Antimicrotubule agents are important in the management of several cancers treatments notably breast, ovarian, and lung. Unfortunately, their effect is often hampered by chemoresistance, and they exhibit biopharmaceutical properties suitable for the treatment of only a limited number of cancers [
1]. New antimicrotubule agents are therefore highly desirable, and several laboratories are eager to develop new drugs exhibiting improved antitumor efficacy and better biopharmaceutical properties [
2]. One of the key end-points used in the screening of new molecules, beside the antiproliferative activity, is the binding affinity of the drug to one or another of the specific binding sites present on tubulin. Although binding sites for pironetin [
3], tryprostatin [
4], epothilone [
5],
Vinca [
6],
Taxus [
7], and
Colchicum [
8] alkaloids have been identified so far, the colchicine-binding site remains the main target of numerous research programs. The determination and the characterization of the binding site(s) for each and every newly prepared drug to the latter binding sites are costly, time-consuming, and require specialized equipment, highly purified tubulin, and expensive radiolabeled ligands and are therefore applied only to the most promising molecules.
To circumvent these limitations, we have developed an inexpensive and simple detection technique to assess the binding of antimicrotubule agents acting on the colchicine-binding site. This method is based on the property of
N,
N'-ethylene-bis(iodoacetamide) (EBI), a homobifunctional thioalkylating agent, to crosslink the Cys-239 and the Cys-354 residues of β-tubulin involved in the colchicine-binding site [
9]. The covalent binding of EBI to β-tubulin forms an adduct that is easily detected by Western blot as a second immunoreacting band of β-tubulin that migrates faster than the native β-tubulin band on SDS-PAGE (sodium dodecyl sulphate (SDS) polyacrylamide gel electrophoresis) [
10]. Luduena has used that property to characterize, in vitro, the activity of several tubulin inhibitors [
11,
12]. In the course of our research program on 1-aryl-3-(2-chloroethyl)ureas [
13], we found that the incubation of colchicine and iodoacetamide prevents the covalent binding of EBI in living cells resulting in the disappearance of the second immunoreacting band of β-tubulin on SDS-PAGE while vinblastine did not prevent the covalent binding of EBI. That observation provides us with a simple and inexpensive method to detect quickly and directly on living cells the ability of a molecule to bind to the colchicine-binding site without used of purified tubulin and the expensive [
3H]-colchicine. Furthermore, the study being performed on living cells, it provides information on the ability of the drug to cross cell membranes, to diffuse through the cytosol and to react with dynamically functional microtubules that are in contact with several cytoskeleton-associated proteins, information that are unavailable when using only purified tubulins in a in vitro setting.
In this study, we assessed the robustness of the method using: (1) drugs binding to the three main binding sites found on the α-,β-tubulin heterodimer, namely the taxol-, the colchicine-, and the vinca-binding sites, (2) drugs unrelated to tubulins and the cytoskeleton such as daunorubicin and verapamil, and (3) four cell lines found in most laboratories. As shown in Table 1, in Additional file
1, drugs binding to the colchicine-binding site inhibit the bisthioalkylation of β-tubulin by EBI and consequently the formation of the EBI: β-tubulin adduct and the second immunoreacting β-tubulin band. Conversely, vinca-binding site inhibitors and tubulin-unrelated drugs did not inhibit the covalent binding of EBI and the formation of the second immunoreacting β-tubulin band. Of note, the distinct colchicine-binding site and the paclitaxel-binding site are close (16–17 Å) and share a slight overlap between their interaction domains [
14]. This overlap between the colchicine-binding site and the paclitaxel-binding site seems to explain the effect of paclitaxel when challenged by EBI, as seen in Table 1,
1 and may show a screening advantage for dual inhibitors. In addition, as shown in Table 2, in Additional file
2, the assay can be performed not only with MDA-MB-231 but also with cell lines such as HT-29, M21, and MCF-7 that contain β-tubulin isoforms bearing the Cys-239 and the Cys-354 residues.
As illustrated in Figure , the inhibition of EBI binding to tubulin is concentration-dependant. When the experiments are performed using escalating concentrations of colchicine and combretastatin A-4 (10 to 31 250-fold the IC50), a semi-quantitative assessment of the affinity of the drug for the binding site is possible. Moreover, the purity of the sodium dodecyl sulfate (SDS) used to perform the experiments is crucial, since SDS at 65% significantly improved the separation between native β-tubulin and the EBI: β-tubulin adduct.
1,2 Jacques Lacroix,1 Marie-France Côté,1 Emmanuel Moreau,3 Éric Petitclerc,1 and René C-Gaudreault1,4
This article has been 
