A novel norindenoisoquinoline structure reveals a common interfacial inhibitor paradigm for ternary trapping of the topoisomerase I-DNA covalent complex

Abstract

We show that five topoisomerase I inhibitors (two indenoisoquinolines, two camptothecins, and one indolocarbazole) each intercalate between the base pairs flanking the cleavage site generated during the topoisomerase I catalytic cycle and are further stabilized by a network of hydrogen bonds with topoisomerase I. The interfacial inhibition paradigm described for topoisomerase I inhibitors can be generalized to a variety of natural products that trap macromolecular complexes as they undergo catalytic conformational changes that create hotspots for drug binding. Stabilization of such conformational states results in uncompetitive inhibition and exemplifies the relevance of screening for ligands and drugs that stabilize ("trap") these macromolecular complexes.

Figure 1

Structure of the Top1 cleavage complex trapped by the norindenoisoquinoline AI-III-52. A–B: Top1 nicking-closing reaction. A: Top1 is generally bound non-covalently to DNA. The Top1 catalytic tyrosine (Y723 for human nuclear Top1) is represented in red (Y). A to B: Top1 cleaves one strand of the duplex as it forms a covalent phosphodiester bond between the catalytic tyrosine and the 3’- DNA terminus. The other DNA terminus is a 5’-hydroxyl (OH). B: The Top1 cleavage complex allows rotation of the 5’-terminus around the intact strand, which relaxes DNA supercoiling (purple dotted circle with arrowhead). B to A: Following DNA relaxation, Top1 religates the DNA. Under normal conditions, the religation (closing) reaction rate constant is much higher than the cleavage (nicking) rate constant. More than 90% of the Top1-DNA complexes are non-covalent (46) (as in A). C: Each of the Top1 inhibitors shown in panel F traps the Top1 cleavage complex by binding at the enzyme-DNA interface between the base pairs flanking the Top1-mediated DNA cleavage site (by convention positions −1 and +1). The colors for the base pairs −1 (light blue), +1 (dark blue) and +2 (orange) are the same as in Fig. 3. D & E: Lateral views of a Top1-DNA complex trapped by the norindenoisoquinoline AI-III-52 (shown in green and red). D: Top1 (yellow) is shown in a surface view to represent the depth of the norindenoisoquinoline binding pocket. E: Top1 is represented in ribbon diagram to allow visualization of the catalytic tyrosine (Y; red) and to show the drug intercalation between the −1 and +1 base pairs. E: Chemical structure of the five drugs co-crystallized with Top1-DNA complexes (15, 16, 24).

Figure 2

Stacking of a drug molecule between two base pairs flanking the Top1 cleavage complex is a common mechanism for the indenoisoquinolines (A, B), the camptothecins (C, D) and an indolocarbazole derivative (E). Each panel shows 2 views for each drug within the Top1 cleavage complex. The left views are oriented as in Fig. 1 D & E with the DNA viewed from the minor groove. The right views are rotated 90° and show the +1 nucleotides covering the drug molecules. In each panel, the catalytic Top1 tyrosine is shown in red (Y) at the top left of each left view. The −1 and +1 base pairs are marked by dashed arrows. In the right views, the colored numbers correspond to the drug atoms numbered in Figure 4. The orientation of SA315F is the same in the middle and right panels in Figure 2E.

Figure 3

DNA unwinding by the norindenoisoquinoline AI-III-52. A: Twist angle between the base pairs flanking the Top1 cleavage site in the absence of inhibitor (from (34)). The thin dashed lines correspond to the base pair long axes. The +1 base pair is colored dark blue and the −1 base pair cyan. B: Twist angle between the base pairs flanking the Top1 cleavage site trapped by AI-III-52. The drug has been removed to only show the −1 and +1 nucleotides. The +1 nucleotides are positioned similarly to panel A. The thin dashed lines correspond to the base pair long axes. C: Twist angle between the +1 and +2 base pairs. The color code is the same as in Fig. 1A–C.

Figure 4

Hydrogen bond networks between drugs and Top1 amino acid residues in the drug-Top1-DNA ternary complexes. A: AI-III-52 forms two direct hydrogen bonds with Asn722 and Arg364. B: MJ-238 forms only one hydrogen bond with Arg364 (16). C: The natural camptothecin forms 3 hydrogen bonds with Asp533, Asn722 and Arg364 (16). D: Two hydrogen bonds have been reported for topotecan with Asn722 and Arg364 (15). The third hydrogen bond with Arg364 has been proposed (see Discussion). Note that the water-mediated Asp722 hydrogen bond is represented with the amino group of N722. It is also plausible to be with the carbonyl of N722 (not shown).

Figure 5

Generalization of the interfacial inhibitor paradigm. A: Schematic cartoon representing the interfacial inhibition paradigm applied to an enzymatic system: 1) A substrate (S) binds to an enzyme (E); 2) The enzyme converts the substrate to product (S - > P); 3) P dissociates and the enzyme is regenerated for a new catalytic cycle; 4) The interfacial inhibitor (I) binds at the enzyme-substrate "interface" and traps the enzyme-substrate complex, thereby reversibly preventing conversion of S to P. B: Equilibrium representation of interfacial inhibition. K_S: substrate equilibrium dissociation constant; K_I: inhibitor equilibrium dissociation constant; k: catalytic constant. C: Equations corresponding to uncompetitive scheme shown in panel B. V: reaction velocity; K_M: Michaelis-Menton equilibrium dissociation constant (K_M = K_S when k is rate limiting).