Tubulin binding site

This is a draft paper for discussion. It should not be quoted, cited or reproduced. MUT/06/14

COMMITTEE ON MUTAGENICITY OF CHEMICALS IN FOOD, CONSUMER
PRODUCTS AND THE ENVIRONMENT
CONCLUSIONS ON BENZIMIDAZOLE ANEUGENICITY: COMBINED
ASSESSMENT
Introduction

1.
At the May 2006 meeting Members commented that it may be possible to develop a scoring system to determine whether a chemical could be
included in a benzimidazole common mechanism group. Establishing a
common mechanism group is a critical first step towards enabling risk
assessments to be performed for exposures to multiple tubulin-active
compounds. We have considered this further and would appreciate Members’
advice on the below information and questions.
Demonstration of additivity

2.
Members considered that it was necessary to demonstrate that any chemicals to be included in a benzimidazole common mechanism group bind to mammalian β-tubulin at the same binding site, and that they act additively. Members also commented on the structure-activity relationship for benzimidazoles and aneugenicity. The current state of knowledge is summarised below: Tubulin binding site 3. There are data for some of the benzimidazoles to indicate that they bind to the colchicine binding site of mammalian tubulin. Mebendazole and oncodazole (a benzimidazole developed as a chemotherapeutic agent) have been shown to competitively inhibit the binding of cochicine to purified rat brain tubulin in competition binding assays using radiolabelled colchicine (Laclette et al., 1980; Ireland et al., 1979; Tahir et al., 2000; Russell and Lacey, 1995; Hoebeke et al., 1976). Oxibendazole and fenbendazole have been shown to competitively inhibit the binding of colchicine to bovine brain tubulin (Friedman and Platzer, 1978). Colchicine has also been shown to competitively inhibit mebendazole binding to rat brain tubulin, providing further evidence that these two compounds share a common binding site (Russell and Lacey, 1995). 4. In contrast, a recent study reported that while benomyl inhibited the polymerisation of purified goat brain tubulin, it did not inhibit colchicine binding to tubulin, as tested using an assay based on the principle that colchicine fluoresces when bound to tubulin (Gupta et al., 2004). Benomyl also did not inhibit binding of a fluorescent analogue of vinblastine to tubulin, indicating that benomyl binds at a novel site. The implications of this are unclear since This is a draft paper for discussion. It should not be quoted, cited or reproduced. benomyl is metabolised to carbendazim, which is generally considered to be the biologically active form (EC Scientific Committee on Plants, 2001). 5. Nguyen et al. (2005) employed docking studies for a range of different chemicals all able to bind within the colchicine binding domain of β-tubulin, including the benzimidazoles nocodazole and mebendazole. They constructed binding models for all these compounds and proposed a common pharmacophore model that linked these diverse chemicals. This included seven points: three hydrogen bond acceptors, one hydrogen bond donor, two hydrophobic centres and one planar group. None of the compounds were characterised by all seven points, which it was suggested may explain some of the differences in the activity of these compounds. 6. Robinson et al. (2004) proposed a site in helminth tubulin at which benzimidazoles may bind to β-tubulin based on clues from existing data, including correlation of benzimidazole resistance in the helminth Haemonchus contortus with a phenylalanine to tyrosine substitution at a particular position. The authors also modelled docking of albendazole oxide into H. contortus β-tubulin. It is not clear from this paper how the binding site might compare to that of other compounds. It is also not clear how binding to mammalian tubulin would compare. Role of the benzimidazole moeity 7. Presence of the benzimidazole ring appears to be important for the pesticidal/veterinary mode of action of the benzimidazoles. Since most of these compounds have been shown to inhibit the polymerisation of mammalian tubulin, the benzimidazole moiety appears to be important for this effect also. 8. Lacey and Watson (1985) studied the effect of the differences in the R1 substituent (see Figure 1) on the concentrations of benzimidazoles required to produce 50% inhibition of polymerisation of sheep brain tubulin (IC50). Carbendazim, which contains simply a hydrogen at this position, had weak activity, but replacing the hydrogen with one of the larger halides (Cl or Br) resulted in a progressive increase in activity. The presence of polar groups (OH, NH2 or NO2) resulted in loss of activity. An increase in alkoxy chain length from methyl to propyl resulted in increased activity but further increasing the chain length had little effect; a similar relationship was shown for alkyl chain length. The presence of branching in the substituent group at the α or β positions of R1 reduced potency. The molecular geometry and the polarity in the α or β regions of the substituent also appeared important to the tubulin binding potency of the molecule. 9. Jayasekhar and Kasture (1999) tested a number of benzimidazoles with an ethoxy group at the R1 position and various alkyl groups at the R2 position and showed that increased pKa was associated with increased This is a draft paper for discussion. It should not be quoted, cited or reproduced. Figure 1: Molecular structure of the benzimidazole ring No studies of the potential additive effects of benzimidazoles in relation to tubulin depolymerisation or aneugenicity have been identified. 11. A study has been identified in which two aneugens which bind to tubulin at different binding sites (paclitaxel and the synthetic vinca alkaloid
vinorelbine) were tested individually and in combination at various doses for
their ability to inhibit cell proliferation of human StM11 1a and G361
melanoma cells (Photiou et al., 1997). Dose-response data indicated
combined effects which the authors interpreted as being synergy at most,
predominantly low, concentrations, with additivity or sub-additivity at some
concentrations. This paper is attached as Annex A. In another study,
phenytoin and vinblastine were reported to have additive effects on porcine
brain tubulin polymerisation, despite binding to tubulin at different sites
(phenytoin binds at the colchicine binding site, vinblastine at the vinca alkaloid
binding site) (Lobert et al., 1999). The authors suggest that phenytoin
sequesters free tubulin heterodimers whereas vinblastine affects forming
microtubules. This paper is attached as Annex B.
Inclusion in a common mechanism group

12.
Figure 2 contains a proposed decision tree by which it could be decided whether to include a benzimidazole in a common mechanism group. This takes into account Members’ comments that non-mammalian data (i.e. aneugenicity in fungi, or helminth or fungal tubulin binding) were poor predictors of in vivo aneugenicity and that there were good correlations between mammalian tubulin binding and in vivo aneugenicity and aneugenicity in vitro in mammalian cells and in vivo aneugenicity. 13. It is suggested that other tubulin acting compounds could subsequently be added to the common mechanism group, providing they are: 1) aneugenic in vivo; and 2) additivity with at least one compound in the benzimidazoles CMG is This is a draft paper for discussion. It should not be quoted, cited or reproduced. Figure 2: Possible decision tree for including a substance in a benzimidazole common mechanism group Evaluate nature of combined effects on tubulin polymerisation of substances that bind at the same and different binding sites Evidence that the substance is aneugenic in Do not include in CMG. Any evidence for other vivo, e.g. positive micronucleus assay? combined effects, e.g. synergy, requires further Substance is negative in an in vivo study and there are pharmacokinetic data to indicate that the target tissue was sufficiently exposed? aneugenicity in mammalian cells in vitro? This is a draft paper for discussion. It
should not be quoted, cited or reproduced.
Questions

14.
The Committee is invited to consider the following questions and raise any other matters that arise from the data in this paper: i) Do Member’s agree with, or have any comments on, the suggested decision tree in Figure 2? Is there a need to consider not only aneugens which act by binding to tubulin at the same binding site but also those that bind to tubulin at different binding sites? What types of combined effect may be anticipated from mixtures of substances which bind to tubulin at different binding sites? Do all combinations of benzimidazole have to be tested to demonstrate additivity or could selected example compounds be tested? Could other aneugens be added to the common mechanism group if, for example they are aneugenic in vivo and additivity with at least one compound in the common mechanism group is demonstrated?
Secretariat
September 2006
References

EC Scientific Committee on Plants (2001). Opinion of the Scientific Committee
on Plants regarding the evaluation of Benomyl, Carbendazim and
Thiophanate-methyl in the context of council directive 91/414/EEC concerning
the placing of plant protection products on the market (opinion adopted by the
Scientific Committee on Plants on 7 March 2001). Available at
http://ec.europa.eu/comm/food/fs/sc/scp/out98_ppp_en.html.
Friedman PA, Platzer EG (1978). Interaction of enthelmintic benzimidazoles
and benzimidazole derivatives with bovine brain tubulin. Biochimica et
Biophysica Acta 544: 605-614.
Gupta K, Bishop J, Peck A, Brown J, Wilson L, Panda D (2004). Antimitotic
antifungal compound benomyl inhibits brain microtubule polymerization and
dynamics and cancer cell proliferation at mitosis, by binding to a novel site in
tubulin. Biochemistry 43: 6645-6655.
Hoebeke J, Van Nijen G, De Brabander M (1976). Interaction of oncodazole
(R 17934), a new anti-tumoral drug, with rat brain tubulin. Biochem. Biophys.
Res. Commun. 69: 319-324.
This is a draft paper for discussion. It
should not be quoted, cited or reproduced.
Ireland CM, Gull K, Gutteridge WE, Pogson CI (1979). The interaction of
benzimidazole carbamates with mammalian microtubule protein. Biochem.
Pharmacol. 28: 2680-2682.
Jayasekhar P, Kastrure AV (1999). Physico-chemical relationship of some
substituted benzimidazoles. Boll. Chim. Farm. 138: 489-492.

Lacey E, Watson TR (1985). Structure-activity relationships of benzimidazole
carbamates as inhibitors of mammalian tubulin, in vitro. Biochem. Pharmacol.
34: 1073-1077.
Laclette JP, Guerra G, Zetina C (1980). Inhibition of tubulin polymerization by
mebendazole. Biochem. Biophys. Res. Commun. 92: 417-423.
Lobert S, Ingram JW, Correia JJ (1999). Additivity of dilantin and vinblastine
inhibitory effects on microtubule assembly. Cancer Res. 59: 4816-4822.
Photiou A, Shah P, Leong LK, Moss J, Retsas S (1997). In vitro synergy of
Paclitaxel (Taxol) and Vinorelbine (Navelbine) against human melanoma cell
lines. Eur. J. Cancer 33: 463-470.
Robinson MW, McFerran N, Trudgett A, Hoey L, Fairweather I (2004). A
possible model of benzimidazole binding to β-tubulin disclosed by invoking an
inter-domain movement. J. Mol. Graph. Model. 23: 275-284.
Russell GJ, Gill JH, Lacey E (1992). Binding of [3H]benzimidazole carbamates
to mammalian brain tubulin and the mechanism of selective toxicity of the
benzimidazole anthelmintics. Biochem. Pharmacol. 43: 1095-1100.
Tahir SK, Kovar P, Rosenberg SH, Ng S-C (2000). Rapid colchicine
competition-binding scintillation proximity assay using biotin-labeled tubulin.
BioTechniques 29: 156-160.

Source: http://tna.europarchive.org/20060829080949/http://www.advisorybodies.doh.gov.uk/pdfs/mut0614.pdf