The interaction of spongistatin 1 with tubulin

Abstract

A tritiated derivative of the sponge-derived natural product spongistatin 1 was prepared, and its interactions with tubulin were examined. [³H]Spongistatin 1 was found to bind rapidly to tubulin at a single site (the low specific activity of the [³H]spongistatin 1, 0.75 Ci/mmol, prevented our defining an association rate), and the inability of spongistatin 1 to cause an aberrant assembly reaction was confirmed. Spongistatin 1 bound to tubulin very tightly, and we could detect no significant dissociation reaction from tubulin. The tubulin-[³H]spongistatin 1 complex did dissociate in 8 M urea, so there was no evidence for covalent bond formation. Apparent K_D values were obtained by Scatchard analysis of binding data and by Hummel-Dreyer chromatography (3.5 and 1.1 μM, respectively). The effects of a large cohort of vinca domain drugs on the binding of [³H]spongistatin 1 to tubulin were evaluated. Compounds that did not cause aberrant assembly reactions (halichondrin B, eribulin, maytansine, and rhizoxin) caused little inhibition of [³H]spongistatin 1 binding. Little inhibition also occurred with the peptides dolastatin 15, its active pentapeptide derivative, vitilevuamide, or diazonamide A, nor with the vinca alkaloid vinblastine. Strong inhibition was observed with dolastatin 10, hemiasterlin, and cryptophycin 1, all of which cause aberrant assembly reactions that might actually mask the spongistatin 1 binding site. Spongistatin 5 was found to be a competitive inhibitor of [³H]spongistatin 1 binding, with an apparent K_i of 2.2 μM. We propose that the strong picomolar cytotoxicity of spongistatin 1 probably derives from its extremely tight binding to tubulin.

Keywords: HPLC Hummel-Dreyer analysis; Halichondrin B; Inhibitors of microtubule assembly; Spongistatin 1; Spongistatin 5; Vinca domain.

Fig. 1.

Structures of spongistatins 1 and 5 and of halichondrin B.

Fig. 2.

Effect of spongistatin 1 concentration on the stoichiometry of binding to 5.0 μM tubulin. Reaction mixtures contained the indicated concentrations of [³H]spongistatin 1, 5.0 μM (0.5 mg/mL) tubulin, 0.1 M Mes (pH 6.9), 0.5 mM MgCl₂, and 4% (v/v) dimethyl sulfoxide. Reaction mixtures were incubated for 15 min at room temperature before two aliquots of each reaction mixture were applied to microcolumns.

Fig. 3.

A. Competitive inhibition of [³H]spongistatin 1 binding to tubulin by spongistatin 5, as demonstrated by Hanes analysis. Reaction mixtures contained the indicated concentrations of [³H]spongistatin 1, 5.0 μM (0.5 mg/mL) tubulin, 0.1 M Mes (pH 6.9), 0.5 mM MgCl₂, 4% (v/v) dimethyl sulfoxide, and either no spongistatin 5 (o) or spongistatin 5 at either 1.0 (Δ), 2.0 (∇), or 3.0 (□) μM. Reaction mixtures were incubated for 15 min at room temperature before two aliquots of each reaction mixture were applied to microcolumns. B. Dixon analysis of the data of Fig. 3A. Symbols: o, 1.0 μM spongistatin 1; Δ, 0.8 μM spongistatin 1; ∇, 0.6 μM spongistatin 1;□, 0.4 μM spongistatin 1. The intercept on the negative side of the abscissa domain indicated an apparent Ki of about 2.2 μM.

Fig. 4.

Demonstration of [³H]spongistatin 1 binding to tubulin by HPLC chromatography. The reaction mixture, containing 0.75 mg/mL (7.5 μM) tubulin and 10 μM [³H]spongistatin 1, was injected into a single HPLC column (plus guard column) in a 100 μl volume. The flow rate was 1.0 mL/min. There was a small peak of aggregated tubulin in the void volume, and this was present when no drug was added to the tubulin, but no oligomers or polymers of tubulin induced by spongistatin 1 were observed. Detection of radiolabel. The printout of the flow detector labeled the ordinate as “cps” with the scale being from 0.0 to 220.0. Detection of protein. The printout of the HPLC system labeled the ordinate as “mV” with the scale being from 0.0 to 116.0 and the baseline at 55.0.

Fig. 5.

Scatchard analysis of the binding of [³H]spongistatin 1 to tubulin. Analysis of data from multiple experiments in which reaction mixtures contained the indicated concentrations of [³H]spongistatin 1, 5.0 μM (0.5 mg/mL) tubulin, 0.1 M Mes (pH 6.9), 0.5 mM MgCl₂, and 4% (v/v) dimethyl sulfoxide, with the reaction mixtures incubated for 15 min at room temperature before two aliquots of each reaction mixture were applied to microcolumns. The data were analyzed in the Scatchard format, as indicated in the Figure.

Fig. 6.

Hummel-Dreyer analysis of the binding of [³H]spongistatin 1 to tubulin. After the column was equilibrated with spongistatin 1, 50 μg of tubulin was injected in 100 μL, and the column was developed at a flow rate of 1.0 μL/min. Fractions (1.0 mL) were collected and analyzed for protein (200 μL) and radiolabel (100 μL), and the data were analyzed to determine the apparent K_D for the binding of spongistatin 1 to tubulin, (0.67 μM in this experiment). A. Detection of radiolabel. The printout of the flow detector labeled the ordinate as “cps” with the scale being from 350.0 to just over 700.0. B. Detection of protein. The printout of the HPLC system labeled the ordinate as “mV” with the scale being from 0.0 to 30.0 and the baseline at 5.5.