The pulling module is further characterized in standard flow cells of different thicknesses and glass capillaries, using two types of micrometer size superparamagnetic beads, long DNA molecules, and a home-built device to rotate capillaries with mrad precision. The force range achieved by the magnetic pulling module was between 0.1 and 30 pN. A formalism for estimating forces in flow-stretched tethered beads is also proposed, and the results compared with those of lateral MT, demonstrating that lateral MT achieve higher forces with lower dispersion. Finally, we show the compatibility with TIRF microscopy and the parallelization of measurements by characterizing DNA binding by the centromere-binding protein ParB from Bacillus subtilis. Simultaneous MT pulling and fluorescence imaging demonstrate the non-specific binding of BsParB on DNA under conditions restrictive to condensation.

The ParB protein forms DNA bridging interactions around parS to condense DNA and earmark the bacterial chromosome for segregation. The molecular mechanism underlying the formation of these ParB networks is unclear. We show here that while the central DNA binding domain is essential for anchoring at parS, this interaction is not required for DNA condensation. Structural analysis of the C-terminal domain reveals a dimer with a lysine-rich surface that binds DNA non-specifically and is essential for DNA condensation in vitro. Mutation of either the dimerisation or the DNA binding interface eliminates ParB-GFP foci formation in vivo. Moreover, the free C-terminal domain can rapidly decondense ParB networks independently of its ability to bind DNA. Our work reveals a dual role for the C-terminal domain of ParB as both a DNA binding and bridging interface, and highlights the dynamic nature of ParB networks in Bacillus subtilis.