Recognition Dynamics of p53 and MDM2: Implications for Peptide

Subscriber access provided by GAZI UNIV

Article

Recognition Dynamics of p53 and MDM2: Implications for Peptide Design Karim M. ElSawy, David P Lane, Chandra S. Verma, and Leo S. D. Caves J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.5b11162 • Publication Date (Web): 24 Dec 2015 Downloaded from http://pubs.acs.org on December 31, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry B is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Recognition Dynamics of p53 and MDM2: Implications for Peptide Design

Karim M ElSawy1,2,*, David P Lane3, Chandra S Verma4,5,6*and Leo SD Caves1,7,* 1

York Centre for Complex Systems Analysis (YCCSA); University of York; York, UK 2

Department of Chemistry; College of Science; Qassim University; Saudi Arabia 3

p53 Laboratory (A*STAR); Singapore

4

Bioinformatics Institute (A*STAR); Singapore

5

Department of Biological Sciences; National University of Singapore; Singapore 6

School of Biological Sciences; Nanyang Technological University; Singapore 7

Department of Biology; University of York; York, UK

*Correspondence to: Karim M ElSawy; Email: [email protected] ; Chandra S Verma: Email: [email protected] ; Leo SD Caves: Email: [email protected]

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 29

Abstract Design of peptides that inhibit MDM2 and attenuate MDM2-p53 interactions, thus activating p53, is currently being pursued as anticancer drug leads for tumours harboring wild type p53. The thermodynamic determinants of peptide-MDM2 interactions have been extensively studied. However, a detailed understanding of the dynamics that underlie these interactions is largely missing. In this study, we explore the kinetics of binding of a set of peptides using Brownian dynamics simulations. We systematically investigate the effect of peptide C-terminal substitutions (Ser, Ala, Asn, Pro) of a Q16ETFSDLWKLLP27 p53-based peptide and a M1PRFMDYWEGLN12 12/1 phage-derived peptide on their interaction dynamics with MDM2. The substitutions modulate peptide residence times around the MDM2 protein. In particular, the highest affinity peptide, Q16ETFSDLWKLLS27, has the longest residence time (t~25 µs) around MDM2, suggesting its potentially important contribution to binding affinity. The binding of the p53-based peptides appears to be thermodynamically driven while that of the phage-derived series appears to be kinetically driven. The phage-derived peptides were found to adopt distinctly different modes of interaction with the MDM2 protein compared to their p53-based counterparts. The p53-based peptides approach the N-terminal region of the MDM2 protein with the peptide C-terminal end oriented towards the protein, whilst the M1PRFMDYWEGLN12-based peptides adopt the reverse orientation. To probe the determinants of this switch in orientation, a designed mutant of the phagederived peptide, R3E (M1PEFMDYWEGLN12), was simulated and found to adopt the orientation adopted by the p53-based peptides and also to result in almost a 5-fold increase in the peptide residence time (~120 µs) relative to the p53-based peptides. On this basis, we suggest that the R3E mutant phage-derived peptide has a higher affinity for MDM2 than the p53-based peptides and would therefore, competitively inhibit MDM2-p53. The study, therefore, provides a novel computational framework for kinetics-based lead optimization for anti-cancer drug development strategies. 2 ACS Paragon Plus Environment

Page 3 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction p53 is a major regulator of cellular signalling and check-pointing and its deregulation has been implicated in most cancers 1-3. Activation of p53 in stressed cells, usually due to DNA damage, aberrant growth signals or interaction with chemotherapeutic agents or kinase inhibitors 2, disrupts multiplication of damaged cells, inhibits cell cycle progression and could lead to programmed cell death (apoptosis) 1. In general, the population p53 in cells is primarily determined by the rate of degradation 1 a process in which the MDM2 protein, a ubiquitin E3 ligase, is centrally involved 4. 2,5

In normal cells, p53 MDM2 interaction involves a negative feed-back loop

such that

transcription of the MDM2 protein is stimulated by p53; ubiquitination of MDM2, and its subsequent degradation, is triggered by p53 binding to the trans-activation N-terminal domain of the MDM2. Due to this feedback loop, the p53 concentration is maintained at low levels in normal cells. Upon cellular stress, post-translational modifications of p53 and MDM2 are triggered, inducing p53 dissociation from MDM2 and blocking p53 ubiquitination. This results in the accumulation of p53 and the activation of its role as a transcription factor 2. Thus, the development of molecular entities to inhibit p53-MDM2 association is a viable route for anticancer drug therapy 2,3,6,7

. For the past two decades, small molecules have been used almost exclusively in drug lead

development strategies 8, largely due to technical difficulties in using other biological molecular alternatives such as small peptides 9 and short nucleotide segments 10. Difficulties in using peptides in drug development have been largely alleviated by advances in peptide drug delivery protocols 1113

and emergence of new techniques that enhance peptide structural stability

14-16

. Peptides are

therapeutically interesting agents, as their properties offer the advantages of both small-molecule drugs and proteins. Peptides, therefore, offer higher specificity and lower toxicity compared to small-molecule drugs that are conventionally used in drug discovery

17,18

. However, search for

potential peptide lead structures poses a challenging task due to the vast dimensionality of the 3 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 29

peptides sequence and conformational space. One approach to reduce the dimensionality of this search space is to use peptide-based mimics

19-21

of protein-protein interfaces that competitively

prevent the binding of endogenous biological inhibitors/regulators. High-resolution NMR and crystal structures of MDM2 in complex with a variety of peptides and small molecules are currently available

22-27

, providing a plethora of information about their

interaction interfaces and providing clues of potential strategies for the development of inhibitors of the p53-MDM2 interaction. Importantly, the crystal structure of the MDM2 protein bound to a 12residue (E17TFSDLWKLLP27) peptide of the p53 trans-activation domain (wild-type p53 peptide) reveals that residues F19, W23 and L26 of the trans-activation domain of p53 are critical for p53 MDM2 association

27

. Mimicking these interactions has been key to the development of new

anticancer peptide-based lead structures

7,28

. However, affinity measurements have shown that the

C-terminal residue of this sequence, despite being located just outside the p53 binding pocket on MDM2 27, also plays a role in modulating the association of p53 with MDM2 29. For example, the P27S C-terminal substitution resulted in a 2.3 kcal mol-1 improvement (~ 200 fold increase in Ka) in peptide binding affinity to MDM2 relative to wild-type p53 peptide. This was attributed, in part, to an increase in the helical propensity of the peptide upon the P27S substitution, thereby reducing the entropic penalty for immobilizing a helical motif onto the binding site of MDM2

30

. Systematic

analysis of the role of single C-terminal substitutions (Ser, Thr, Ala, Asn and Pro) in a Q16ETFSDLWKLLP27 p53-based

peptide and an

equivalent

phage

optimized

peptide

M1PRFMDYWEGLN12 (12/1) revealed two distinct thermodynamic modes of interactions with the MDM2

31

. One mode correlates with peptides that have a lower helical propensity (the

M1PRFMDYWEGLN12 series) and is more enthalpically driven while the other mode correlates with peptides that are more helical in their apo states (the p53-based Q16ETFSDLWKLLP27 series) and is more entropically driven. Interestingly, the structure of the two series of peptides were found to be largely similar when complexed to MDM2 31. However, given that C-terminal substitutions in 4 ACS Paragon Plus Environment

Page 5 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


both peptide series lie just outside the p53 MDM2 binding pocket, this alludes to the possibility of more subtle effects that underlie the interactions of these peptides with the MDM2 protein. Traditionally, analyses of protein-peptide/ligand interactions have been limited to either probing the complex state and/or the state of the peptide/ligand or protein in their apo states. Any kinetic information has been gleaned from experimental techniques such as SPR which provide the rates of association and dissociation. However, details of these processes have only been available from a few Brownian dynamics simulations and more recently long timescale atomistic simulations coupled to markov chain analyses

32,33

. In a recent Brownian dynamics study, based on kinetic

arguments, we showed that residues on the MDM2 surface outside the p53 binding pocket could stabilize the p53-MDM2 encounter complex, a state that is formed prior to the formation of the final bound complex, by increasing p53 residence times and allowing for allosteric interactions with MDM2

34

. Kinetics of drug interactions are important and are now increasingly being explored to

investigate drug binding affinities with the realization that biology is governed by open-system conditions that are usually far from thermodynamic equilibrium

35-40

. Computational methods

supporting drug discovery have traditionally focused on estimating equilibrium thermodynamic parameters of binding. Docking is clearly a static methodology, and is typically used for estimating/approximating the free energy of binding - a thermodynamic measure

41-44

. Molecular

dynamics (MD) is an explicit dynamical method that could, in principle, be used to compute the kinetic parameters we estimate in this study

45,46

. However, MD is computationally far more

expensive than Brownian Dynamics (BD), and thus its use would become intractable to generate the statistical results that we utilise in this study. Thus we employ BD as the method underpinning our new computational approach to investigating the kinetics of receptor-ligand interactions, and suggest it as the basis for new tools for kinetics-based drug discovery. In this work, we investigate the kinetic factors that underlie the interactions of MDM2 with the two series of peptides, Q16ETFSDLWKLLX27 p53-based and M1PRFMDYWEGLX12 12/1 5 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

phage-derived series (X = Ser/Ala/Asn/Pro)

Page 6 of 29

29

. Using Brownian dynamics (BD) simulations, we

examine the configurations of the peptide-MDM2 encounter complexes for the two series and compute their life times viz. residence times. We show that the two series show distinct encounter complex configurations indicating distinct mechanisms of interactions with MDM2 well before formation of the final bound complex. We also show that the residence time trends of these encounter complexes mimic that of the observed affinities of the two series. A testable hypothesis is then developed by designing a new mutant peptide.


Page 7 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Methods Modelling

MDM2

diffusional

association

to

the

Q16ETFSDLWKLLP27

and

M1PRFMDYWEGLN12 peptide series BD simulations of the diffusional association of MDM2 to a two series of C-terminally substituted

peptides;

a

Q16ETFSDLWKLLX27

p53-based

peptide

series

and

a

M1PRFMDYWEGLX12 phage-derived 12/1 peptide series (X = Ser/ Ala/Asn/Pro), were carried out using the SDA package version 4.23b

47

. In each simulation, ten conformations of the MDM2

protein were extracted from molecular dynamics trajectories of the p53:MDM2 complex 48 and used as representative initial structures for the MDM2 protein. Representative structures for the Q16ETFSDLWKLLX27 and the M1PRFMDYWEGLX12 peptide series were prepared as follows: Since the crystal structure of these two series of peptides is not known, the crystal structure of the backbone of a p53 peptide (residues 17–29) bound to MDM2 (PDBID: 1YCR)27 was used as a template for the backbone structure of the peptides of the two series. Coordinates of the p53 Q16 residue, missing from the 1YCR structure, were built from standard bond lengths, angles and torsions as provided in the CHARMM22 parameters. Acetylated N-terminus and amidated Cterminus patches were applied to peptide ends. This was followed by 2000 steps of steepest descent minimization whilst the peptide backbone atoms were constrained by a 0.5 kcal mol-1 Ǻ-1 force constant. MD simulation of each peptide was then carried out using CHARMM force field 49. The MD trajectories were used for constructing a collective coordinate subspace

50

for describing the

peptide conformational variation using principal component analysis (PCA). The distribution of peptide conformations within this subspace was then partitioned into distinct regions (“basins of attraction”) using a contour following algorithm

51

. Representative structures for the

Q16ETFSDLWKLLX27 and the M1PRFMDYWEGLX12 series of peptides, used to initiate the BD simulations were selected from the point of (local) maximum probability density within each basin. BD simulation details 7 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 29

Brownian Dynamics (BD) simulations were performed using the SDA package (version 4.23b) 47 using the Ermak-McCammon algorithm 52 for generating the diffusional trajectories. The HYDROPRO software

53

was used for calculating the translational and rotational diffusion

coefficients of the peptide and the MDM2 protein. At the start of each BD trajectory, the MDM2 centre of mass (COM) was placed at the origin, while the peptide COM was placed at a distance b = 150.0 Å relative to the MDM2 COM. It is noted that, the electrostatic potential of the MDM2 is nearly isotropic at about 80 Å from its COM, therefore placing the peptide at 150.0 Å distance relative to the MDM2 COM results in negligible orientational bias to the peptide at the start of the BD trajectory. During the BD simulations, when the distance between the protein and the peptide was less than 90 Å, a time step of 0.1 ps was used. At larger distances, a varying time step was used which increases linearly with a slope of 0.5 ps Å-1. 50,000 BD trajectories were propagated for each MDM2 peptide simulation. BD trajectories were terminated if the peptide-protein COM-COM separation was greater than a distance c (c = 3b Å). In the BD simulations; steric, electrostatic and desolvation interaction forces between the peptide and the MDM2 protein were computed. Steric interactions were modelled through the use of exclusion grids which prevent overlap of peptide and protein: grids were centred on the respective COM, with a grid spacing of 1.0 Å. Electrostatic interaction forces were calculated by multiplying the electrostatic potential of the MDM2 protein at the positions of the peptide atoms by their respective point charges. The electrostatic potential of the peptide/protein was computed by solving the nonlinear Poisson-Boltzmann equation 54,55 using the APBS program

56

on a 161 × 161

× 161 Å grid with 1.0 Å spacing. A dielectric constant of 78.5 was used for the solvent while the protein/peptide dielectric constant was set to 4 and a salt concentration of 0.15 M was used. The PDB2PQR program

57,58

was used to specify the atomic charges and radii of the peptide and the

protein. In order to account for electrostatic desolvation interactions, desolvation penalty grids

59

were calculated around the peptide and the MDM2 protein using a scaling factor of 1.67 60. For the 8 ACS Paragon Plus Environment

Page 9 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


sake of computational efficiency, the atomic charges of the protein/peptide were replaced during the BD simulations by a set of effective charges 61 that accurately reproduce the calculated electrostatic potential of the protein/peptide in a 3 Å layer extending outwards from each structure using the ECM module in the SDA package. Analysis of the BD simulations The residence time radial profiles of peptide-MDM2 interaction were calculated in spherical concentric radial slabs of 1 Å thickness, centred at the MDM2 COM, by weighting the peptide radial probability density in each slab by the corresponding BD time step 47,62. In order to construct a 3D spatial probability density of the peptide, a grid of 201 × 201 × 201 Å with 1 Å was constructed around the MDM2 protein. The probability density at each individual spatial grid cell was calculated from the average frequency of occupancy of the peptide (though its COM position). The basins of attraction within the 3D probability density landscape were then constructed by the contour following algorithm described previously 40,51. In order to compute the ligand residence time (τ) for individual basins, the time-weighted ligand probability density was numerically integrated within its respective contour surfaces. Using a small contour interval such that the ligand density D is effectively constant between consecutive contours, the integral can be approximated by: N −1

τ = ∑ ∆t (V −V )( D + D ) / 2 n=2

n

( n −1 )

n

n +1

(V − V( n −1) ) is the where ∆t is the time step used in the Brownian dynamics simulation, n volume enclosed between the two consecutive contour surfaces at ligand densities

Dn

and

Dn +1

while N is the number of contour surfaces. The peptide MDM2 association rate constants, kon, were calculated by combining the peptide average probability (β) of reaching individual basins with the steady state rate constant



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 29

k(b) , k (b) = 4πDb , of the peptide being at a distance b of the centre of mass of the MDM2 protein

(b=150 Å) 63:

ka =

k (b) β 1 − (1 − β )k (b) / k (c) .

In the expression for ka, the denominator accounts for the termination of trajectories at a finite separation c (c=3b).


Page 11 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Results and Discussion In order to understand the kinetic determinants of the interactions of MDM2 with the Cterminal variants of a Q16ETFSDLWKLLX27 p53-based peptide and a M1PRFMDYWEGLX27 phage-derived 12/1 peptide (X=Ser/Ala/Asn/Pro), we studied their dynamical interactions with the MDM2 protein using BD simulations. The resulting interaction landscapes characterize molecular events that are diffusively bound where electrostatic interactions predominate 62. We partitioned the interaction landscape into regions of space where the rate of peptide association to MDM2 is larger than its rate of dissociation (Figures 1 & 2) using a methodology that we developed previously 40. These regions are termed ‘basins of attraction’. Each basin of attraction corresponds to an ensemble of molecular configurations of MDM2-peptide interactions termed the ‘encounter complex’, representing an essential step distinct from the formation of the non-diffusive final bound complex. Previously, we found that identification of the locations of the basins of attraction, characterization of the molecular configurations of the encounter complexes and computation of the corresponding residence times could provide detailed explanations for the mechanisms of interactions of MDM2 with p53, phosphorylated-p53 and the geometrical isomers of a small molecule inhibitor of the p53MDM2 interaction, Nutlin

34,64,65

. Here, we use this approach to explore the interaction dynamics

variations underlying the affinity of MDM2 for the C-terminal variants of the QETFSDLWKLLX p53-based peptide and the MPRFMDYWEGLX phage-derived 12/1 peptide (X=Ser/Ala/Asn/Pro), and relate these to the observed binding affinities.

Effect of QETFSDLWKLLP C-terminal substitution on peptide residence times around MDM2 Inspection of the radial residence time profiles of the p53-based peptide mutants, (Figure 1a) reveals that Asn, Ala and Ser C-terminal substitutions lead to a steady increase in their average residence times around the MDM2 protein. The increase in average residence times can be 11 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 29

attributed to contributions from the peptide basins of attractions around the MDM2 protein (Figure 1b). The Ser substitution results in the largest increase in the residence time (t~25 µs) of the dominant basin. This increase is in agreement with the experimentally observed trend of the affinity of these peptides for MDM2

31

. The increase in residence time is commensurate with

increase in the volume of the dominant basins of attraction (Figure 1 c) that gradually cover greater proportions of the MDM2 surface, especially towards the MDM2 N-terminal region. This is consistent with our previous finding that an ETFSDLWKLLP p53-based peptide interacts with MDM2 through prolonged interactions with the N-terminal region of the MDM2, away from the p53 binding site, allowing for allosteric changes in the MDM2 conformation 34,48.

Effect of MPRFMDYWEGLN C-terminal substitution on peptide residence time around the MDM2 In the phage-derived peptide series, the Ser, Ala and Pro C-terminal substitutions result in more complex changes in the peptide average residence times around the MDM2 protein (Figure 2 a). Consistent with the observed affinity for MDM2

31

, the Asn substituted peptide shows the

shortest average residence time while the Ser substituted peptide shows the longest average residence time. The radial residence time profiles of the Pro and Ala peptides, however, do not clearly follow the observed affinity trend. The computed dissociation constants (Kd = koff/kon ; koff ~ 1/τ ; τ = residence time) of the encounter complexes corresponding to the dominant basins do not follow the observed affinity trend either (Figure 3a). Of course, it must be noted that the observed affinities characterize the formation of the final non-diffusive bound complex whilst the current method enables the computation of kon and koff that characterizes the formation of the diffusivelybound encounter complex. This suggests that post-encounter binding events are important for the phage-derived series and could impact on the computed kon and koff. In contrast, the computed values of the peptide association rate constants (kon), follow the experimentally-observed trend in


Page 13 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


peptide affinity for MDM2 (Figure 3b) indicating that post-encounter binding events affect koff rather than kon which suggests that the formation of the final bound complex for this series is most likely thermodynamically driven. Inspection of the locations of the basins of attraction of the four peptides (Figure 2 c) reveals that they span a large part of the MDM2 protein with prominent extensions towards the N-terminal end of MDM2 indicating a mode of interaction similar to the p53-based peptide series. However, the shapes of the basins of attraction of the Pro and Ala mutants (Figure 2 b; 2nd and 3rd column) are markedly different. The dominant basin of attraction of the Pro mutant shows a broad arm that extends away from the N-terminal of the MDM2, which is not the case for the Ala mutant. This could explain, in part, the observed lower affinity of the Pro mutant relative to the Ala mutant through providing an exit channel that leads to depletion of the Pro peptide away from the Nterminal region of the MDM2.

Modes of interaction of the p53-based and the phage-derived peptide series with MDM2 The modes of interactions of MDM2 with the p53-based and the phage-derived peptide series correspond to the encounter complex ensembles (molecular configurations within the basins of attraction) (Figures 1d & 2d). In both peptide series, the encounter complex ensembles exhibit ordered orientations of the peptides. This indicates that peptide interactions with MDM2 incur an entropic cost that should be compensated by enthalpic interactions for favourable binding to take place. The configurations corresponding to the dominant basins of attractions of the p53-based series reveal that the peptides approach the N-terminal region of the MDM2 protein through their C-terminal ends (Figure 1d). This is consistent with our previous study of an ETFSDLWKLLP p53based peptide that interacts with MDM2 through its C-terminal end, approaching the N-terminal region away from the p53 binding site and allowing for allosteric changes in MDM2 conformations



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 29

34,48

. Interestingly, the encounter complex configurations corresponding to the dominant basins of

attractions of the phage-derived peptide series (Figure 2d) reveal that the peptides approach MDM2 through their N-terminal ends. This orientation is opposite to that adopted by the p53-based series. Given the observation that the p53-based and the phage-derived peptides bind MDM2 in the same orientation in the final (non-diffusive) complex

31

, this suggests that the phage-derived peptides

need to undergo a reorientation in the final approach to binding. This would incur an additional entropic cost to the binding process, suggesting an enthalpic compensation, in accord with the observation that their binding is more enthalpically driven compared to the p53-based peptide series 31

.

Modulating the mode of interaction of the phage-derived peptides with the MDM2 protein Since formation of the encounter complexes takes place away from the binding site, their formation is predominated by long-range electrostatic interactions. The electrostatic potential of the peptides (Figure 4) reveals that the phage-derived peptides exhibit regions of positive electrostatic potential that are negative in the p53-based peptides. It is tempting to hypothesize that the opposite orientations that the two series adopt upon approaching the MDM2 protein, result from these electrostatic differences between the peptides. To test this, we carried out two mutations (R3E and R3E & E9R) of the phage-derived MPRFMDYWEGLS peptide and computed their encounter complex configurations and residence times around MDM2 (Figure 5). These mutations lead to a change of the orientation of the phage-derived peptide back to that adopted by the p53-based peptides, in the vicinity of the N-terminal region of MDM2 (Figure 1d; leftmost basin). Notably, the R3E mutation has a residence time (Figure 5 a & b) that is much longer than any of the peptides (~120 µs) which is commensurate with its large contiguous/connected basins of attraction that encompass most of the MDM2 surface around the p53-binding site (Figure 5 c). Interestingly, the encounter complex ensemble of the R3E mutant (the MPEFMDYWEGLS peptide; Figure 5d) 14 ACS Paragon Plus Environment

Page 15 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


exhibits a range of order, from being more disordered away from the N-terminal region of MDM2 (rightmost basin of attraction; Figure 5 d) to being highly ordered in the vicinity of the N-terminal region of MDM2. By contrast, the R3E & E9R double mutant (MPEFMDYWRGLS peptide; Figure 5 d) shows more order away from N-terminal region of MDM2 and a shorter residence time nearby (~20µs). Previous studies revealed that Nutlin 34, a potent MDM2 inhibitor, adopts more disordered configurations (low entropic cost) upon approaching the MDM2, whereas wild-type p53 peptide adopts well-ordered configurations (high entropic cost) upon approaching the MDM2 N-terminal region through its C-terminal end

34

. The R3E mutant strikes a compromise between these two

behaviours, whereby the entropic costs are cancelled out for the overall interaction process. This suggests that the R3E phage-derived mutant MPEFMDYWEGLS could have an affinity for MDM2 that is much larger than p53-based peptides. Finally the question arises: if the R3E mutant were more potent, why did the phage display not give rise to it? A phage-display experiment explores a large sequence space (e.g. ~1081011 peptides), but the full configurational search space for a 12 amino acid peptide, assuming a repertoire of 20 naturally occurring amino-acids, is 2012. Additionally, phage-display techniques typically exhibit codon bias which reduces the diversity of the peptides generated

66

. Peptides

raised using mRNA display technology cover up to a 105-fold larger peptide space and lead to potent peptides that activated p53 by binding to MDM2 67; the amino acid at the equivalent position was again Arg. Given the complexities of the underlying mechanisms that govern selection, which includes the effects of these peptides on bacterial growth, the tolerance to substitutions, the reasons for the selection of Arg and not Glu in the phage experiments remain undetermined.

Conclusion We carried out BD simulations of two series of C-terminally substituted peptides; a p53based Q16ETFSDLWKLLX27 peptide series and a phage-derived M1PRFMDYWEGLX12 peptide



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 29

series (X = Asn, Ala, Ser and Pro), in order to investigate the kinetic and mechanistic determinants of peptide interactions with the MDM2 protein. Based on characterization of the encounter complex - a diffusively bound state that is distinct from the non-diffuse state of the final bound complex - of the peptide-MDM2 interaction, the two series of peptides showed distinct mechanisms of interactions with MDM2. The peptides of the p53-based series were all found to adopt the same mode of interaction as on approach with their C-terminal ends facing the N-terminal region of MDM2. Although, Cterminal substitution (from Pro to Ser, Ala and Asn) does not affect the mode of interaction, it was found to affect the peptide average residence times which increase in the order X=S>A>N>P. This matches the experimentally-observed trend in the binding affinities

31

. The phage-derived peptide

series, however, showed a distinct mode of interaction, approaching MDM2 in an orientation opposite to that adopted by the p53-based series. C-terminal substitutions (from Asn to Ser, Ala and Pro) do not affect the mode of interaction, but affects the peptide average residence times (in the order X=S>P>A>N) around the MDM2 surface. The computed dissociation constants of the encounter complexes suggests that post-encounter binding events are important for the phagederived series and could impact on the computed kon and koff. In contrast, our calculations suggest that for the p53-derived series the final bound complex is most likely thermodynamically driven. A designed mutant (R3E) of the MPRFMDYWEGLS phage-derived peptide reverses the peptide orientation back to the orientation adopted by the p53-based peptide series and results in a 5-fold increase in the peptide residence time (~120 µs) around MDM2. This suggests that the R3E phagederived peptide mutant could have affinity for MDM2 that is much larger than any of the p53-based series. The MPEFMDYWEGLS peptide could, therefore, be a potential lead candidate for the development of peptide-based anticancer drugs.


Page 17 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Acknowledgment This work has been supported by the National Science, Technology and Innovation Plan (NSTIP) program by Qassim University, Project Number: 13-BIO889-09. The authors gratefully acknowledge provision of computing resources by Drs. Garib Murshudov and Seishi Shimizu at the University of York.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 29

References (1) Vogelstein, B.; Lane, D.; Levine, A. J. Surfing the p53 Network. Nature 2000, 408, 307-310. (2) Brown, C. J.; Lain, S.; Verma, C. S.; Fersht, A. R.; Lane, D. P. Awakening Guardian Angels: Drugging the p53 Pathway. Nat. Rev. Cancer 2009, 9, 862-873. (3) Lane, D. P.; Hupp, T. R. Drug Discovery and p53. Drug Discov. Today 2003, 8, 347355. (4) Momand, J.; Wu, H.-H.; Dasgupta, G. MDM2 -- Master Regulator of the p53 Tumor Suppressor Protein. Gene 2000, 242, 15-29. (5) Shmueli, A.; Oren, M. Regulation of P53 by MDM2: Fate Is in the Numbers. Mol. Cell 2004, 13, 4-5. (6) Madhumalar, A.; Lee, H. J.; Brown, C. J.; Lane, D.; Verma, C. Design of a Novel MDM2 Binding Peptide Based on the p53 Family. Cell Cycle 2009, 8, 2828-2836. (7) Chen, L.; Yin, H.; Farooqi, B.; Sebti, S.; Hamilton, A. D.; Chen, J. p53 Alpha-Helix Mimetics Antagonize p53/MDM2 Interaction and Activate p53. Mol. Cancer Ther. 2005, 4, 10191025. (8) Fox, J. L. Antivirals Become a Broader Enterprise. Nat. Biotechnol. 2007, 25, 13951402. (9) Verma, C.; Seebah, S.; Low, S. M.; Zhou, L.; Liu, S. P.; Li, J.; Beuerman, R. W. Defensins: Antimicrobial Peptides for Therapeutic Development. Biotechnol. J. 2007, 2, 1353-1359. (10) Cejka, D.; Losert, D.; Wacheck, V. Short Interfering RNA (siRNA): Tool or Therapeutic? Clin. Sci. (Lond.) 2006, 110, 47-58. (11) Branco, M. C.; Schneider, J. P. Self-Assembling Materials for Therapeutic Delivery. Acta Biomater. 2009, 5, 817-831. (12) Kogan, M. J.; Olmedo, I.; Hosta, L.; Guerrero, A. R.; Cruz, L. J.; Albericio, F. Peptides and Metallic Nanoparticles for Biomedical Applications. Nanomed. 2007, 2, 287-306. (13) Maher, S.; Brayden, D. J.; Feighery, L.; McClean, S. Cracking the Junction: Update on the Progress of Gastrointestinal Absorption Enhancement in the Delivery of Poorly Absorbed Drugs. Crit. Rev. Ther. Drug Carrier Syst. 2008, 25, 117-168. (14) Guo, Z.; Mohanty, U.; Noehre, J.; Sawyer, T. K.; Sherman, W.; Krilov, G. Probing the α-Helical Structural Stability of Stapled p53 Peptides: Molecular Dynamics Simulations and Analysis. Chem. Biol. Drug Des. 2010, 75, 348-359. (15) Walensky, L. D.; Kung, A. L.; Escher, I.; Malia, T. J.; Barbuto, S.; Wright, R. D.; Wagner, G.; Verdine, G. L.; Korsmeyer, S. J. Activation of Apoptosis in Vivo by a HydrocarbonStapled Bh3 Helix. Science 2004, 305, 1466-1470. (16) Borghouts, C.; Kunz, C.; Groner, B. Current Strategies for the Development of Peptide-Based Anti-Cancer Therapeutics. J. Pept. Sci. 2005, 11, 713-726. (17) Craik, D. J.; Fairlie, D. P.; Liras, S.; Price, D. The Future of Peptide-Based Drugs. Chem. Biol. Drug Des. 2013, 81, 136-147. (18) Mendoza, F. J.; Espino, P. S.; Cann, K. L.; Bristow, N.; McCrea, K.; Los, M. AntiTumor Chemotherapy Utilizing Peptide-Based Approaches: Apoptotic Pathways, Kinases, and Proteasome as Targets. Arch. Immunol. Ther. Exp. (Warsz.) 2005, 53, 47-60. (19) Vagner, J.; Qu, H.; Hruby, V. J. Peptidomimetics, a Synthetic Tool of Drug Discovery. Curr. Opin. Chem. Biol. 2008, 12, 292-296. (20) Tsomaia, N. Peptide Therapeutics: Targeting the Undruggable Space. Eur. J. Med. Chem. 2015, 94, 459-470. (21) Fischer, P. M. Peptide, Peptidomimetic, and Small-Molecule Antagonists of the p53HDM2 Protein-Protein Interaction. Int. J. Pept. Res. Ther. 2006, 12, 3-19. 18 ACS Paragon Plus Environment

Page 19 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(22) Rew, Y.; Sun, D.; Yan, X.; Beck, H. P.; Canon, J.; Chen, A.; Duquette, J.; Eksterowicz, J.; Fox, B. M.; Fu, J. et al. Discovery of Am-7209, a Potent and Selective 4Amidobenzoic Acid Inhibitor of the MDM2-p53 Interaction. J. Med. Chem. 2014, 57, 10499-10511. (23) Nagata, T.; Shirakawa, K.; Kobayashi, N.; Shiheido, H.; Tabata, N.; SakumaYonemura, Y.; Horisawa, K.; Katahira, M.; Doi, N.; Yanagawa, H. Structural Basis for Inhibition of the MDM2:p53 Interaction by an Optimized Mdm2-Binding Peptide Selected with Mrna Display. PloS one 2014, 9, e109163. (24) Chee, S. M.; Wongsantichon, J.; Soo Tng, Q.; Robinson, R.; Joseph, T. L.; Verma, C.; Lane, D. P.; Brown, C. J.; Ghadessy, F. J. Structure of a Stapled Peptide Antagonist Bound to Nutlin-Resistant MDM2. PLoS One 2014, 9, e104914. (25) Fry, D. C.; Emerson, S. D.; Palme, S.; Vu, B. T.; Liu, C. M.; Podlaski, F. NMR Structure of a Complex between MDM2 and a Small Molecule Inhibitor. J. Biomol. NMR 2004, 30, 163-173. (26) Uhrinova, S.; Uhrin, D.; Powers, H.; Watt, K.; Zheleva, D.; Fischer, P.; McInnes, C.; Barlow, P. N. Structure of Free MDM2 N-Terminal Domain Reveals Conformational Adjustments That Accompany p53-Binding. J. Mol. Biol. 2005, 350, 587-598. (27) Kussie, P. H.; Gorina, S.; Marechal, V.; Elenbaas, B.; Moreau, J.; Levine, A. J.; Pavletich, N. P. Structure of the MDM2 Oncoprotein Bound to the p53 Tumor Suppressor Transactivation Domain. Science 1996, 274, 948-953. (28) Fuller, J. C.; Jackson, R. M.; Edwards, T. A.; Wilson, A. J.; Shirts, M. R. Modeling of Arylamide Helix Mimetics in the p53 Peptide Binding Site of HDM2 Suggests Parallel and AntiParallel Conformations Are Both Stable. PLoS One 2012, 7, e43253. (29) Bottger, A.; Bottger, V.; Sparks, A.; Liu, W. L.; Howard, S. F.; Lane, D. P. Design of a Synthetic MDM2-Binding Mini Protein That Activates the p53 Response in Vivo. Curr. Biol. 1997, 7, 860-869. (30) Zondlo, S. C.; Lee, A. E.; Zondlo, N. J. Determinants of Specificity of MDM2 for the Activation Domains of p53 and p65: Proline27 Disrupts the MDM2-Binding Motif of p53. Biochemistry 2006, 45, 11945-11957. (31) Brown, C. J.; Dastidar, S. G.; Quah, S. T.; Lim, A.; Chia, B.; Verma, C. S. CTerminal Substitution of MDM2 Interacting Peptides Modulates Binding Affinity by Distinctive Mechanisms. PLoS One 2011, 6, e24122. (32) Bisignano, P.; Doerr, S.; Harvey, M. J.; Favia, A. D.; Cavalli, A.; De Fabritiis, G. Kinetic Characterization of Fragment Binding in Ampc Î²-Lactamase by High-Throughput Molecular Simulations. J. Chem. Inf. Model. 2015, 54, 362-366. (33) Shukla, D.; HernÃ¡ndez, C. X.; Weber, J. K.; Pande, V. S. Markov State Models Provide Insights into Dynamic Modulation of Protein Function. Acc. Chem. Res. 2015, 48, 414-422. (34) ElSawy, K. M.; Verma, C. S.; Joseph, T. L.; Lane, D. P.; Twarock, R.; Caves, L. S. On the Interaction Mechanisms of a p53 Peptide and Nutlin with the MDM2 and MDMX Proteins: A Brownian Dynamics Study. Cell Cycle 2013, 12, 394-404. (35) Copeland, R. A.; Pompliano, D. L.; Meek, T. D. Drug-Target Residence Time and Its Implications for Lead Optimization. Nat. Rev. Drug Discovery 2006, 5, 730-739. (36) Zhang, R.; Monsma, F. The Importance of Drug-Target Residence Time. Curr. Opin. Drug Discov. Devel. 2009, 12, 488-496. (37) Tummino, P. J.; Copeland, R. A. Residence Time of Receptor-Ligand Complexes and Its Effect on Biological Function. Biochemistry 2008, 47, 5481-5492. (38) Copeland, R. A. The Dynamics of Drug-Target Interactions: Drug-Target Residence Time and Its Impact on Efficacy and Safety. Expert Opin. Drug. Discov. 2010, 5, 305-310. (39) Lu, H.; Tonge, P. J. Drug-Target Residence Time: Critical Information for Lead Optimization. Curr. Opin. Chem. Biol. 2010, 14, 467-474.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 29

(40) ElSawy, K. M.; Twarock, R.; Lane, D. P.; Verma, C. S.; Caves, L. S. D. Characterization of the Ligand Receptor Encounter Complex and Its Potential for in Silico KineticsBased Drug Development. J. Chem.Theory Comput. 2012, 8, 314-321. (41) B-Rao, C.; Subramanian, J.; Sharma, S. D. Managing Protein Flexibility in Docking and Its Applications. Drug Discov. Today 2009, 14, 394-400. (42) McCammon, J. A. Target Flexibility in Molecular Recognition. Biochim. Biophys. Acta 2005, 1754, 221-224. (43) Stefan, H.; Outi, M. H. S.-A.; Bingding, H.; Friedrich, F. R.; Gabriele, C.; Rebecca, C. W. Computational Approaches to Identifying and Characterizing Protein Binding Sites for Ligand Design. J. Mol. Recognit. 2010, 23, 209-219. (44) Gilson, M. K.; Zhou, H. X. Calculation of Protein-Ligand Binding Affinities. Annu. Rev. Biophys. Biomol. Struct. 2007, 36, 21-42. (45) Shan, Y.; Kim, E. T.; Eastwood, M. P.; Dror, R. O.; Seeliger, M. A.; Shaw, D. E. How Does a Drug Molecule Find Its Target Binding Site? J. Am. Chem. Soc. 2011, 133, 9181-9183. (46) Pande, V. S.; Beauchamp, K.; Bowman, G. R. Everything You Wanted to Know About Markov State Models but Were Afraid to Ask. Methods 2010, 52, 99-105. (47) Gabdoulline, R. R.; Wade, R. C. Brownian Dynamics Simulation of Protein-Protein Diffusional Encounter. Methods 1998, 14, 329-341. (48) Dastidar, S. G.; Lane, D. P.; Verma, C. S. Modulation of p53 Binding to MDM2: Computational Studies Reveal Important Roles of Tyr100. BMC Bioinformatics 2009, 10 Suppl 15, S6. (49) Brooks, B. R.; Brooks, I. C. L.; Mackerell, J. A. D.; Nilsson, L.; Petrella, R. J.; Roux, B.; Won, Y.; Archontis, G.; Bartels, C.; Boresch, S. et al. CHARMM: The Biomolecular Simulation Program. J. Comput. Chem. 2009, 30, 1545-1614. (50) Caves, L. S. D.; Evanseck, J. D.; Karplus, M. Locally Accessible Conformations of Proteins: Multiple Molecular Dynamics Simulations of Crambin. Protein Sci. 1998, 7, 649-666. (51) ElSawy, K. M.; Hodgson, M. K.; Caves, L. S. D. The Physical Determinants of the DNA Conformational Landscape: An Analysis of the Potential Energy Surface of Single-Strand Dinucleotides in the Conformational Space of Duplex DNA. Nucleic Acids Res. 2005, 5749-5762. (52) Donald, L. E.; McCammon, J. A. Brownian Dynamics with Hydrodynamic Interactions. J. Chem. Phys. 1978, 69, 1352-1360. (53) García de la Torre, J.; Huertas, M. L.; Carrasco, B. Calculation of Hydrodynamic Properties of Globular Proteins from Their Atomic-Level Structure. Biophys. J. 2000, 78, 719-730. (54) Im, W.; Beglov, D.; Roux, B. Continuum Solvation Model: Computation of Electrostatic Forces from Numerical Solutions to the Poisson-Boltzmann Equation. Comput. Phys. Commun. 1998, 111, 59-75. (55) Coalson, R.; Beck, T. L. Numerical Methods for Solving Poisson and PoissonBoltzmann Type Equations. In Encyclopedia of Computational Chemistry; von Rague Schleyer, P., Ed.; John-Wiley: New York, 1998; Vol. 3; pp 2086-2100. (56) Baker, N. A.; Sept, D.; Joseph, S.; Holst, M. J.; McCammon, J. A. Electrostatics of Nanosystems: Application to Microtubules and the Ribosome. Proc. Natl. Acad. Sci. U. S. A. 2001, 98, 10037-10041. (57) Dolinsky, T. J.; Czodrowski, P.; Li, H.; Nielsen, J. E.; Jensen, J. H.; Klebe, G.; Baker, N. A. Pdb2pqr: Expanding and Upgrading Automated Preparation of Biomolecular Structures for Molecular Simulations. Nucleic Acids Res. 2007, 35, W522-W525. (58) Czodrowski, P.; Dramburg, I.; Sotriffer, C. A.; Klebe, G. Development, Validation, and Application of Adapted PEOE Charges to Estimate pKa Values of Functional Groups in Protein–Ligand Complexes. Proteins: Struct., Funct., Bioinf. 2006, 65, 424-437.


Page 21 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(59) Elcock, A. H.; Gabdoulline, R. R.; Wade, R. C.; McCammon, J. A. Computer Simulation of Protein-Protein Association Kinetics: Acetylcholinesterase-Fasciculin. J. Mol. Biol. 1999, 291, 149-162. (60) Gabdoulline, R. R.; Wade, R. C. Protein-Protein Association: Investigation of Factors Influencing Association Rates by Brownian Dynamics Simulations. J. Mol. Biol. 2001, 306, 1139-1155. (61) Gabdoulline, R. R.; Wade, R. C. Effective Charges for Macromolecules in Solvent. J. Phys. Chem. 1996, 100, 3868-3878. (62) Gabdoulline, R. R.; Wade, R. C. Simulation of the Diffusional Association of Barnase and Barstar. Biophys. J. 1997, 72, 1917-1929. (63) Scott, H. N.; Stuart, A. A.; McCammon, J. A. Brownian Dynamics Simulation of Diffusion-Influenced Bimolecular Reactions. J. Chem. Phys. 1984, 80, 1517-1524. (64) ElSawy, K. M.; Verma, C. S.; Lane, D. P.; Caves, L. S. On the Origin of the Stereoselective Affinity of Nutlin-3 Geometrical Isomers for the MDM2 Protein. Cell Cycle 2013, 12, 3727-3735. (65) ElSawy, K. M.; Sim, A.; Lane, D. P.; Verma, C. S.; Caves, L. S. A Spatiotemporal Characterization of the Effect of p53 Phosphorylation on Its Interaction with MDM2. Cell Cycle 2014, 14, 179-188. (66) Gray, B. P.; Brown, K. C. Combinatorial Peptide Libraries: Mining for Cell-Binding Peptides. Chem. Rev. 2014, 114, 1020-1081. (67) Shiheido, H.; Takashima, H.; Doi, N.; Yanagawa, H. Mrna Display Selection of an Optimized MDM2-Binding Peptide That Potently Inhibits MDM2-p53 Interaction. PLoS One 2011, 6, e17898.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Page 22 of 29

22

ACS Paragon Plus Environment

Page 23 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48


Figure 1 (a) The radial profile of the peptide residence times (RT) around the MDM2 protein, (b) the peptide residence time in the dominant basin of attraction, (c) the location of the peptide MDM2 basins of attraction, and (d) the corresponding ensemble of structures of the peptide MDM2 encounter complex for the QETFSDLWKLLS, QETFSDLWKLLA, QETFSDLWKLLN, QETFSDLWKLLP peptides column-wise. The peptides are arranged from left to right in order of decreasing experimental binding affinity to MDM2 (-9.54, -9.31, -9.08 and -7.79 kCal/mol) 31. In a, the residence time profiles were spatially normalized by dividing total residence time at a radius r by the spherical volume slab 4πr2 and then averaged over the number of BD trajectories. In (b & c) the basins of attraction are coloured in grey, green, orange and red for respective residence time intervals centred at 0, 10, 20 and 25 ns. In (c &d) the MDM2 protein is shown in blue and the p53 peptide in its final bound state to MDM2 is shown in cyan based on PDBID:1YCR. In d, in order to indicate the direction of the peptides, each peptide structure is coloured according to its atom index using a colour palette that changes smoothly from blue (N-terminal) to red (C-terminal).

23



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Page 24 of 29

24


Page 25 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48


Figure 2 (a) The radial profile of the peptide residence times (RT) around the MDM2 protein, (b) the peptide residence time in the dominant basin of attraction, (c) the location of the peptide MDM2 basins of attraction, and (d) the corresponding ensemble of structures of the peptide MDM2 encounter complex for the MPRFMDYWEGLS, MPRFMDYWEGLA, MPRFMDYWEGLP and MPRFMDYWEGLN peptides column-wise. The peptides are arranged from left to right in order of decreasing experimental binding affinity to MDM2 (-10.35, -9.82, -8.94 and -8.87 kCal/mol) 31. The colour schemes used in (b, c & d) are similar to those used in Figure 1 (b, c & d).

25



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 29

a)

b)

Figure 3 The computed dissociation constants; Kd (a) and the association rate constants; kon (b) corresponding to the formation of the dominant encounter complex basins for the p53-based peptide series (left) and the phage-derived 12/1 peptide series (right). The colour scheme follows that used in Figure 1 (b, c & d).


Page 27 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48


Figure 4 The electrostatic potential of (a) the p53-based peptide series and (b) the phage optimized 12/1 peptide series mapped to the peptide molecular surface defined by a probe radius of 1.4 Ǻ. The electrostatic potential colour ramp changes smoothly from –kT (red) to +kT (blue) (kT is ~0.6 kcal mol-1 at 298K).

27



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Page 28 of 29

Figure 5 (a) The radial profile of the peptide average residence times (RT) around the MDM2 protein, (b) the peptide residence time in the basins of attraction, (c) the location of the peptide MDM2 basins of attraction, and (d) the corresponding ensemble of structures of the peptide MDM2 encounter complex for the MPEFMDYWEGLS (R3E mutant), MPEFMDYWRGLS (R3E & E9R mutant) columnwise.. The colour schemes used in (b, c & d) are similar to those used in Figure 1 (b, c & d).

28


Page 29 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TOC graphic


Recognition Dynamics of p53 and MDM2: Implications for Peptide

Recommend Documents