WO2009074271A1

WO2009074271A1 - Peptides modulators of tubulin polymerisation

Info

Publication number: WO2009074271A1
Application number: PCT/EP2008/010347
Authority: WO
Inventors: Maurizio Sironi; Stefano Pieraccini; Giorgio Saladino; Paolo Mario Manitto; Giovanna Speranza; Pierangelo Francescato; Graziella Cappelletti
Original assignee: Universita' Degli Studi Di Milano
Priority date: 2007-12-12
Filing date: 2008-12-05
Publication date: 2009-06-18
Also published as: WO2009074271A8; ITMI20072323A1

Abstract

A peptide having one of the following sequences: - Ser-Pro-Lys-Val-Ser-Asp-Thr-Val-Val; - Ala-Trp-Xaa-Pro-Thr-Gly-Phe-Lys-Val-Gly (Xaa is Cys or Ala); - Phe-Arg-Arg-Lys-Ala-Phe-Leu-His-Trp-Tyr-Thr-Gly, analogues or derivatives therof.

Description

PEPTIDES MODULATORS OF TUBULIN POLYMERISATION

The invention refers to modulators of tubulin polymerisation. Background of the invention

Tubulin is an heterodimeric protein consisting of two similar monomers exhibiting a high homology degree, which are defined as the α and the β subunits. Tubulin (α, β) dimers self-assemble in a head-to-tail fashion to form protofilaments and microtubules. _N

In the polymerisation process an αβ unit binds to a subsequent αβ unit through a series of complex non-covalent interactions on an intercation surface of about 3000 A². Tubulin is presently one of the most interesting target for anti-mitotic drugs able to block the growth of tumor cells. Different compounds, mainly of natural origin, able to modify its polymerisation dynamic, are presently used. In recent years, widespread research efforts have been paid aiming at developing small molecules able to effectively bind to the protein regions acting as interfaces in the protein-protein interactions interfering either with protein aggregation or with the polymerisation. This goal is of primary interest for modulating all the processes where protein interactions play a fundamental role. Tubulin polymerisation causes the formation of microtubules involved in a number of cell functions including the cellular division processes. The growth and break down of microtubules is a process characterised by a quite complex kinetic and the interaction with said process may either stabilise or de-stabilise of microtubules with major effects on the cell cycle.

Microtubules are therefore the targets of many anticancer drugs that exert their cytotoxic action during the cell division process. A few molecules that interfere with microtubule dynamic instability either by stabilising or destabilising them, such as taxol, colchichine derivatives, vinca alkaloids and other antitumoral agents, are currently being used in therapy or are under clinical trials. These molecules were mainly identified by screening natural substances. Despite the great results obtained with these drugs, their success is limited by the onset of multidrug resistance in tumour cells during the treatment, so the development of new, resistance-proof, tubulin-targeted molecules is needed.

It is particularly meaningful that microtubules stabilisation induced by drugs such as paclitaxel and derivatives thereof, as well as their destabilisation, induced by colchicinoids and by vinca alkaloids, stop the mitosis process causing cell apoptosis (Jordan, M. A. & Wilson, L. Microtubules as a target for anticancer drugs. Nature Reviews Cancer. 4, 253-265 (2004)).

In view of the complexity of each heterodimer and of the fact that the interaction region between two subsequent units covers an area of about 3000 A², characterised therefore by a large number of complex interactions, the finding of suitable amino acid subsequences having an inhibitory effect is remarkably difficult and problematic, since the number of theoretically selectable peptides is huge. It should also be pointed out that in this case the target of the inhibitor is a surface between two proteins and there are many reports on the difficulty of developing small molecules which, "mimicking" a protein may block the interaction between the protein surfaces. This is mainly due to two reasons: a) the interprotein surface did not evolve so as to interact with small molecules; b) although some similarities between surfaces of different proteins have been recently observed, there is not yet sufficient information to establish binding analogies and therefore each interproteic surface is a separate case.

Summary of the invention

The present invention relates to peptides with tubulin-derived sequences which interfere with its polymerization process. The peptides of the invention have the following sequences: Ser-Pro-Lys-Val-Ser-Asp-Thr- VaI- VaI (Plug-R), Seq ID 1 Ala-Trp-Xaa-Pro-Thr-Gly-Phe-Lys-Val-Gly (Plug-H; Xaa is Cys or Ala) Seq ID 2 and 3; Phe-Arg-Arg-Lys-Ala-Phe-Leu-His-Trp-Tyr-Thr-Gly (Plug-X) Seq ID 4.

Said sequences are sub-sequences of tubulin β (Plug-R and X) and α

(Plug-H), with the sole exception of the plug-H Alal residue (which has replaced the Asp residue present in the original sequence). The three peptides are located at the interface between the two α and β units, as evidenced by the crystallographic structure deposited in the Protein Data Bank (structure ITUB. pdb). The invention also comprises the derivatives of the said peptides obtained by substitution of the natural amino acids with the corresponding amino acids of the D series and/or by derivatization of the hydroxide or thiol functional groups or of serine, threonine, tyrosine, cysteine, arginine, lysine basic functional groups and/or by functionalization of the N- and C- terminal groups, e.g. with acetate to amide groups. The invention also comprises the derivatives obtained by retro-inversion of one or more peptide bonds, according to known techniques which allow to stabilize the peptides against hydrolytic enzymes, thereby improving their pharmacokinetic characteristics. The present invention also includes the derivatives or the analogues of the said peptides obtained by conservative mutation of amino acids, typically one to three mutations, preferably one to two and most preferably one conservative mutation. A conservative mutation is for example the substitution of a basic residue with another basic residue, of an acidic residue acid with another acidic residue or of a neutral residue with another neutral residue. The present invention also includes peptides comprising two additional amino acids at the N or C terminal end, preferably selected from those present at the corresponding positions in the tubulin sequence. Said sequences are located at the interaction surface between two consecutive αβ units, i.e. in the space region interested by the polymerization process. The interaction takes place in a different domain other than those used by prior art antitumor medicaments targeting tubulin. Indeed, the peptides object of the invention do not interact with one of the already known sites for tubulin, conversely they act by locating at the interface between two tubulin dimers.

Furthermore, the interaction site of the peptides of the invention involves amino acids highly conserved in tubulin in that they are indispensable to its polymerization, therefore drug resistance due to mutations is unlikely.

As a consequence, the peptides of the invention may be used in the presence of other medicaments at reduced doses, thus decreasing the side effects.

The peptides of the invention can be prepared using conventional peptide synthesis techniques, and may be formulated in appropriate pharmaceutical compositions, preferably suited to the parenteral administration, due to the peptide nature of the active ingredients, although other administration routes, such as the oral route, cannot be excluded.

The compositions of the invention may be used for the treatment of tumors responding to antitubulin treatment. The peptides of the invention may be optionally combined with taxane derivatives (paclitaxel, docetaxel), vincamine, etoposide and derivatives or analogues thereof.

The peptides of the invention may also be useful as antiproliferative and anti-inflammatory agents and for the treatment of dermatological or neurodegenerative infections and diseases. The properties of the peptides of the invention were evidenced making use of both computational and experimental methods, as reported below.

Computational methods

Three simulations of "all-atoms" molecular dynamics have been carried out using the force field AMBER on the systems obtained by docking the peptide PLUG-R and the PLUG-X on the α unit and the peptide Plug-H on the tubulin β unit. The binding energies of the peptides have been calculated by the MM-GBSA approach using the structures obtained from the molecular dynamics. The role of each residue in the binding energy has been evaluated by carrying out a "computational alanine scanning". "Docking" studies have also been carried out.

Computational results for Plug-R

The binding energy for Plug-R amounts to 47% with respect to the binding offered by the complete interaction between two pairs of heterodimers.

The more significant interactions of peptide Plug-R with the residues of the α unit of tubulin are reported in Table Ia whereas the values obtained from alanine scanning are reported in Table 2a.

Table Ia: Decomposition of Binding Energy for Plug-R Alanine Scanning of Plug-R residue AAG_pepιideι (kcal/mol)

S£R_! 0.74

PRO₂ 0.33

LYS₃ -0.2

VAL₄ 3.73

SER₅ 2.67

ASP₆ 18.23

THR₁ 1.76

VALi -0.01

VAL₉ -0.66

Table 2a: Alanine Scanning for Plug-R

The results show that Plug-R mainly interacts with the residues of unit α. More particularly, Plug-R is able to establish an electrostatic interaction due to a saline bridge between Asp₆ and LyS₃₅₂ and a series of hydrophobic interactions involving VaI₄, Thr₇ e VaI₉ with Thr₃₄₉, Phe₃₅₁ and LyS₃₅₂. The geometrical analysis of the conformations obtained from the molecular dynamics also shows the formation of hydrogen bonds between the pairs Asp₆-Phe₃₅₁, Asp₆-Val₃₅₃ and Thr₇-Thr₃₄₉. During the simulation, said interactions remain stable, thus confirming that Plug-R can to bind to the α unit of tubulin, exactly on the surface which should be subjected to attack from other heterodimers, influencing thereby the polymerisation process.

The conformations obtained from the molecular dynamics have been subjected for further confirmation to a "docking" procedure which confirmed the binding of the peptide with an interaction energy of about 13.0 Kcal/mol. Computational results for Plug-H (Xaa=Cys)

The binding energy for Plug-R amounts to 30% with respect to the binding offered by the complete interaction between two pairs of heterodimers. The results for Plug-H are almost identical to that for plug-R, since they are located in a face to face position on the tubulin interproteic surface.

The most significant interactions of peptide Plug-H with the residues of the α unit of tubulin are reported in Table Ib whereas the values obtained from alanine scanning are reported in Table 2b.

"Pairwise" Decomposition for PIug-H

Res Res E_int (kcal/mol)

THR₅ VAL₁₄S₅ -1.28

THR₅ SER ₁₄g₆ -2.58

THR₅ ASP_U%1 -1.24

THR₅ THR 1488 -3.07

THR₅ VAL_US9 -1.00

PHE₁ VAL_U%5 -0.71

PHE₁ SER ₁₄₈₆ -2.95

PHE₁ ASP_l4Si -1.46

PHE₁ THRi₄₈₈ -0.31

LYS₉ ASP_14S1 -7.31

LYS₉ THRi₄₈₈ -0.71

VAL₁₀ THR 1484 -0.82

VAL₁₀ SuR₁₄₈₆ -0.61

Table Ib: Decomposition of Binding Energy for Plug-Η

Alanine Scanning of Plug-Η residue ΔΔG_peptide2 (kcal/mol)

TRP₂ 0.22

CYS₃ -0.10

PRO₄ 2.32

THR₅ 3.66

PHE₁ 0.67

LYSs 3.26

VAL₉ 5.83

Table 2b: Alanine Scanning for plug-Η

The results show that Plug-Η mainly interacts with the residues of unit β comprising the subsequence from which Plug-R was derived. More particularly, Plug-H is able to establish an electrostatic interaction due to a saline bridge between Asp₁₄₈₇ and Lysg and a series of hydrophobic interactions involving the same amino acid residues involved in the binding of Plug-R.

During the simulation, said interactions remain stable, thus confirming that plug-H can bind to the β unit of tubulin, exactly on the surface which should be subjected to attack from other heterodimers, influencing thereby the polymerisation process. The conformations obtained from the molecular dynamics have been subjected for further confirmation to a "docking" procedure which confirmed the binding of the peptide with an interaction energy of about 16.0 Kcal/mol.

Computational results for Plug-X

The binding energy for Plug-X amounts to 40% with respect to the binding offered by the complete interaction between two pairs of heterodimers.

The most significant interactions of Plug-X with the residues of the tubulin α unit are reported in Table Ic whereas the results of alanine scanning are reported in Table 2c.

"Pairwise" Decomposition for Plug-X

Table Ic: Decomposition of the Binding Energy for Plug-X

Alanine Scanning del Plug-X residue AAG_peptide2 (kcal/mol)

PHE₁ 2.45

ARG₂ -0.01

ARG₃ 4.49

LYS₄ 5.87

PHE₆ 8.47

LEU₇ 0.06

HIS₈ 2.18

TRP₉ 9.84

TYR₁₀ 0.39

THR₁₁ -0.06

Table 2c: Alanine Scanning for Plug-X

The results show that Plug-X interacts with different residues of the α subunit, and it establishes an electrostatic interaction due to the contact Lys₄ and Asp₄₃₈, hydrogen bonds between Trp₉ and VaI_26O ^and between Arg₃ and Ser₄₃₉. Hydrophobic interactions occur also between Phe₆ and VaI₂₆₀, His₈ and Tyr₂₆₂, Trp₉, VaI₂₆₀ and Tyr₂₆₂.

During simulation, said interactions remain stable, thus confirming that plug-X con bind to tubulin α unit, exactly on the surface which should be subjected to attack from other heterodimers, influencing thereby the polymerisation process.

The conformations obtained from the molecular dynamics have been subjected for further confirmation to a "docking" procedure which confirmed the binding of the peptide with an interaction energy of about 16.0 Kcal/mol. Description of the drawings

Fig. 1. The influence of Plug-X and Plug-H on tubulin assembly in vitro, a, Tubulin assembly was recorded as a function of time by measuring the increase in absorbance at 350 nm. Tubulin (20 μM) was polymerised in assembly buffer (filled circles) and in the presence of 50 μM Plug-X (open circles), 50 μM Plug-H (filled triangles), or 50 μM inactive peptide control

(open squares), b, The variation of the absorbance maximum as a function of time was recorded with various initial tubulin concentrations in the absence

(filled circles) or presence of 50 μM Plug-X (open circles) or 50 μM Plug-H

(filled triangles). The tubulin critical concentration was calculated as the x-intercept of the fitted lines.

In Figure 2 tubulin assembly was recorded as a function of time by measuring the increase in absorbance at 350 nm. Control (black diamonds) and Plug-R (open triangles) curves are reported.

In Figure 3 cell proliferation assay was performed on A549 cells incubated for 48 h in the presence of Plug-R, Plug-X and Plug-H.

Experimental Methods

Synthesis and characterization

The three capped peptides Plug-R (Ac-Ser-Pro-Lys-Val-Ser-Asp-Thr- VaI-VaI-NH₂), Plug-H (Ac-Ala-Trp-Ala-Pro-Thr-Gly-Phe-Lys-Val-Gly-NH₂), and Plug-X (Ac-Phe-Arg-Arg-Lys-Ala-Phe-Leu-His-Trp-Tyr-Thr-Gly-NH₂) were prepared by standard fluorenyl-9-methoxy-carbonyl (Fmoc) solid-phase synthetic protocols (Applied Biosy stems mod 433 A synthesizer) on a Rink Amide resin support (Novabiochem). Following standard Fmoc-amine deprotection, the acetyl group was introduced onto the N-terminal amino acid using acetic anhydride and HOBt/DIPEA as the coupling agent. The capped peptide was side chain-deprotected and cleaved from the resin with trifluoroacetic acid (TFA)-H₂O-thioanisole (90:5:5) and purified by reverse phase HPLC (C₁₈ column, gradient elution with acetonitrile/H₂O containing 0.1% TFA). The identity and purity (>95%) of each sample were assessed by electrospray ionization mass spectrometry (ESI-MS, ThermoFinnigan LCQ Advantage instrument) and analytical HPLC.

The biological activity of the designed peptides was assessed in vitro using purified tubulin and cultured cells. We initially investigated the ability of peptides to interfere with microtubule assembly in vitro and found that Plug-X and Plug-H inhibit tubulin polymerisation (Fig. 1). Quantitative parameters describing the mechanism of tubulin assembly can be deduced from the kinetics with theoretical models. We calculated the apparent first-order rate constant of elongation, the steady-state extent of assembly, and the critical concentration of tubulin. From assembly kinetics recorded with 20 μM tubulin (Fig. Ia), values (± s.e.m.) of 0.039 ± 0.008 ΔA min^'1 and 0.043 ± 0.009 ΔA min^"1 were derived for the elongation rate in the presence of Plug-X and Plug-H, respectively, which were significantly decreased with respect to the values of 0.073 ± 0.008 ΔA min^"1 that were derived from control kinetics. Similarly, the presence of Plug-X and Plug-H significantly affected the absorbance maximum at the end of assembly, which is proportionally related to the mass concentration of the tubulin polymer. The presence Plug-R (Fig.2) on the other hand resulted in an increased initial polymerization speed (0.1303 ± 0.02 ΔA min^'1). Plug-H and Plug-X were further tested to evaluate their effect on tubulin critical polymerisation concentration.. The variation of the absorbance maximum as a function of time was recorded with various tubulin concentrations (Fig. Ib). A linear relationship between the two values allows for the determination of the x-intercept, which is the minimum concentration of tubulin necessary to obtain polymerisation. We determined the critical concentration of tubulin is 9.52 μM with tubulin polymerised in the absence of plugs, but is 14.54 μM and 13.67 μM with tubulin polymerised in the presence of Plug-X and Plug-H, respectively. Therefore, Plug-X and Plug- H competitively affect tubulin assembly in vitro by decreasing the elongation rate and increasing the tubulin critical concentration. The interference of Plug-X and Plug-H with tubulin polymerisation in vitro is consistent with the prediction from the in silico design. However, it was also crucial to confirm these anti-microtubule effects in cells. To move to the lung adenocarcinoma A549 cell line, a proliferation assay was initially performed, and the dose-dependent cytotoxic effect of the plugs is described in Fig. 3. Values of about 184 μM and 197 μM were derived for the IC₅₀ in the presence of Plug X and Plug H, respectively. The IC_5O value derived for Plug-R was about 20 μM. In order to correlate the anti-proliferative effects evoked by the plugs with microtubule damage, immunofluorescence and confocal microscopy analyses on A549 cells were performed, and the microtubule network was investigated. In control cells, we observed a widespread network of long microtubules and the typical accumulation of microtubules at one side of the nucleus in the region called the microtubule organizing centre (MTOC). This conventional microtubule distribution underwent dramatic rearrangements in the presence of Plug-X, Plug-H and Plug-R, showing the evident disorganization of the network, lack of microtubule accumulation in the MTOC region, and fragmentation of microtubules. Tubulin purification and assembly assay

Tubulin was purified from bovine brain purchased from a local slaughterhouse, conserved before use in ice-cold PIPES buffer (IM K-PIPES, pH 6.9, 2 mM EGTA, and 1 mM MgCl₂) and used as soon as possible. Pure tubulin was obtained by two cycles of polymerisation-depolymerisation in a high-molarity buffer (Castoldi and Popov, 2003), and protein concentration was determined by the MicroBCA assay kit (Pierce). Stock solutions of plugs were prepared by dissolving the powders at a concentration of 4 mM in assembly buffer (80 mM K-PIPES pH 6.9, 2 mM EGTA, 1 mM MgCl₂, and 10% glycerol). The kinetics of tubulin polymerisation was followed turbidimetrically at 350 nm in an Ultraspec 300 spectrophotometer (Pharmacia) equipped with a temperature controller. After the addition of GTP (final concentration 1 mM), the reaction mixture containing tubulin (amount varied from 10 to 28 μM) and plugs in assembly buffer was transferred to cuvettes in the spectrophotometer, and the polymerisation reaction was followed at 37°C for 30 min. The steady-state extent of assembly was also examined as a function of the initial tubulin concentration and a linear fitting method was used to fit the data (Johnson and Borisy, 1977). The x-intercept of the fitted lines was designated to be the critical concentration. Cell culture and immunofluorescence

Human lung carcinoma cell line A549 (CCL- 185; American Type Culture Collection, Rockville, MD, USA) was grown in minimal essential medium with Earle's (E-MEM), supplemented with 10% fetal bovine serum (Hyclone Europe, Oud-Beijerland, Holland), 2 mM 1-glutamine, 100 U/ml penicillin, and non-essential amino acids. Cells were maintained at 37°C in a humidified atmosphere at 5% CO₂. For the cell proliferation assay, A549 cells (45000 cell/well) were seeded onto 24-well tissue culture plates, incubated for 48 h in the absence or presence of increasing concentrations of the plugs, harvested and counted. All experimental data were fitted and analysed by computer using a sigmoidal dose-response function (Sigma Plot, Jandel, CA). For microtubule organization analysis, cells were plated on glass coverslips at a density of 1.5^χ l O⁴ cells/cm² and grown for 24 h in control medium in the absence or presence of 50 μM Plug-X or Plug-H. At the end of the treatments, cells were fixed and permeabilized for 10 min with methanol at -20⁰C, washed with PBS, and blocked in PBS + 1% bovine serum albumin (BSA) for 15 min at room temperature. To localize tubulin, the cells were incubated with monoclonal anti-α-tubulin antibody (clone B-5-1-2, Sigma-Aldrich) 1 :500 in PBS for 1 h at 37°C. We used goat anti-mouse Alexa Fluor™ 594 (Molecular Probes) 1 : 1000 in PBS + 5% BSA for 45 min at 37°C as secondary antibodies. Nuclei staining was performed by incubation with DAPI (0.25 μg/ml in PBS) for 15 min at room temperature The coverslips were mounted in Mowiol^® (Calbiochem)-DABCO (Sigma-Aldrich) and examined with a confocal laser scan microscope imaging system (TCS SP2 AOBS, Leica Microsystems, Heidelberg, Germany) equipped with Ar, He-Ne and UV lasers.

Claims

1. A peptide having one of the following sequences:

- Ser-Pro-Lys-Val-Ser-Asp-Thr- VaI- VaI; - Ala-Trp-Xaa-Pro-Thr-Gly-Phe-Lys-Val-Gly (Xaa is Cys or Ala);

- Phe-Arg-Arg-Lys-Ala-Phe-Leu-His-Trp-Tyr-Thr-Gly, analogues or derivatives therof.

2. The peptide of claim 1 which is Ser-Pro-Lys-Val-Ser-Asp-Thr-Val-Val.

3. The peptide of claim 1 Ala-Trp-Xaa-Pro-Thr-Gly-Phe-Lys-Val-Gly (Xaa is Cys o Ala).

4. The peptide of claim 1 which is Phe-Arg-Arg-Lys-Ala-Phe-Leu-His- Trp-Tyr-Thr-Gly.

5. Pharmaceutical compositions comprising as active ingredient a peptide of claims 1-4.

6. Use of the peptides of claims 1-4 for the preparation of antitumor medicaments.

7. Use of the peptides of claims 1-4 for the per Ia preparation of antiproliferative, antinflammatory medicament and for the treatment of infections and for the treatment of dermatological and neurodegenerative diseases.