Interaction of diazonamide A with tubulin

Abstract

[³H]Diazonamide A ([³H]DZA), prepared from the natural product isolated from Diazona angulata, bound to tubulin in larger aberrant assembly products (>500 kDa by sizing HPLC) but not to the αβ-tubulin heterodimer. The binding reaction was rapid, but stoichiometry was low. Stoichiometry was enhanced up to 8-fold by preincubating the tubulin in the reaction mixture prior to adding the [³H]DZA. Although Mg²⁺ did not affect binding stoichiometry, the cation markedly increased the number of tubulin rings (diameter about 50 nm) observed by negative stain electron microscopy. Bound [³H]DZA did not dissociate from the tubulin oligomers despite extensive column chromatography but did dissociate in the presence of 8 M urea. With preincubated tubulin, a superstoichiometric amount of [³H]DZA appeared to bind to the tubulin oligomeric structures, consistent with observations that neither nonradiolabeled DZA nor DZA analogues inhibited binding of [³H]DZA to the tubulin oligomers. Only weak inhibition of binding was observed with multiple antimitotic compounds. In particular, no inhibition occurred with vinblastine, and the best inhibitors of those examined were dolastatin 10 and cryptophycin 1. We compared the aberrant assembly reaction induced by DZA to those induced by other antimitotic peptides and depsipeptides, in particular dolastatin 10, cryptophycin 1, and hemiasterlin, but the results obtained varied considerably in terms of requirements for maximal reactions, polymer morphology, and inhibitory effects observed with antimitotic compounds.

Fig. 1.

Structures of DZA and a synthetic analogue DZ-2384. DZ-2384, co-crystallized with a tubulin complex in the vinca site, differs from the natural product, as indicated, in four ways. First, a tert-butyl group replaces the naturally occurring isopropyl group. Second, an ethoxy group replaces the naturally occurring chloro group. Third, a fluoro group has been added. Fourth, the bridging chloroindole group has been eliminated.

Fig. 2.

Pre-existing rings in the tubulin preparation used in studies of the binding of [³H]DZA to tubulin. The reaction mixture contained 4.0 μM tubulin, 4.0% (v/v) dimethyl sulfoxide, and 0.1 M Mes (pH 6.9) at 22 °C. The grid was prepared as soon as possible after diluting the stock tubulin solution (66.5 mg/mL) into the reaction mixture. Arrows indicate some of the rings observed on the grid, and magnification as indicated.

Fig. 3.

Effects of [³H]DZA concentration on the binding of [³H]DZA to tubulin. Each reaction mixture contained 0.1 M Mes (pH 6.9), 4.0 μM tubulin, 4.0% (v/v) dimethyl sulfoxide, and the indicated concentration of [³H]DZA. The reaction mixtures were incubated for 2 h at room temperature, at which point they were applied to syringe-columns. The symbol ● represents samples in which tubulin was the last component added, and its addition initiated the incubation. The symbol ○ represents samples in which the tubulin was added prior to the [³H]DZA. In these samples, the reaction mixtures were incubated for 90 min at room temperature, followed by addition of the [³H]DZA with a further incubation for 2 h at room temperature prior to the samples being applied to the syringe columns.

Fig. 4.

Scatchard analysis of different reaction sequences for the binding of either [³H]DZA or [³H]D10 to tubulin. The symbol ○ represents reaction mixtures containing 2.5 μM tubulin, 0.1 M Mes (pH 6.9), 4.0% (v/v) dimethyl sulfoxide, 0.5 mM MgCl₂, and varying concentrations of [³H]D10 (ranging from 0.2 to 4.0 μM). D10 was the last reaction component added. Incubation was for 15 min at room temperature before the samples were applied to syringe-columns. The symbol □ represents reaction mixtures containing the same components as in the D10 reaction sequence, except that they contained varying concentrations of [³H]DZA (ranging from 0.2 to 4.0 μM), which was the last reaction component added. Similar Bound/Free values were also obtained from reaction mixtures that did not contain MgCl₂ and/or in which the tubulin was preincubated prior to addition of the [³H]DZA. The symbol Δ represents the [³H]DZA containing reaction mixtures represented by the symbol ● in Fig. 3 (tubulin was the last reaction component added). The symbol ▽ represents the [³H]DZA containing reaction mixtures represented by the symbol ○ in Fig. 3 (tubulin was preincubated for 90 min before [³H]DZA, the last reaction component, was added).

Fig. 5.

Ring formation induced by DZA. Reaction mixtures contained 0.1 M Mes at either pH 6.9 or 6.5, as indicated, and 2.0 mM MgCl₂, if indicated, 4.0 μM tubulin, and 4.0% (v/v) dimethyl sulfoxide and were incubated for 90 min at 22 °C in volumes of 39.7 μL. DZA (4.0 μM) was added, bringing the reaction volume to 40 μL, and, after 30 min at 22 °C, the grids were prepared. All concentrations refer to the final reaction volumes of 40 μL. Magnification as indicated in panel D. Arrows in panels A and B indicate some of the scattered rings observed in the absence of Mg²⁺.

Fig. 6.

Structures of two DZA analogues that enhance, rather than inhibit, the binding of [³H]DZA to tubulin. Arrows indicate structural changes from DZA.

Fig. 7.

Size exclusion HPLC analysis of mixtures of tubulin and [³H]DZA. In all experiments the tubulin was preincubated for 30 min at room temperature, followed by the [³H]DZA for 15 min at room temperature. All reaction mixtures contained 0.1 M Mes (Ph 6.9) and 4.0% (v/v) dimethyl sulfoxide. A. The reaction mixture contained 2.0 μM tubulin and 0.5 μM [³H]DZA. The flow rate was 0.5 mL/min. B. The reaction mixture contained 4.0 μM each of tubulin and [³H]DZA. The flow rate was 0.5 mL/min. C. The reaction mixture contained 2.0 μM tubulin and 8.0 μM [³H]DZA. The flow rate was 1.0 mL/min.