Topobexin targets the Topoisomerase II ATPase domain for beta isoform-selective inhibition and anthracycline cardioprotection

Abstract

Topoisomerase II alpha and beta (TOP2A and TOP2B) isoenzymes perform essential and non-redundant cellular functions. Anthracyclines induce their potent anti-cancer effects primarily via TOP2A, but at the same time they induce a dose limiting cardiotoxicity through TOP2B. Here we describe the development of the obex class of TOP2 inhibitors that bind to a previously unidentified druggable pocket in the TOP2 ATPase domain to act as allosteric catalytic inhibitors by locking the ATPase domain conformation with the capability of isoform-selective inhibition. Through rational drug design we have developed topobexin, which interacts with residues that differ between TOP2A and TOP2B to provide inhibition that is both selective for TOP2B and superior to dexrazoxane. Topobexin is a potent protectant against chronic anthracycline cardiotoxicity in an animal model. This demonstration of TOP2 isoform-specific inhibition underscores the broader potential to improve drug specificity and minimize adverse effects in various medical treatments.

Fig. 1. TOP2 catalytic inhibition protects primary rat cardiomyocytes from toxicity of daunorubicin.

a General strategy for development of improved TOP2 inhibitors. Created in BioRender. Cong, A. (2025) https://BioRender.com/tncql1j. b Chemical structure of compound 5c. c Protective effects of 5c against toxicity (LDH release) induced by DAU [1.2 µM] in NVCM 48 h after DAU addition, n = 4 independent replicates, mean ± SD. Statistical significance (P ≤ 0.05, one-way ANOVA) against untreated cells in column 1 is indicated as (c) or DAU treated cells in column 2 indicated as (d). d Inhibition of decatenation activity of recombinant human TOP2A and TOP2B by 5c (n = 3 independent replicates, mean ± SD normalized to untreated control). e Inhibition of ATPase activity of recombinant human TOP2A and TOP2B by compound 5c (n = 3 independent replicates, mean ± SD normalized to untreated control). Source data are provided as a Source Data file. Precise n and P values for each experiment are included in Supplementary Data Table 5.

Fig. 2. TOP2 inhibition through binding in a newly discovered pocket that differs between isoforms.

a Overall structure of the human TOP2B ATPase bound to AMPPNP (a nonhydrolyzable ATP analogue, representing the pre-hydrolysis conformation) and compound 5c. TOP2 ATPase subdomains are colour-coded and the second monomer coloured grey. b Molecular architecture of the obex pocket occupied by 5c in TOP2A and TOP2B. Electron density corresponding to 5c from a composite omit map (green mesh, contoured at 1σ) reveals the location and conformation of the compound (pink) within this pocket. c Sequence alignment of selected regions of the TOP2 ATPase domain. Pink circles indicate residues that interact with compound 5 c. d Overlay of the compound 5c inhibited ATPase structure (containing AMPPNP) with the ATPase structure in post ATP-hydrolysis state (ADP + SO₄, PDB ID: 7QFN). e Surface rendering showing the obex pocket of TOP2B ATPase is open in the AMPPNP (ATP)-bound form (left) but occluded in the ADP + SO₄ structure (PDB ID: 7QFN, right). Domains are coloured as in a.

Fig. 3. Topobexin (9) is a highly selective TOP2B inhibitor that binds to non-identical residues in the obex pocket.

a Chemical structure of topobexin (9) (additional substitution is shown in orange). b Inhibition of decatenation activity of recombinant human TOP2A and TOP2B by topobexin (9) (n = 3 independent replicates, mean ± SD normalized to untreated control). c, Molecular architecture of the obex pocket binding topobexin (9) in TOP2A and TOP2B. Electron density corresponding to topobexin (9) from a composite omit map (green mesh, contoured at 1σ) reveals the location and conformation of topobexin (9) (pink) within this pocket. d Overlay of topobexin (9) binding in the obex pocket of TOP2A and TOP2B. e Details of the amide dihedral angle of topobexin (9) when bound to TOP2A, TOP2B or in unbound (free) state. Source data are provided as a Source Data file.

Fig. 4. Topobexin (9) inhibits TOP2B with high selectivity and protects cardiomyocytes in vitro.

a Fluorescence recovery after photobleaching (FRAP) analysis in HEK293F cells expressing the indicated YFP-tag protein. Bleached area is shown in a purple box, scale bar = 5 μm. b RADAR assays show the levels of TOP2-DNA covalent complex modulated by each TOP2 isoform in HEK293F cells evaluated by Western blotting (n = 5 independent replicates, P ≤ 0.05 against untreated control cells (c) or DAU (d) (one-way ANOVA)). c Protective effects of topobexin (9) in isolated rat neonatal ventricular cardiomyocytes (NVCM) against toxicity (LDH release) induced by DAU [1.2 µM] 48 h after DAU addition, n = 4 independent replicates, mean ± SD. Statistical significance (P ≤ 0.05, one-way ANOVA) against untreated cells in column 1 is indicated as (c) or DAU treated cells in column 2 indicated as (d). d The levels of TOP2B protein in NVCM evaluated by Western blotting (n = 4 independent replicates, mean ± SD, P ≤ 0.05 against untreated cells (c) (one-way ANOVA)). e, The levels of phosphorylated γH2AX in NVCM evaluated by Western blotting (n = 4 independent replicates, P ≤ 0.05 against untreated control cells (c) or DAU (d) (one-way ANOVA)). f Activation of caspases after 3 h incubation with DAU and the effect of 3 h pre-treatment with topobexin (9) (n = 6 independent replicates, P ≤ 0.05 against untreated control cells (c) or DAU (d) (ratio paired t-test, two-tailed)). g Live-cell imaging of NVCM cells in the schedule corresponding to the cytotoxicity/protection experiments (48 h after DAU addition). Changes in morphology are shown in DIC mode while fluorescence signal from JC-1 probe corresponds to the state of mitochondria. Red signal represents actively respiring mitochondria while green indicates mitochondrial depolarization. Scale bar  = 100 μm. For all box and whisker plots, center line represents the median, “+” represents the mean. Bounds of box indicate 25th to 75th percentile, whiskers indicate minimal and maximal value. Source data are provided as a Source Data file. Precise n and P values for each experiment are included in Supplementary Data Table 5.

Fig. 5. Selective TOP2B inhibitor, topobexin (9), prevents anthracycline-induced cardiotoxicity in a rabbit model.

a Plasma pharmacokinetics of topobexin (9) after single i.v. dose administration to rabbits (10 mg/kg via 20 min infusion, n = 6 independent replicates) were determined using LC-MS/MS. b, Western blotting analysis of γH2AX in the left (LV) ventricular myocardium 1.5 h after a single administration of daunorubicin (DAU, 3 mg/kg). For the combination treatment, DAU was administered 40 min after the end of topobexin (9) infusion (n ≥ 6 independent replicates in each group). The quantitative comparisons involved multiple membranes, but all samples were processed and analyzed in parallel and the data from each membrane were normalized on the same internal standard prepared by pooling aliquots from all samples in the study. c–k Chronic cardiotoxicity experiment in rabbits: topobexin (9) and DAU were administered as in previous single dose experiments (A, B), repeated once weekly for 10 weeks and the indicated parameters were analyzed one week after the last dose. c Dosing regime in the chronic study. d Body weight gain. e Left ventricular fractional shortening (LV FS) examined by echocardiography – last measured values (LMV) in all animals. f Treatment-induced change in LV FS (Δ LV FS) during the experiment (the difference between last measured values and values obtained at the beginning of the experiment in individual animals). g LV dP/dt_max as an index of systolic function examined at the of the study via left ventricular catheterization. h Brain Natriuretic Peptide (BNP) mRNA levels in the LV myocardium were determined by quantitative RT-PCR. i Area Under Curve (AUC) of plasma concentrations of cardiac troponin T (cTnT), j, Fibronectin 1 and k, collagen I alpha-1 mRNA levels as determined by quantitative RT-PCR. *P < 0.05 compared to the indicated samples or # as compared with the topobexin (9)-alone group, determined using one-way ANOVA. For all box and whisker plots, center line represents the median, “+” represents the mean. Bounds of box indicate 25th to 75th percentile, whiskers indicate minimal and maximal value. Source data are provided as a Source Data file. Precise n and P values for each experiment are included in Supplementary Data Table 5.

Fig. 6. Model of topobexin-mediated cardioprotection.

a The TOP2 reaction cycle is inhibited by the catalytic inhibitors topobexin (9) (TOP2B) or dexrazoxane prior to ATP hydrolysis, thus preventing the cleavage complex from forming, and trapping TOP2 in a closed-clamp conformation on DNA. TOP2 poisons anthracyclines (daunorubicin) and etoposide block the religation step. Catalytic inhibitors block the TOP2 reaction cycle prior to cleavage, and thus fewer TOP2 cleavage complexes are available for poisoning. b Anthracyclines poison both TOP2A and TOP2B (left) leading to cancer killing and cardiotoxicity. TOP2B can be preferentially inhibited by topobexin (9) which binds to the obex binding site on TOP2B to prevent cardiotoxicity without impairing anthracycline effects on cancer cells. Created in BioRender. Cong, A. (2025) https://BioRender.com/atckl2z.