PM534, an Optimized Target-Protein Interaction Strategy through the Colchicine Site of Tubulin

Abstract

Targeting microtubules is the most effective wide-spectrum pharmacological strategy in antitumoral chemotherapy, and current research focuses on reducing main drawbacks: neurotoxicity and resistance. PM534 is a novel synthetic compound derived from the Structure-Activity-Relationship study on the natural molecule PM742, isolated from the sponge of the order Lithistida, family Theonellidae, genus Discodermia (du Bocage 1869). PM534 targets the entire colchicine binding domain of tubulin, covering four of the five centers of the pharmacophore model. Its nanomolar affinity and high retention time modulate a strikingly high antitumor activity that efficiently overrides two resistance mechanisms in cells (detoxification pumps and tubulin βIII isotype overexpression). Furthermore, PM534 induces significant inhibition of tumor growth in mouse xenograft models of human non-small cell lung cancer. Our results present PM534, a highly effective new compound in the preclinical evaluation that is currently in its first human Phase I clinical trial.

Figure 1

Crystal structure of PM534 bound to tubulin. (A) Chemical structures of PM534 and natural compound PM742. (B) Time-lapse polymerization assay of 18 μM tubulin in the presence of increasing PM534 concentrations (0.625, 1.25, 2.55, 5 μM, salmon color gradient) compared with control (0.5% DMSO, black line). (C) Overall structure of the T₂R-TTL tubulin complex in the presence of PM534 (PDB 7ZYW), where proteins are in ribbon representation (α-tubulin in gray, β-tubulin in white, TTL in purple, and RB3 in green), and PM534 (salmon) is represented as surface bound to the colchicine domain of β-tubulin. (D) The sigma A weighted 2mFo-DFc (dark blue) electron density maps contoured at 1 sigma of PM534 (stick representation in salmon) at the colchicine domain of each of the protein chains, where it was identified. (E) Zoom into the colchicine domain in β-tubulin (ribbon representation chain D in white) showing colchicine (sticks purple), nocodazole (sticks blue), and PM534 (sticks salmon). The three zones of the pocket are labeled as dotted-line squares. (F) Zoom into the colchicine domain in β-tubulin (ribbon representation chain D in white), highlighting secondary structure elements involved in drug interaction and residues involved in hydrophobic and hydrogen bonding interactions (sticks representation).

Figure 2

PM534 binding, dissociation kinetics, and effect on microtubular network. (A) Displacement of the fluorescent probe R-PT (0.2 μM) bound to tubulin (0.2 μM) by PM534 at 25 °C. The solid line was generated with the best fit (EQUIGRA 5.0) value of the binding equilibrium constant of the competitor, assuming a one-to-one binding to the same site. Data are the mean value of three experiments ± SD. (B) Dissociation kinetics of PM534 from tubulin at 37 °C (salmon line, average of 3 curves). Purple line shows the association kinetics of 60 μM colchicine to 2.5 μM tubulin in the absence of PM534. (C) Effects of PM534 on A549 cells. Each cell triplet shows α-tubulin immunostaining (DM1A antibody), the DNA labeling (DAPI staining), and the merge of the previous ones (tubulin in green and DNA in magenta). Panels a, c, e, g, and i are interphase cells, while panels b, d, f, h, and j correspond to mitotic cells. Panels a and b (0.5% of drug vehicle, DMSO); c and d (100 nM colchicine); e and f (50 nM podophyllotoxin); g and h (1 nM PM534) and; i and j (2 nM PM534). f panels show two cells, the upper one with a less severe phenotype than the bottom one. Scale bar: 10 μm.

Figure 3

PM534 inhibition of brain tubulin vs βIII isotype. Time course assembly of 25 μM bovine brain tubulin (A) or recombinant human α1β3-tubulin (B) in the presence of DMSO (vehicle; black line), 27.5 μM paclitaxel (gray line), 27.5 μM podophyllotoxin (purple line), and increasing concentrations of PM534 (5 μM, 10 μM and, 27.5 μM, salmon color gradient).

Figure 4

Antitumor effect of PM534 on NCI-H460 xenografts. (A) Representative DNA staining from tumor xenografts cells with Hoechst 33258, 24 h post placebo (top) or post administration of 2.5 mg/kg of PM534 (bottom), highlighting the increase of apoptosis in PM534-treated NCI-H460 tumors (magnification 40X and scale bar: 50 μm). (B) Box-and-Whisker Plot (top) and dose–response curve (bottom) of tumor volume on Day 7 after single administration of PM534 (on Day 0) in mice bearing NCI-H460 xenografts. (C) Tumor growth curve for mice bearing NCI-H460 xenografts treated with PM534 (on Days 0, 7 and 14). (D) Kaplan–Meier survival curve in mice bearing NCI-H460 xenografts and treated with a placebo or PM534. Color code in these graphs is: placebo (gray) and doses of mice treatment 0.75, 1.1, 1.7, and 2.5 mg/kg (salmon color gradient). (E)Body weight change in the experimental groups of mice bearing NCI-H460 xenographs treated with placebo (gray) or 2.5 mg/kg PM534 (salmon). Lines represent mean weights ± SD (n = 6).