Single-Molecule Supercoil Relaxation Assay as a Screening Tool to Determine the Mechanism and Efficacy of Human Topoisomerase IB Inhibitors

Abstract

Human nuclear type IB topoisomerase (Top1) inhibitors are widely used and powerful anticancer agents. In this study, we introduce and validate a single-molecule supercoil relaxation assay as a molecular pharmacology tool for characterizing therapeutically relevant Top1 inhibitors. Using this assay, we determined the effects on Top1 supercoil relaxation activity of four Top1 inhibitors; three clinically relevant: camptothecin, LMP-400, LMP-776 (both indenoisoquinoline derivatives), and one natural product in preclinical development, lamellarin-D. Our results demonstrate that Top1 inhibitors have two distinct effects on Top1 activity: a decrease in supercoil relaxation rate and an increase in religation inhibition. The type and magnitude of the inhibition mode depend both on the specific inhibitor and on the topology of the DNA substrate. In general, the efficacy of inhibition is significantly higher with supercoiled than with relaxed DNA substrates. Comparing single-molecule inhibition with cell growth inhibition (IC50) measurements showed a correlation between the binding time of the Top1 inhibitors and their cytotoxic efficacy, independent of the mode of inhibition. This study demonstrates that the single-molecule supercoil relaxation assay is a sensitive method to elucidate the detailed mechanisms of Top1 inhibitors and is relevant for the cellular efficacy of Top1 inhibitors.

Figure 1

Comparison of Top1 mediated DNA cleavage induced by camptothecin (CPT), LMP-400, LMP-776 and lamellarin D (LamD). (a) Compound structures. (b) (lane 1) DNA alone; (lane 2) Top1 alone; (lane 3–16) Top1 + indicated drugs at 0.3, 0.9, 2.7, and 8.1 μM, respectively. The numbers on the left and arrows indicate cleavage site positions. A 117-bp DNA substrate was used in this assay, but the cleavage sites are numbered to be consistent with the 161-bp DNA substrate traditionally used in this assay in order to facilitate comparison (25).

Figure 2

Overview of single-molecule Top1 DNA supercoil relaxation and inhibition assay. (a) Cartoon of magnetic tweezers instrument (not to scale). A magnetic bead is attached to the surface via a rotationally constrained DNA molecule. DNA tension and supercoiling are controlled by the height and rotation of magnets above the sample chamber. During Top1 supercoil relaxation assay, the DNA is supercoiled by rotating the magnets, which results in a linear decrease in the DNA extension once the DNA forms a plectoneme. As Top1 relaxes DNA supercoils, the DNA extension increases, indicated as the unwinding phase (blue dashed line). (b) Two distinctive effects of inhibitors on Top1 catalytic activity: First, the inhibitor can prevent DNA religation, resulting in relatively long but finite duration (life time = τ_rf) events during which the DNA cannot be supercoiled (Top panel). Second, the inhibitor can decrease the relaxation rate. The red trace shows DNA relaxation without CPT for comparison with the black trace that shows dramatically reduced relaxation rate in the presence of CPT. The duration of these slow relaxation-rate events is τ_sl (Bottom panel).

Figure 3

Single-molecule measurements of DNA topology-dependent Top1 inhibition. (a) Two modes of Top1 inhibition and effective binding probability as a function of inhibitor concentration and chirality of DNA supercoiling. Religation inhibition (Top panel). Note that religation inhibition values are relative to values in the absence of inhibitor. Relative change in the probability of low relaxation-rate events (Middle panel). Inhibitor binding determined from combining the probabilities of religation inhibition and low relaxation-rate events (Bottom panel). (b) Inhibitor binding time for different concentrations of each inhibitor for both positively and negatively supercoiled DNA.

Figure 4

DNA topology modulates the level of Top1cc formation induced by Top1 inhibitors. (a) Visualization of Top1cc formation with increasing CPT concentration (0– 50 μM) with three different DNA substrates: negative supercoiled pBR322 (top), positive supercoiled pBR322 (middle), and relaxed pBR322 DNA (bottom). (b) Fraction of nicked DNA as a function of drug concentration for three different DNA topological substrates. The amount of nicked DNA indicates the amount of Top1cc trapped by an inhibitor. Red dotted line indicates no difference in the fraction of nicked DNA relative to measurements without inhibitor.

Figure 5

Correlation between single-molecule measures of inhibition and cellular measures of cytotoxicity. (a) The drug off-rate (k_drug-off, solid circles), the inverse drug binding probability (1/P_drug-binding, solid triangles), and the inverse nicked DNA fraction (1/F_nicked, solid squares) for each inhibitor plotted versus IC₅₀ (P388 cell line). Individual linear fits provide the Pearson correlation coefficient ranging from 0 to 1, P_r, as the level of linear correlation between in-vitro quantities and cytotoxicity. P_r measured for k_drug-off was 0.97, for 1/P_drug-binding was 0.97, and for 1/F_nicked was 0.96. (b) k_drug-off (solid circles), 1/P_drug-binding (solid triangles), and 1/F_nicked (solid squares) for each inhibitor plotted versus <IC_50/CPT>. <IC_50/CPT> indicates the average IC₅₀ relative to that of CPT for each cell line. The x-axis error bars correspond to the standard error from IC_50/CPT values of different cell lines for individual inhibitors. P_r measured for k_drug-off was 0.97, for 1/P_drug-binding was 0.95, and for 1/F_nicked was 0.97. Individual IC₅₀ values and their sources are listed in Table 1.