[5]-Helistatins: Tubulin-Binding Helicenes with Antimitotic Activity

Abstract

Helicenes are high interest synthetic targets with unique conjugated helical structures that have found important technological applications. Despite this interest, helicenes have had limited impact in chemical biology. Herein, we disclose a first-in-class antimitotic helicene, helistatin 1 (HA-1), where the helicene scaffold acts as a structural mimic of colchicine, a known antimitotic drug. The synthesis proceeds via sequential Pd-catalyzed coupling reactions and a π-Lewis acid cycloisomerization mediated by PtCl₂. HA-1 was found to block microtubule polymerization in both cell-free and live cell assays. Not only does this demonstrate the feasibility of using helicenes as bioactive scaffolds against protein targets, but also suggests wider potential for the use of helicenes as isosteres of biaryls or cis-stilbenes-themselves common drug and natural product scaffolds. Overall, this study further supports future opportunities for helicenes for a range of chemical biological applications.

Figure 1

(A) Structures of colchicine and CA4. An SAR overview for CA4 is provided. (B) Syntheses of HA-1 and HA-2, both structural analogues of CA4, are reported in this work. HA-1 was designed as the active tubulin inhibitor and HA-2, where the ortho electron-donating −OH is located on the inner groove of the helicene, was designed to act as a negative control.

Figure 2

Molecular docking of (P)-HA-1 against (A) DAMA-colchicine and (B) CA4 in the tubulin-colchicine stathmin-like domain complex (PDB code: 1SAO). There is excellent overlap in both polar and nonpolar interactions, and the helical pitch provides the correct conformation for binding. See Movies S1–S4 to visualize the full three-dimensional fit.

Scheme 1. (A) Reaction conditions: (a) LDA, TMSCl, and THF, −78–0 °C, 24 h; (b) 2, ICl and CH₂Cl₂, −10–0 °C, 18 h; (c) 3, TIPS-Acetylene, CuI (10 mol %), Pd(dppf)Cl₂ (5 mol %), and PhMe/DIPA (2:1), 70 °C, 6 h; (d) 4, (3,4,5-trimethoxyphenyl)boronic acid, Pd(dppf)Cl₂ (12 mol %), K₂CO₃ and 1,2-Dimethoxyethane/H₂O (10:1), 85 °C, 18 h. (B) Reaction conditions: (a) 6, cyclopropyl methyl bromide, K₂CO₃, THF, 70 °C, 18 h; (b) 7, nBuLi, 2-isopropoxy-4,4,5,5-tetramethyl-1,3,2-dioxaborolane, −78–0 °C, 24 h; (c) KF (10 M aq.) MeCN/MeOH (1:1), 21 °C, 1 min, then l-(+)-tartaric acid in THF, 2–5 min, then filter. (C) Reaction conditions: (a) 4-methoxyphenylboronic acid or 9, Pd(dppf)Cl₂ (10 mol %), K₂CO₃ or Cs₂CO₃, and 1,2-dimethoxyethane/H₂O (10:1), 85 °C, 18 h; (b) TBAF, THF, r.t., 2 h; (c) PtCl₂ (40 mol %), PhMe, 80 °C, 18 h; (d) 12 M HCl, MeOH/iPrOH (1:2), 60 °C, 18 h.

Figure 3

Cellular bioactivity and mechanism of action of HA-1. (A) Antiproliferative activity (HeLa cells, 40 h incubation; one experiment representative of three independent experiments is shown). (B) Suppression of tubulin polymerization in cell-free conditions. Docetaxel is a positive control for polymerization enhancement, roughly indicating the magnitude of a meaningful difference from DMSO-only control behavior. HA-1 at 10 μM shows meaningful polymerization inhibition, while HA-2 up to 20 μM and colchicine or HA-1 at the substoichiometric concentration of 2 μM are all insignificantly inhibiting conditions (turbidimetric assay; greater absorbance indicates a greater degree of polymerization). (C, D) Live cell EB3-GFP “comets”, indicating growing MTs during HA-1/HA-2 treatments (HeLa cells). Cells were first imaged in media for baseline MT dynamics, then treated with 1% DMSO and imaged again, then 20 μM HA-1/HA-2 was applied, and the same cells were again imaged, which allows longitudinal comparisons over each treatment (10/12 cells acquired for HA-1/HA-2). (C) Each cell’s comet count was normalized to that in the first five frames of the no-cosolvent control before statistics (Wilcoxon) calculated for groups; mean ± SD with data points overlaid. (D) Stills from representative movies, Movies S5 and S6, showing that HA-1 suppresses polymerizing microtubule count, but HA-2 does not.