Journal of Molecular Graphics and Modelling 44 (2013) 286–296 Contents lists available at ScienceDirect Journal of Molecular Graphics and Modelling j ourna l h om epa ge: www.elsevier.com/locate/JMGM Insights from simulations into the mechanism of human topoisomerase I: Explanation for a seeming controversy in experiments Neslihan Ucuncuoglu a , Ioan Andricioaei b , Levent Sari a,c,∗ a Department of Physics, Fatih University, Istanbul 34500, Turkey b Department of Chemistry, University of California, Irvine, CA 92697, United States c Medical School, Fatih University, Istanbul 34500, Turkey a r t i c l e i n f o Article history: Accepted 5 July 2013 Available online 24 July 2013 Keywords: Human topoisomerase I DNA supercoiling Molecular dynamics Force-field Mutation a b s t r a c t Human topoisomerase-I is a vital enzyme involved in cellular regulation of DNA supercoiling. We extend our previous work on wild type enzyme [13] to study how different enzyme mutants with various parts of the protein clamped by disulfide mutations affect DNA rotation. Three different mutants have been simulated; they are clamped enzyme-DNA systems in which the disulfide bridge is formed by replac- ing His367 and Ala499, Gly365 and Ser534, and, respectively, Leu429 and Lys436 with Cys pairs. The first of these mutants, a ‘distally clamped’ enzyme, mimics the experimental study of Carey et al. [11], which reports DNA rotation within the clamped enzyme. The second one, a ‘proximal clamp’, mimics the study of Woo et al. [12], who do not observe DNA rotation. The third is a newly suggested mutant that clamps the hinge for protein opening; we use it to test a hypothesis on negative supercoil relaxation. Our simulations show that the helical domain ˛5 totally melts in relaxation of positive supercoils when the enzyme is proximally clamped, while it preserves its structure very well within the distally clamped one. Moreover, a distally clamped protein permits DNA rotations in both directions, while the proximal clamp allows rotations only for negatively supercoiled DNA. These observations reconcile the two seem- ingly contradictory experimental findings, suggesting that subtle changes in the location of the disulfide bridge alter the mechanism significantly. © 2013 Elsevier Inc. All rights reserved. 1. Introduction Topoisomerases (topos) are enzymes that adjust the topology of DNA, with very important roles in replication [1], transcription [2], recombination [3,4], and repair [5]. They relieve supercoils, knots catenanes [6] generated during the genetic transactions in the cell [7,8]. These enzymes are classified as type I or type II, depending on whether they cleave one or both strands of the DNA. Type I topos are further subcategorized as type IA if the enzyme attaches to 5 ′ end of the DNA, or type IB if it the attachment happens to be with the 3 ′ end of the DNA. Human topoisomerase I (a type IB topo) is a monomeric protein of 765 amino acids which is composed of four major regions [8,9]: a N-terminal, a core, a linker, and a C-terminal domain. As illustrated in Fig. 1, the core subdomain I (from Ile215 to Glu232 and from Ser320 to Ser433) and core subdomain II (from Pro233 to Met319) ∗ Corresponding author at: Medical School, Fatih University, Istanbul 34500, Turkey. Tel.: +90 2128663300. E-mail address: lsari@fatih.edu.tr (L. Sari). constitute an ‘upper cap’, which is connected by a flexible ‘hinge’ (a loop approximately from Leu429 to Lys436, between ˇ11 and ˛8) to the bottom part of the clamp that has the C-terminal (Gln713 to Phe765) and the core subdomain III (Arg434 to Ala635). The linker domain is formed by an amino acid sequence from Pro636 to Lys712 (˛18 and ˛19). The ‘nose-cone’ helices, ˛5 (Thr303 to Gln318) and ˛6 (Lys321 to Tyr338) that form a “V” shape structure believed to be important in the topoisomerization mechanism [10] belong to core subdomain II and I, respectively. The upper cap and the lower one interact via the ‘lips’ region opposite to the hinge side. The upper side of the lips is composed of a loop from core subdomain I, and the lower lip of two loops from core subdomain III. The protein is believed [10] to open up by parting the lips region to bind the DNA. Champoux et al. [9,10] provided two crystal structures, pdb codes 1A31 and 1A36. In one of them (pdb code 1A31), the protein is missing the non-essential N-terminal part (the first 214 residues), the linker (residues 636–712), 627–659 of the core and 713–719 of the C-terminal domains, and is covalently bound to a 22 base pair DNA duplex. In the other one (pdb code 1A36), the protein has the full amino acid sequence except the N-terminal part and 634–640 of the core, and bounds to DNA non-covalently. 1093-3263/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jmgm.2013.07.003