Our group is advancing the state-of-the-art in the field of biomolecular modeling to tackle fundamental questions into the mechanisms of DNA replication, genome maintenance, gene regulation and the structures/functions of the large nucleo-protein molecular machines that carry out these processes. This work has direct biomedical relevance, specifically, for the etiology of cancer and inherited genetic disorders.
A sampling of recently completed research projects is provided below.
Posttranslational modifications of PCNA and the DNA damage response In this project we employed a hybrid computational protocol to model ubiquitinated and SUMOylated PCNA. Proliferating cell nuclear antigen (PCNA) is a pivotal replication protein, which also controls cellular responses to DNA damage. PCNA binds core replisomal constituents, numerous repair and cell cycle control proteins, acting as a platform for the assembly of the replication machinery. Posttranslational modifications of PCNA by ubiquitin (Ub) and SUMO play a critical role in coordinating DNA damage responses to suppress genome instability. How the modifiers alter PCNA-dependent DNA repair and damage tolerance pathways has so far remained elusive. We used integrative computational modeling methods to identify atomic models of PCNAK107-Ub, PCNAK164-Ub and PCNAK164-SUMO complexes consistent with solution small angle X-ray scattering (SAXS) data. We show that Ub and SUMO have distinct modes of association to PCNA. Ubiquitin adopts discrete docked binding conformations and the position of ubiquitin attachment, 107 versus 164, alters conformation. By contrast, SUMO associates by simple tethering and adopts extended flexible conformations. These structural differences are the result of the opposite electrostatic potentials of SUMO and Ub. The unexpected contrast in conformational behavior of Ub-PCNA and SUMO-PCNA has implications for interactions with partner proteins, interacting surfaces accessibility, and access points for pathway regulation.
- Tsutakawa, S.E., Yan, C., Xu, X., Weinacht, C., Frudental, B., Zhuang, Z., Washington, M.T., Tainer, J.A. & Ivanov, I.* Structurally distinct ubiquitin- and SUMO-modified PCNA: Implications for their distinct roles in the DNA Damage response Structure (2015) 23, 724-733, doi:10.1016/j.str.2015.02.008
Modeling of transcription pre-initiation assemblies in gene regulation
Gene expression, transcriptional regulation and transcription-coupled nucleotide excision repair (TC-NER) are dependent on large complexes involving RNA Polymerase II (Pol lI). The size and complexity of Pol II assemblies complicates mechanistic understanding. To make progress, we collaborated with the groups of Prof. Eva Nogales (Berkeley) and Prof. Yuan He (Northwestern) in an effort to combine advanced computation with cryo-EM to elucidate the structures and mechanisms of Pol II complexes. A critical step preceding transcription is the process of promoter opening resulting in a nascent transcription bubble. This is accomplished by Pol II in association with a multitude of general transcription factors that assemble into a transcription preinitiation complex (PIC). The three-dimensional layout of the eukaryotic PIC is currently under debate. Furthermore, the events and mechanical forces leading to DNA unwinding and transcription bubble opening within the PIC are still insufficiently understood. Promoter opening depends on the transcription factor TFIIH, which contains two helicase subunits XPD and XPB (Figure 1C). While chemical cross-linking and crystallography had provided glimpses into the architecture of yeast Pol II complexes in various functional states, none of the conventional structural approaches had obtained structures of complete PIC assemblies. Recently, cryo-EM density maps from the Nogales group have revealed the organization of the PIC in unprecedented structural detail. We applied integrative computational modeling to construct atomically detailed PIC models in four different functional states along the path from promoter recognition to transcription initiation. Collectively, these models shed light on critical events in the early stages of transcription. Importantly, we also show that the judicious combination of complementary techniques – molecular dynamics flexible fitting (MDFF) and refinement of atomic coordinates in Phenix – leads to pseudoatomic cryo-EM models comparable in quality to crystal structures in the same resolution range.
- He, Y., Yan, C., Inouye, C., Fang, J., Tjian, R., Ivanov, I. & Nogales E. Structural basis of transcription promoter opening using single particle cryo-EM Nature
Evolution and Allostery in Homologous Transcription Factors
Transcription factors (TFs) bind to specific DNA sequences in response to biological signaling pathways to affect transcription activation or repression, depending on the stimulus and target gene. The biophysical basis for the recognition of target DNA sequences by TFs is a difficult question. By studying the dynamics of paralogous TF-DNA complexes selecting in the 3-ketosteroid family, we have identified interesting allosteric differences caused by the interplay between evolutionary mutations, otherwise known as epistasis. By applying selected analytical methodologies from network theory, we have been able to highlight an allosteric rewiring in these homologues that ultimately underlies the observed epistatic relationships between mutations. We are also able to correlate enhanced binding affinity with an increase in network communication throughout the complex. While the roles of allostery and epistasis remain far from being fully understood, our recent collaborative efforts on the 3-ketosteroid TFs clearly demonstrates a strong relationship between the two concepts.
- Hudson, W., Kossmann B., de Vera, I., Chuo, S., Weikum, E., Eick, G., Thornton, J., Ivanov, I., Kojetin, D. & Ortlund, E. Distal substitutions drive divergent sequence specificity among paralogous transcription factors through subdivision of conformational space Proc. Natl. Acad. Sci. (2016) doi:10.1016/pnas.15189600113
A highlight of our work from the Oak Ridge Leadership Computing Facility.