The cell nucleus is highly dynamic in nature. A myriad of motor proteins translocate along, and rotate around, DNA as they separate DNA strands, carry out polymerization reactions, resolve topological issues, repair DNA damage, and modify DNA-binding proteins. All these activities continually subject DNA to forces and torques, causing it to be stretched and twisted, and in turn, DNA architecture regulates motor activities (Figure 1). Thus dynamics constitutes an integral part of the nucleus infrastructure. By investigating one molecular complex at a time, single-molecule approaches provide controlled and quantitative approaches to visualize, measure, and manipulate these processes at unprecedented resolution.
My lab develops innovative techniques to investigate many of these processes at the single-molecule level. These techniques allow us to stretch, unzip, and twist single DNA molecules, to track the motions of motor proteins that translocate along and rotate around DNA, and to measure forces and torques generated by these motors. By applying these single-molecule techniques, we have provided insights into the mechanisms of DNA packaging, transcription, and replication.
Development of Novel Single-Molecule Techniques
A number of our technical innovations have enabled new capabilities in single-molecule manipulation and detection.
DNA-based processes critically rely on sequence-specific binding proteins. We have developed a DNA-unzipping technique that is a versatile and powerful single-molecule method to probe protein-DNA interactions (Figure 2A). Using an optical trap, we can mechanically and sequentially separate a single DNA double helix containing bound proteins into two single strands. When the unzipping fork reaches a bound protein molecule, a dramatic increase in the required unzipping force is detected, followed by a sudden force reduction as the protein is displaced. The unzipping force pattern maps the location of the interaction to near single–base-pair accuracy and the disruption force reveals the strength of the interaction. We have used this approach to study restriction enzymes, DNA repair enzymes, RNA polymerases, and nucleosomes.
Because of the double-helical nature of DNA, motor proteins that translocate along DNA may have to rotate around the DNA helical axis. For example, during transcription, the translocation of RNA polymerase along the double-helical DNA will generate torque to introduce positive supercoils in front and negative supercoils behind. Torque introduced by motor proteins is an essential regulatory factor in biology; however, direct measurements of torque have proved to be challenging. To meet this challenge, we have developed an angular optical trap (AOT) that permits direct and simultaneous measurements of force and torque (Figure 3). A molecule of interest is specifically attached to a nanofabricated quartz cylinder held in the AOT. Torque is applied to the molecule by controlling the orientation of the cylinder via a linearly polarized trapping beam. Using the AOT, we have conducted extensive experiments to characterize the structural properties of DNA as it is subject to twist and torque. We have directly measured the torque required to buckle DNA to form plectonemes, to melt DNA during the B-DNA to L-DNA transition, and to melt DNA during the B-DNA to P-DNA transition. These studies serve as the first step in understanding the role of DNA supercoiling in the regulation of cellular processes. We have furthered the utility of the AOT by also using it to twist DNA molecules containing a single nucleosome. This has allowed us to investigate the regulation of nucleosome stability by torque.
A main limitation of tabletop optical trapping experiments is low throughput—only a single molecule can be manipulated by a focused laser spot. Therefore, to collect data on multiple molecules, experiments must be carried out sequentially, making data collection time-consuming and challenging. We are developing novel bionanophotonics methods for high-throughput optical trapping. This approach promises parallel measurements of multiple molecules on a single silicon chip. Such high-throughput, miniaturized instruments will broaden future applications of optical trapping for both single-molecule and cellular studies.
Mechanical Stability of Nucleosomes
Our extensive studies of the mechanics of nucleosome packaging and remodeling include a number of highlights. (1) By unzipping DNA molecules, each containing a single nucleosome, we generated a high-resolution map for histone-DNA interactions within a nucleosome to near base-pair resolution (Figure 2B). (2) By stretching out nucleosomal arrays, we measured the mechanical stability of nucleosomes and the contributions of histone tails and their acetylation to nucleosomal stability (Figure 4). (3) Our direct measurements of nucleosome repositioning and structure after nucleosome remodeling created the framework for understanding the nature of the obstacle a nucleosome presents to a protein that needs to gain access to nucleosomal DNA.
Mechanism of Transcription
During transcription, RNA polymerase translocates processively along a DNA template while incorporating nucleotides into the nascent transcript RNA. By studying transcription mechanically, at the single-molecule level, we have shown that RNA polymerase is a powerful molecular motor. We developed and verified the first sequence-dependent thermal ratchet model of transcription elongation. This model is able to predict sequence-dependent transcription pausing. In addition, our single-molecule experiments, in which we challenged RNA polymerase with a nucleosome barrier, have shown that multiple RNA polymerases can work synergistically to transcribe nucleosomal DNA (Figure 5).
Unwinding of DNA by Helicase
During DNA replication, helicases catalyze the strand separation of duplex nucleic acids. We have developed single-molecule assays to monitor how T7 helicase, a model hexameric helicase, translocates on DNA and separates the two strands of duplex DNA. We found that T7 helicase unwinds DNA in a force- and sequence-dependent fashion, which is consistent with a model in which T7 helicase actively destabilizes the downstream double-stranded DNA (dsDNA; Figure 6). More recently, we have also determined that T7 helicase can unwind DNA using ATP (contrary to prior belief), but slips frequently. This discovery led to experiments that revealed that all, or nearly all, helicase subunits coordinate nucleoside triphosphate (NTP) binding and catalysis, and DNA affinity, all in such a way as to maintain processivity (Figure 7).
This work was supported in part by grants from the National Institutes of Health and the National Science Foundation.
As of March 24, 2016