Doing the methylene shuffle - Further insights into the inhibition of mitotic kinesin Eg5 with S-trityl l-cysteine

Article abstract of DOI:10.1016/j.ejmech.2012.05.034

S-Trityl l-cysteine (STLC) is an inhibitor of the mitotic kinesin Eg5 with potential as an antimitotic chemotherapeutic agent. We previously reported the crystal structure of the ligand-protein complex, and now for the first time, have quantified the interactions using a molecular dynamics based approach. Based on these data, we have explored the SAR of the trityl head group using the methylene shuffle strategy to expand the occupation of one of the hydrophobic pockets. The most potent compounds exhibit strong (<100 nM) inhibition of Eg5 in the basal ATPase assay and inhibit growth in a variety of tumour-derived cell lines.

Full text of DOI:10.1016/j.ejmech.2012.05.034

Supporting information for:
Doing the methylene shuffle – further insights into the inhibition of mitotic kinesin
Eg5 with S-trityl L-cysteine
Murad N. Abualhasana*, James A. D. Goodb*, Kitiyaporn Wittayanarakula, Nahoum G. Anthonya,
Giacomo Berrettaa, Oliver Rathb, Frank Kozielskib Oliver B. Sutcliffea, Simon P. Mackaya†.
,
aStrathclyde Institute of Pharmacy and Biomedical Sciences, 161 Cathedral Street, University of
Strathclyde, Glasgow, G4 0RE, Scotland, UK.
and
bThe Beatson Institute for Cancer Research, Garscube Estate, Switchback Road, Glasgow G61 1BD,
Scotland, UK.
* = These authors contributed equally to the work.
†= Corresponding author.
Tel.: +44 1415482866; fax: +44 1415526443
E-mail address: s.mackay@strath.ac.uk (S. P. Mackay)
S1

Contents:
Molecular Modelling
13C NMR of 8c
S2
S5
13C NMR of 8d
S8
13C NMR of 8f
S11
S14
Supporting Information References
S2

Classical Molecular Dynamics methodology
Two minimisation cycles were applied as follows: first, the solute was restrained, while water molecules
and ions were minimised over 500 steps each of steepest descents (SD) and conjugate gradients (CG).
Solute restraints were then removed and 1000 steps of SD and 1500 steps of CG applied. MD
simulations initially involved the NVT ensemble, with the system being heated from 0 to 300 K over 20
ps, during which the solutes were constrained with a force constant of 10.0 Kcal/mol/Å2. Solute
restraints were then removed and NVT was replaced by the NPT ensemble at a constant temperature and
pressure of 300 K and 1 atm for the production phase [1], with bonds involving hydrogen atoms
constrained using the SHAKE algorithm [2] and a 2 fs time step employed throughout all simulations
(15 ns for all systems). Periodic boundary conditions with a 12 Å cutoff for non-bonded interactions
were used, with the particle mesh Ewald (PME) method [3] applied to account for the long-range
electrostatic interactions.
Free energy calculations
The total binding free energies (Gbinding) of the complexes were calculated using the MM/PBSA
method based on Equation 1.
G


G


(Equation 1)
G
Gbinding
complex
receptor
ligand

The composite terms of the binding free energies for the complex, receptor (in this case Eg5) and the
ligand(s) were calculated from Equation 2, which contains the enthalpic (H), entropic (S) and
solvation free energy (Gsolvation) contributions. H is the summation of the standard bonded and non-
bonded terms obtained from the force field equation and represents the molecular mechanics energy. S
is determined from translational, rotational and vibrational contributions based on classical statistical
thermodynamics [4, 5]. The solvation free energies (Gsolvation) consist of polar (Gpolar) and non-polar
(Gnon-polar) contributions (Equation 3), with the former being calculated by Poisson-Boltzmann (PB)
calculations [6] and the latter by a function of the solvent-accessible surface area (SASA) [7]. To
calculate Gpolar, grid spacing was set at 0.5 Å with evaluation of all pairwise interactions using an
internal and external dielectric constant of 1.0 and 80, respectively. The SASA term was based on
Equation 4, where  and b are 0.0054 Kcal/mol.Å2 and 0.92 Kcal/mol, respectively. The binding free
S3

energies of Eg5 in a complex with STLC and/or ADP were averaged from 200 snapshots collected from
the last 5 ns of the MD simulations.
(Equation 2)
Gbinding  H TS  Gsolvation
Gsolvation  Gpolar  Gnonpolar
Gnonpolar    SASA b
(Equation 3)
(Equation 4)
Solute entropic contributions were estimated from the sampled structures based on mass-weighted
covariance matrix analysis (quasiharmonic analysis) using the ptraj modules in AMBER [8]. This
method can calculate the entropy, heat capacity, and internal energy from the structure of a molecule.
The masses and the moments of inertia of the complexes are calculated for the translational and the
rotational degrees of freedom [1]. The vibrational contribution to the entropy of the complexes is
calculated using quasiharmonic analysis from the mass-weight covariance matrix. Due to demanding
computational times, the first, 50th, and 100th, 150th, and 200th configurations of the last 5 ns of the
trajectory were selected.
The MM-PBSA approach calculates free energies (G) based on Equation S1. We used 200 snaphots of
the solute sampled regularly from the last ns of the MD trajectories, with the water and counterions
stripped away. This method combines the enthalpic or molecular mechanics energies (EMM) that
represent the internal energies (bond, angle and dihedral; EBADH) along with van der Waals (EvdW) and
electrostatic interactions (Eelec), with the solvation free energies (Gsol) calculated by the finite difference
Poisson-Boltzmann (PB) model for polar solvation (GPB or Gpolar) [9] and the non-polar contribution
(Gnon-polar) as a function of the solvent-accessible surface area (SASA). All terms were computed from
the MM-PBSA module in AMBER. Solute entropic contributions were estimated from the sampled
structures based on mass-weighted covariance matrix analysis (Quasiharmonic analysis) using the ptraj
modules in AMBER [8].

 T 
S
G  EMM Gsol
(Eq. S1)
S4

Figure S1 – 13C NMR spectrum for 8c: full spectrum.
S5

Figure S2 – 13C NMR spectrum for 8c: expansion 1.
S6

Figure S3 – 13C NMR spectrum for 8c: expansion 2.
S7

Figure S4 – 13C NMR spectrum for 8d: full spectrum.
S8

Figure S5 – 13C NMR spectrum for 8d: expansion 1.
S9

Figure S6 – 13C NMR spectrum for 8c: expansion 2.
S10

Figure S7 – 13C NMR spectrum for 8f: full spectrum.
S11

Figure S8 – 13C NMR spectrum for 8f: expansion 1.
S12

Figure S9 – 13C NMR spectrum for 8f: expansion 2.
S13

Supporting Information References
[1] K. Wittayanarakul, S. Hannongbua, M. Feig, Accurate prediction of protonation state as a
prerequisite for reliable MM-PB(GB)SA binding free energy calculations of HIV-1 protease inhibitors,
Journal of Computational Chemistry, 29 (2008) 673-685.
[2] J.-P. Ryckaert, G. Ciccotti, H.J.C. Berendsen, Numerical integration of the cartesian equations of
motion of a system with constraints: molecular dynamics of n-alkanes, Journal of Computational
Physics, 23 (1977) 327-341.
[3] T. Darden, D. York, L. Pedersen, Particle mesh Ewald: An N [center-dot] log(N) method for Ewald
sums in large systems, The Journal of Chemical Physics, 98 (1993) 10089-10092.
[4] D.A. McQuarrie, Statistic Mechanics, University Science Books, New York, (1976).
[5] B. Tidor, M. Karplus, The Contribution of Vibrational Entropy to Molecular Association : The
Dimerization of Insulin, Journal of Molecular Biology, 238 (1994) 405-414.
[6] J.G. Kirkwood, F.H. Westheimer, The electrostatic influence of substituents on the dissociation
constants of organic acids I and II, J. Chem. Phys., 7 (1939) 437-437.
[7] E. Durham, B. Dorr, N. Woetzel, R. Staritzbichler, J. Meiler, Solvent accessible surface area
approximations for rapid and accurate protein structure prediction, Journal of Molecular Modeling, 15
(2009) 1093-1108.
[8] D.A. Case, T.A. Darden, T.E.C. III, C.L. Simmerling, J. Wang, R. E. Duke, R. Luo, K. M. Merz, D.
A. Pearlman, M. Crowley, R. C. Walker, W. Zhang, B. Wang, S. Hayik, A. Roitberg, G. Seabra, K. F.
Wong, F. Paesani, X. Wu, S. Brozell, V. Tsui, H. Gohlke, L. J. Yang, C. H. Tan, J. Mongan, V. Hornak,
G. Cui, P. Beroza, D. H. Mathews, C. Schafmeister, W. S. Ross, P.A. Kollman, AMBER 9, University
of California, San Francisco, (2006).
[9] K. Chin, K.A. Sharp, B. Honig, A.M. Pyle, Calculating the electrostatic properties of RNA provides
new insights into molecular interactions and function, Nature Structural Biology, 6 (1999) 1055-1061.
S14