Substrate Sequence Determines Catalytic ... - ACS Publications

Subscriber access provided by University of Massachusetts Amherst Libraries

B: Biophysics; Physical Chemistry of Biological Systems and Biomolecules

Substrate Sequence Determines Catalytic Activities, DomainBinding Preferences, and Allosteric Mechanisms in Pin1 Mohamed Faizan Momin, Xin-Qiu Yao, Waygen Thor, and Donald Hamelberg J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.8b03819 • Publication Date (Web): 31 May 2018 Downloaded from http://pubs.acs.org on May 31, 2018

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 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 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.

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 26 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

Substrate Sequence Determines Catalytic activities, DomainBinding Preferences, and Allosteric Mechanisms in Pin1

Mohamed Momin1, Xin-Qiu Yao1, Waygen Thor1, and Donald Hamelberg1,2*

1

Department of Chemistry and 2Center for Diagnostics and Therapeutics, Georgia State University, Atlanta, GA 30302-3965, United States

*To whom correspondence should be addressed, Dr. Donald Hamelberg, P. O. Box 3965, Atlanta, GA 30302-3965. Email: [email protected]; Tel: 404-413-5564, Fax: 404-513-5505

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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 26

ABSTRACT Pin1 is a unique phosphorylation dependent peptidyl-prolyl isomerase that regulates diverse subcellular processes and an important potential therapeutic target. Functional mechanisms of Pin1 are complicated because of the two-domain structural organization: the catalytic domain both binds the specific pSer/Thr-Pro motif and catalyzes the cis/trans isomerization, whereas the WW domain can only bind the trans configuration and is speculated to be responsible for substrate binding specificity. Numerous studies of Pin1 have led to two divergent conclusions on the functional role of the WW domain. One opinion states that the WW domain is an allosteric effector and substrate binding to this domain modulates the binding and catalysis in the distal catalytic domain. The other opinion, however, argues that the WW domain does not have any allosteric role. Here, using molecular dynamics and binding free energy calculations, we examine catalysis and allosteric mechanisms in Pin1 under various substrate- and WW-binding conditions. Our results reveal a strong substrate sequence dependency of catalysis, domain-binding preferences, and allosteric outputs in Pin1. Importantly, we show that the different opinions about the WW domain can be unified in one framework, in which substrate sequences determine whether a positive, negative, or neural allosteric effect will be elicited. Our work further elucidates detailed mechanisms underlying the sequence dependent allostery of Pin1 and finds that inter-domain contacts are key mediators of intra-protein allosteric communications. Our findings collectively provide new insights into the function of Pin1, which may facilitate the development of novel therapeutic drugs targeting Pin1 in the future.


Page 3 of 26 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 The modulation of protein function via allosteric effects is a biological phenomenon prevalent in diverse subcellular processes and probably an intrinsic nature of all proteins that are inherently dynamic.1-3 Allosteric modulation represents the transmission of a signal, such as prolyl isomerization, ligand/effector protein binding, mutation, post-translational modification, or response to pH changes, from a distal site to the active site. Allosteric communications can be mediated through either conformational changes4-5 or dynamical couplings6 and underlying mechanisms are usually system dependent. Allostery is a critical factor that must be considered in protein design7 and the biological foundation for developing allosteric drugs.8-9 Despite its importance, the allosteric mechanism in many biological systems is still poorly understood. Elucidating the process of an allosteric transition at the atomistic level remains a grand challenge in biochemistry.

Human Pin1 is a model allosteric system that has gained increasing interests in recent years. Pin1 is a unique phosphorylation-dependent peptidyl-prolyl cis/trans isomerase (PPIase), which belongs to the parvulin family of isomerases and regulates mitosis.10-11 Pin1 interacts with a variety of phosphoproteins and accelerates the conversion of phosphorylated pSer/Thr-Pro peptide bonds between cis and trans conformations by >1000-fold.12 Pin1 is overexpressed in many types of cancer while its down-regulation is related to neurodegenerative diseases such as Alzheimer’s disease, making Pin1 an important potential therapeutic target.13-17 A recent study suggested that Pin1 could also be a target of dietary therapies.18 Pin1 has two domains: the carboxy-terminus PPIase domain contains the catalytic site and the amino-terminus non-catalytic WW domain is mainly responsible for substrate binding specificity 3 ACS Paragon Plus Environment


Page 4 of 26

(Figure 1).19 While both domains of Pin1 recognize pSer/Thr-Pro motifs (via Ser16, Tyr23, Ser32, and Trp34 in the WW domain and Lys63, Arg68, Arg69, Leu122, Met130, and Phe134 in the catalytic domain),19 the WW domain has a ten-fold higher binding affinity for peptides than the PPIase or catalytic domain.20 Recent experiments revealed that the WW domain also plays as an allosteric effector of the catalytic domain. Using nuclear magnetic resonance (NMR), Peng and his colleagues characterized dynamical responses of Pin1 to a short Pin1 substrate (FFpSPR) and its peptidomimetic analogs and identified an internal conduit comprised of conserved hydrophobic residues connecting the active site to the domain-domain interface.21 This hydrophobic conduit likely mediates the communication within the catalytic domain. They later showed that substrate binding to the WW domain disrupted inter-domain contacts leading to enhanced catalytic activities, suggesting a negative allosteric modulation by WW binding.22 Recent computational studies supported the allosteric role of the WW domain. Using molecular dynamics (MD) and a network analysis approach, Zhou lab revealed a concerted stabilization of catalytic loops and substrate binding to the WW domain and discovered two allosteric pathways connecting WW to catalytic loops.23 Using MD and linear response theory based methods, Ozkan and colleagues found overall alterations of dynamic flexibility profiles, dynamic couplings, and allosteric pathways in Pin1 upon substrate binding to WW.24 In addition, our previous simulation study revealed conformation dependent modulations of the dynamics and binding affinity in the catalytic domain upon substrate-WW binding or a point mutation at the domain-domain interface.25 These studies collectively show that the WW domain not merely enhances the local concentration of pSer/Thr-Pro motifs by binding to specific substrates, as suggested by bivalent binding models,26 but allosterically modulates binding and catalytic activities in the catalytic domain through intraprotein communications.


Page 5 of 26 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


Surprisingly, a recent experimental study by Kern lab challenged several existing models about the function of Pin1, including the allosteric role of the WW domain.27 They found that the binding of peptides derived from human tau, an important protein promoting microtubule assembly and related to Alzheimer’s disease, to the WW domain did not affect either binding or catalysis in the catalytic domain. The authors also questioned whether tau is a biological substrate of Pin1 and pointed out that pThr231, a site of tau that has been related to significant biological phenotypes in previous studies, could not be catalyzed. A notable feature around pThr231 is that there is an additional proline at the +1 position after the pSer/Thr-Pro motif. A previous study showed that substrates following such a sequence pattern had a strong preference to bind the WW domain.28 It is unclear whether the findings of Kern lab are specific to certain substrates or general mechanisms of Pin1.

Perhaps it is not so surprising that tau-WW binding did not lead to an increase of local pSer/ThrPro concentration to gain an enhanced binding affinity in the catalytic domain – the two phosphorylation sites examined, pThr231 and pSer235, are only four residues separated, which is not long enough to accommodate a bivalent binding.26 Why were allosteric regulations through intra-Pin1 communications absent either? One possible explanation is that the substrates used in Kern lab’s (tau-derived peptides) and previous studies (Cdc25C- or FFpSPR-derived peptides) are different, which may cause different effects. It is likely that the amino acids flanking the pSer/ThrPro motif determine whether a positive, negative, or neutral allosteric effect will be elicited. To test this hypothesis, we examined conformational dynamics and binding free energies in the catalytic domain upon binding several distinct peptides in the WW domain using long-time multiple-microsecond MD simulations and associated analyses. We first show that substrates 5 ACS Paragon Plus Environment


Page 6 of 26

containing a +1 proline are indeed WW preferred and cannot be catalyzed, validating our methodologies. Then, we argue that the direction and magnitude of allosteric effects are dependent on the sequence of substrate bound in the WW domain. In particular, while FFpSPR caused a remarkable positive allosteric effect, a peptide derived from tau had a negative to zero effect, thus unifying the otherwise contradicting previous studies. The sequence dependent allosteric mechanism of Pin1 echoes the widely studied biased agonisms in GPCR, where binding of distinct ligands to the same membrane receptor elicits either G protein or β arrestin mediated signaling pathways.29


Page 7 of 26 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


Figure 1. Pin1 with phosphorylated substrates bound in both the catalytic domain and WW domain. Pin1 (PDB: 2Q5A) is represented as molecular surface and substrates are as licorice. The substrate in the catalytic domain (green) is in the cis conformation, whereas the substrate in the WW domain (pink) is in the trans conformation. The flexible linker is colored white. The graphic is generated with VMD.30

COMPUTATIONAL METHODS Molecular dynamics simulations were carried out using the Amber16 suite of programs31 and the AMBER ff14SB32, a modified version of the Cornell et al. (1995)33 force field. Initial coordinates of systems were taken from a previous simulation of Pin1/substrates complex.25 The substrate in the structure was modified to match sequences to be simulated (See Table 1 for full sequences of all substrates). Side-chain atoms of substrates were removed from the initial structure and then added back by the LEaP module of AmberTools.31 The re-optimized dihedral parameters34



Page 8 of 26

developed to properly describe the free energy of a peptidyl-prolyl bond were used. The parameters for the phosphorylated tyrosine were taken from Steinbrecher et al.35

Each system was solvated in a periodic octahedron box with approximately 8,000 TIP3P explicit water molecules,36-37 ensuring that the system is at least 10 Å away from the nearest face of the box. All systems were neutralized by adding counter ions (Na+ or Cl-), and were then subjected to a total of 5,000 steps of energy minimization (2,000 steps of steepest descent followed by 3,000 steps of conjugate gradient) and six steps of 1-ns equilibration. At the first five equilibration steps, positions of all protein atoms were restrained with a harmonic potential, with the force constant of the restraint potential gradually reduced from 500 to 5 kcal/mol/Å2. The final equilibration step was carried out allowing the whole system to move freely. All simulations were conducted at a constant temperature of 300 K (Langevin thermostat with collision frequency γ=1.0 ps-1) and a constant pressure of 1 bar (Monte Carlo barostat with coupling constant τp=1.0 ps). A 2-fs time step was used to solve the equation of motion. To calculate electrostatic interactions, the particle mesh Ewald (PME) summation method was used with a 9-Å cutoff for long-range non-bonded interactions.38 A harmonic dihedral-restraint (with force constant 200 kcal/mol • rad2) was imposed on substrates in the TS state to restrict the peptidyl-prolyl bond torsion angle ω to be ~+90°.

The molecular mechanics/Poisson-Boltzmann surface area (MM/PBSA) approach39 was used to estimate the binding affinity between Pin1 and substrates. Gas phase energies were computed with the ff14SB force field.32-33 The polar term of solvation energy was calculated by solving the linear Poisson-Boltzmann equation using the SANDER program of AMBER16.31 Dielectric constants of


Page 9 of 26 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


protein interior and exterior were set to 1.0 and 80, respectively. The ionic strength was set to the default 0.0 mM. The nonpolar solvation energy was estimated by summing up a dispersion term using a surface-based integration method40 and a cavity term that is linearly proportional to the molecular solvent-accessible surface area. The binding energy was calculated over every 10 frames of a simulation (cumulatively 105 energy values for each system). The residue-residue contact analysis similar to previsouly described25, 41 was carried out to investigate vital conformational changes under various substrate-bound states. A contact between two non-adjacent residues is considered to be formed if any two inter-residue heavy atoms are within 4.5 Å. The probability of contact formation was then calculated over the entire simulation trajectory and compared between systems. The principal component analysis (PCA) was also applied to contact dynamics. To do so, contact formation probability was first averaged over simulation systems. For most residue pairs, contacts are seldom formed (averaged probablity f90%). Only a small percentage of residue-residue contacts are in the middle (10%≤f≤90%), which are referred to “dynamic contacts.”41 The trajectory of dynamic contacts was generated, where status of each dynamic contact at each time point was represented by ‘1’ (formed) and ‘0’ (broken). PCA was then performed on the binary contact trajectory in the same manner as PCA on normal Cartesian coordinates. Dynamical (Pearson) cross-correlation was calculated for every pair of Cα atoms using the Bio3D R package.42-43 Calculations were performed over 104 equally spaced frames for each simulation. Prior to the calculations, all simulation snapshots were superimposed onto the first frame based on Cartesian coordinates of all Cα atoms of Pin1 excluding the flexible linker.



Page 10 of 26

RESULTS & DISCUSSION Table 1. List of substrates. The subscript “p” before amino acid codes designates phosporylated residues and residues in bold designate the pSer/Thr-Pro motifs. Substrate Name pSPS pTPV pTPP pSPR

Substrate Sequence Ace-TPPKpSPSSAK-Nme Ace-EQPLpTPVTDL-Nme Ace-AVVRpTPPKSP-Nme Ace-FFpSPR-Nme

Protein Name Human tau Human Cdc25C Human tau Artificial peptide11

Extensive multi-microsecond simulations were performed to investigate, on the atomic scale, mechanisms underlying the peptidyl-prolyl isomerization in Pin1. Sixteen simulations (totaling 25+µs) were performed under various combinations of substrate binding and domain-domain interaction conditions (See Table 1 and Table 2 for a list of substrates and a full list of simulations). The (substrate) sequence dependency of catalysis was examined with an isolated catalytic domain, where the free energy profile of full isomerization reaction (represented by the cis, TS, and trans conformation of substrate) was estimated for different substrates. Then, comparing isolated catalytic and WW domains characterized domain-specific substrate binding. At last, sequence- and WW-dependent allosteric effects were tested by evaluating the binding affinity of substrate in the catalytic domain in the absence or presence of the WW domain, or with the WW domain bound with different substrates. In all comparisons, the binding free energy was calculated with MM/PBSA.


Page 11 of 26 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


Table 2. List of simulations. Simulation

Catalytic Domain

X-Pro ω-bond angle (°)

WW Domain

X-Pro ω-bond angle (°)

Time (µs)

1 2 3

pSPS

0 90 180

Removed Removed Removed

-

2.2 2.1 1.2

4 5 6

pTPP pTPP

0 90 180

Removed Removed Removed

-

1.2 1.2 1.3

7 8

Removed Removed

-

pSPS pTPP

180 180

1.2 1.2

9 10 11 12

pTPV

0 0 0 0

Removed Apo pTPP pSPR

180 180

1.8 1.2 3.8 1.5

13 14 15

pSPS

0 0 0

Apo pTPP pSPR

180 180

2.2 2.3 1.2

pSPS pSPS

pTPP

pTPV pTPV pTPV

pSPS pSPS

Sequence dependent catalysis and domain binding preference in Pin1. While Pin1 catalyzes the isomerization of the substrate pSPS, it is unable to catalyze pTPP. The ability of Pin1 to stabilize the transition state of the substrate relative to the cis and trans states is used to gauge if the reaction can occur by measuring the binding free energy of substrate in the form of reactant, transition state mimic, and product. It shows that the mean binding free energy of pSPS in the TS conformation (-16.51 kcal/mol) is much lower than that in the trans (-4.46 kcal/mol) and cis (-2.92 kcal/mol) conformation (Figure 2A). This energetic difference results in a substantially reduced energy barrier for reaction to proceed within Pin1 compared with that in solvent (Figure 2C), indicating that catalysis can occur. In contrast, for pTPP, both cis and TS conformation show an overall positive binding free energy (with mean energy being 16.68 and 11 ACS Paragon Plus Environment


Page 12 of 26

10.26 kcal/mol, respectively), whereas the trans conformation has a negative mean binding energy (-3.36 kcal/mol) (Figure 2B). The results suggest that catalysis cannot happen for this substrate due to the fact that Pin1 does not bind the substrate in the transition state (Figure 2D). These results suggest that an extra proline immediately after the pSer/Thr-Pro motif in Pin1 substrates, which is present in pTPP but absent in pSPS, imposes specific geometric restraints that may impede catalysis. Our results are consistent with the recent experimental study showing that pSer/Thr-ProPro could not be catalyzed.27

Figure 2. Probability distributions of binding energies of substrate binding in an isolated Pin1 catalytic domain and schematic free energy profiles of isomerization reactions in and out of Pin1. (A,B) Distributions of binding energies for the (A) pSPS and (B) pTPP substrate in the 12 ACS Paragon Plus Environment

Page 13 of 26 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


cis (black), trans (green), and transition state (TS; red) conformation. (C,D) Schematic free energy profiles derived based on the binding energies for (C) pSPS and (D) pTPP. E+S and E:S represent the unbound and bound state, respectively.

The pTPP substrate has a strong binding preference toward the WW domain whereas pSPS has either a weak domain binding specificity or a binding preference toward the catalytic domain dependending on the peptidyl-prolyl peptide conformation. MM/PBSA energies of pTPP and pSPS were calculated for both isolated WW and catalytic domain. Since the WW domain is known to not able to bind substrates in the cis conformation, here we only calculated energies of substrates in the trans conformation for the WW domain while considered both cis and trans conformations for the catalytic domain. Measuring binding energies of substrates under the trans conformation shows that pTPP has a much lower mean binding energy (-31.28 kcal/mol) in the WW domain than that in the catalytic domain (-3.36 kcal/mol), indicating a strong domain binding preference of this substrate (Figure 3A). In contrast, pSPS has a much weaker domain binding specificity under the trans conformation, with mean binding free energies in the WW and catalytic domain being -12.31 and -4.46 kcal/mol, respectively, both of which are much higher than that of the pTPP-WW binding (Figure 3B). However, under the cis conformation pSPS does show a domain-binding preference toward the catalytic domain – the binding free energy of pSPS in the cis conformation in the catalytic domain is similar to that in the trans conformation (Figure 3C) while a cis-pSPS cannot bind to the WW. Intriguingly, the pTPP-catalytic binding energy under the cis conformation is much larger than 0 (Figure 3C), indicating that a cis-pTPP does not bind Pin1 at all. The results here suggest that there is a sequence (and cis-trans conformation) dependency of domain-binding preferences. A previous study classified Pin1 substrates into two groups: One has a signature



Page 14 of 26

proline at the +1 position after the pSer/Thr-Pro motif and can bind the WW domain even in the absence of the catalytic domain, whereas the other group does not have the +1 proline and the binding to Pin1 needs the presence of the catalytic domains.28 Our computational simulations largely agree with the experiments.

Figure 3. Probability distributions of binding energies of pSPS and pTPP in isolated domains of Pin1. Probability distributions of binding energies of pSPS (black) and pTPP (red) in isolated (A) WW and (B, C) catalytic domain are shown. For (A, B) cases, the substrate peptidyl-prolyl bond between the phosphorylated residue and the proline is in the trans conformation, while for (C) the substrate peptidyl-prolyl bond is in the cis conformation.

Sequence and WW domain dependent allosteric mechanism. The WW domain can modulate the binding event in the catalytic domain. To further test the sequence dependency of allosteric regulation, we performed multiple microsecond simulations of full-length Pin1, where the substrate pSPS was bound in the active site of the catalytic domain and various substrates were bound in the WW domain. For comparison, we also performed simulations with an apo WW domain and with WW domain completely removed (designated by ∆WW). It shows that in the presence of the WW domain the binding in the catalytic domain is slightly enhanced, with mean binding free energy decreasing from -2.92 to -9.33 kcal/mol (Figure 4A, 14 ACS Paragon Plus Environment

Page 15 of 26 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


brown versus black). Binding affinity in the catalytic domain is also influenced by the identity of substrate bound in the WW domain. In particular, with pSPR bound in the WW domain, the mean binding energy of pSPS to the active site is the lowest (-20.30 kcal/mol) among all the tested conditions (Figure 4A, purple versus others), indicating a positive allosteric modulation by pSPR. In contrast, under pTPP-WW binding conditions, active-site substrate binding is weaker (-7.92 kcal/mol) than that under apo-WW conditions (See Figure 4A, orange versus black), indicating a slightly negative to neutral allosteric modulation by pTPP. These results suggest that the WW domain can indeed modulate substrate binding in the distal catalytic site, and this allosteric effect is dependent on the sequence of substrate bound to the WW domain. Whereas simply the presence of the WW domain enhances active-site binding, binding of specific substrates in the WW domain either amplifies or compromises this enhancement allosterically.



Page 16 of 26

Figure 4. Sequence and WW dependent allosteric regulation on pSPS binding to the active site. (A) Probability distributions of binding energies of pSPS to the catalytic domain under ΔWW (brown), apo-WW (black), pTPP-WW (orange), and pSPR-WW (purple) conditions. (B) Centerof-mass domain-domain distance distributions under pSPR-WW (purple), pTPP-WW (orange), and apo-WW(black) conditions. Inset, a schematic representation of distance measured between the catalytic and WW domain. (C,E) Contact probability difference of (C) pSPR-WW and (E) pTPPWW with respect to apo-WW mapped onto the structure of Pin1. Only probability difference above 20% are shown. Blue and red cylinders indicate contacts that are more and less formed, respectively, upon substrate binding in the WW domain. The width of cylinders is proportional to the magnitude of probability difference. (D) Projection of simulation trajectories under apo-WW


Page 17 of 26 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


(black), pSPR-WW (purple), and pTPP-WW (orange) conditions onto the subspace spanned by the top two PCs of the contact based PCA (See Methods). Inter-domain residue-residue contacts mediate the sequence dependent allosteric effect. The sequence dependent allosteric regulation is correlated with the distance between catalytic and WW domain. Distributions of the center-of-mass distance between domains, excluding the flexible linker region, were calculated under pSPR-WW, pTPP-WW, and apo-WW conditions with pSPS bound in the catalytic domain (Figure 4B). It shows that pSPR binding to the WW domain results in a shorter domain-domain distance (20.7 Å) than that under pTPP -WW (21.7 Å) and apo-WW conditions (21.9 Å). As expected, the apo-WW has the widest distribution. The movement of the WW domain with respect to the catalytic domain causes an alteration of domain-domain interactions, which may underlie the distinct allosteric effects produced by pSPR and pTPP. Intriguingly, a dynamical cross-correlation analysis shows consistent results with domain-domain distance distributions. Overall Cα-Cα correlations under pSPR-WW conditions are much weaker than those under apo-WW (Figure S1A & B), implying a more compact (and thus more rigid) conformational state upon pSPR binding to the WW domain. In contrast, correlations under pTPPWW are similar to those under apo-WW (Figure S1A & C), implying a similar level of molecular compactness between the two systems.

Principal component analysis (PCA) of residue-residue contacts reveals distinct conformational space populated by systems under different substrate-WW binding conditions. The same contact analysis methods described previously were employed.25,

41

Contact dynamics contain the

movement of both backbone and side-chain atoms, and thereby contact based PCA captures more complex protein motions than conventional Cartesian based PCA. It shows that despite some overlap, both pSPR-WW and pTPP-WW systems display a different conformational distribution 17 ACS Paragon Plus Environment


Page 18 of 26

from the apo-WW in the PC1-PC2 subspace (Figure 4D). Intriguingly, pSPR-WW and pTPP-WW separate from the apo-WW in opposite directions along PC1. These results indicate that substrate binding in the WW domain causes a population shift of conformations in a sequence dependent manner, and this shift ultimately affects the binding affinity of substrates in the active site.

Further analysis of residue-residue contacts identifies distinct, sequence dependent, contact dynamical changes in the domain interface. Probability difference of contact formation between the pSPR- or pTPP-WW and apo-WW was calculated (Figure 4C & E). Both comparisons show a set of contacts having either enhanced or reduced formation probability with respect to the apoWW. In particular, in the case of positive allosteric regulation (pSPR-WW), the dynamical changes of contact near the “hinge” region of domain-domain interface are primarily reduced, indicating that upon pSPR binding to WW contacts in the hinge region become more often broken (Figure 4C). In contrast, contacts in the hinge region are more often formed upon binding of pTPP to WW, along with contact breaking in the distal domain interface (Figure 4E). These observations explain the closer domain-domain distance under pSPR-WW binding (Figure 4B), and further indicate that the sequence dependent allosteric effect is mediated by distinct contact re-arrangements in the domain interface. Common inter-domain contacts undergoing significant, but opposite, changes between pSPR-WW and pTPP-WW conditions include Trp11-Arg142, Trp11-Glu145, and Pro37Arg142. Our results support recent experimental and computational studies showing that interface interactions and/or dynamics are involved in the allosteric communications in Pin1.21-24


Page 19 of 26 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


Independent tests with the pTPV substrate support the sequence dependent allosteric mechanism. The sequence and WW domain dependent allosteric regulation is supported by a further test on the binding of the pTPV substrate in the active site. Multiple microsecond simulations of fulllength Pin1 with pTPV bound in the catalytic domain and various alterations of the WW domain (ΔWW, apo-WW, pTPP-WW, and pSPR-WW) were performed. Similar WW and sequence dependency on allosteric regulation to the previous test was observed (Figure 5A). In particular, pSPR-WW

binding results in a higher binding affinity than that of pTPP-WW binding (-28.15 and

-8.91 kcal/mol, respectively; Figure 5A, purple versus orange). Also, the presence of the WW domain enhances the active-site binding, indicating an allosteric modulation by domain-domain interactions (-14.61 and -12.46 kcal/mol, respectively; Figure 5A, black versus brown). Although sequence dependent allosteric effects derived with the pTPV substrate are generally consistent with those derived with pSPS, detailed molecular mechanisms underlying allosteric communications are distinct between the two substrate-catalytic domain binding conditions. When pTPV binds

to the catalytic domain, both pSPR-WW and pTPP-WW only show a slight shift toward

a more compact conformation from apo-WW (21.2 and 21.3 Å, respectively, versus 21.8 Å domain-domain distance; Figure 5B). Also, overall changes of cross-correlations of pSPR-WW and pTPP-WW from apo-WW are mild (Figure S1D-F). Contact analysis suggests that only slight conformational changes occur upon binding of the positive modulator pSPR to the WW domain, including contact formations between domains and contact rearrangements within the catalytic domain (Figure 5C and D). In contrast, upon pTPP-WW binding, much larger conformational changes occur including a net domain-domain contact breakage and substantial contact rearrangements near the active site (Figure 5D and E). Note that pTPV is a stronger binder to the 19 ACS Paragon Plus Environment


Page 20 of 26

catalytic domain than pSPS (-14.61 versus -9.33 kcal/mol binding free energy). Our results suggest that distinct mechanisms can be utilized by a substrate bound in the WW domain to exert the same allosteric output in the catalytic domain and specific conformational changes during the allosteric transition are dependent on substrate identities bound in both the WW and catalytic domain.

Figure 5. Sequence and WW dependent allosteric regulation on pTPV binding to the active site. (A) Probability distributions of binding energies under ΔWW (brown), apo-WW (black), pTPP-WW

(orange), and pSPR-WW (purple) conditions. (B) Center-of-mass domain-domain

distance distributions under pSPR-WW (purple), pTPP-WW (orange), and apo-WW(black) conditions. Inset, a schematic representation of distance measured between the catalytic and WW domain. (C,E) Contact probability difference of (C) pSPR-WW and (E) pTPP-WW with respect to apo-WW mapped onto the structure of Pin1. Only probability difference above 20% are shown.


Page 21 of 26 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


Blue and red cylinders indicate contacts that are more and less formed, respectively, upon substrate binding in the WW domain. The width of cylinders is proportional to the magnitude of probability difference. (D) Projection of simulation trajectories under apo-WW (black), pSPR-WW (purple), and pTPP-WW (orange) conditions onto the subspace spanned by the top two PCs of the contact based PCA (See Methods).

CONCLUSIONS We have carry out a comprehensive in silico study of the catalytic and allosteric mechanisms in Pin1 using various substrates and Pin1 constructs. We show that substrates containing a proline at the +1 position after the pSer/Thr-Pro motif cannot be catalyzed due to a high energy barrier around the transition state and tend to bind preferentially to the WW domain than the catalytic domain, consistent with previous experimental studies.27-28 In addition, our results support the allosteric role of the WW domain and reveal a sequence dependent allosteric mechanism in Pin1. In particular, the substrate (pSPR) that was shown to have a positive allosteric effect in previous experimental and computational studies21-25 causes a remarkable increase in substrate binding affinities in the catalytic domain when it binds to the WW domain. In contrast, the substrate derived from tau (pTPP) only elicits slightly (near zero) negative allosteric effects; no allosteric effect was detected in experiments using the same peptide.27 Hence, our study unifies these otherwise contradicting previous studies. We further delineate detailed mechanisms underlying the sequence dependent allostery and find that residue-residue contacts at the domain-domain interface play a key role in mediating intra-protein allosteric communications. Our findings could facilitate future design of novel therapeutics targeting Pin1 – chemicals mimicking the specific geometry of 21 ACS Paragon Plus Environment


Page 22 of 26

pSer/Thr-Pro-Pro motif are suggested to be good candidates of Pin1 allosteric inhibitors that might preferentially bind to the WW domain and hamper substrate binding in the catalytic domain by either blocking positive substrate modulator binding in the WW domain or exerting a negative allosteric effect directly.

Supporting Information Figure S1. Dynamical cross-correlation matrices under distinct substrate binding conditions.

ACKNOWLEDGMENTS This research was supported by the National Science Foundation (MCB-1517617). We also acknowledge support from Georgia State University and the Georgia Research Alliance.


Page 23 of 26 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


REFERENCES 1. Gunasekaran, K.; Ma, B. Y.; Nussinov, R., Is allostery an intrinsic property of all dynamic proteins? Proteins 2004, 57, 433-443. 2. Tsai, C. J.; Nussinov, R., A unified view of "how allostery works". PLoS Comput. Biol. 2014, 10. 3. Liu, J.; Nussinov, R., Allostery: An overview of its history, concepts, methods, and applications. PLoS Comput Biol 2016, 12, e1004966. 4. Monod, J.; Wyman, J.; Changeux, J. P., On the nature of allosteric transitions: A plausible model. J. Mol. Biol. 1965, 12, 88-118. 5. Koshland, D. E., Jr.; Nemethy, G.; Filmer, D., Comparison of experimental binding data and theoretical models in proteins containing subunits. Biochemistry 1966, 5, 365-385. 6. Cooper, A.; Dryden, D. T., Allostery without conformational change. A plausible model. Eur. Biophys. J. 1984, 11, 103-109. 7. Raman, S., Systems approaches to understanding and designing allosteric proteins. Biochemistry 2018, 57, 376-382. 8. Nussinov, R.; Tsai, C. J., Allostery in disease and in drug discovery. Cell 2013, 153, 293305. 9. Wagner, J. R.; Lee, C. T.; Durrant, J. D.; Malmstrom, R. D.; Feher, V. A.; Amaro, R. E., Emerging computational methods for the rational discovery of allosteric drugs. Chem. Rev. 2016, 116, 6370-6390. 10. Lu, K. P.; Hanes, S. D.; Hunter, T., A human peptidyl-prolyl isomerase essential for regulation of mitosis. Nature 1996, 380, 544-547. 11. Yaffe, M. B.; Schutkowski, M.; Shen, M.; Zhou, X. Z.; Stukenberg, P. T.; Rahfeld, J. U.; Xu, J.; Kuang, J.; Kirschner, M. W.; Fischer, G., et al., Sequence-specific and phosphorylationdependent proline isomerization: A potential mitotic regulatory mechanism. Science 1997, 278, 1957-1960. 12. Lu, K. P.; Zhou, X. Z., The prolyl isomerase pin1: A pivotal new twist in phosphorylation signalling and disease. Nat. Rev. Mol. Cell Biol. 2007, 8, 904-916. 13. Lu, P. J.; Wulf, G.; Zhou, X. Z.; Davies, P.; Lu, K. P., The prolyl isomerase pin1 restores the function of alzheimer-associated phosphorylated tau protein. Nature 1999, 399, 784-788. 14. Liou, Y. C.; Sun, A.; Ryo, A.; Zhou, X. Z.; Yu, Z. X.; Huang, H. K.; Uchida, T.; Bronson, R.; Bing, G.; Li, X., et al., Role of the prolyl isomerase pin1 in protecting against agedependent neurodegeneration. Nature 2003, 424, 556-561. 15. Lu, K. P., Pinning down cell signaling, cancer and alzheimer's disease. Trends Biochem. Sci. 2004, 29, 200-209. 16. Driver, J. A.; Zhou, X. Z.; Lu, K. P., Pin1 dysregulation helps to explain the inverse association between cancer and alzheimer's disease. Biochim. Biophys. Acta 2015, 1850, 20692076. 17. Zhou, X. Z.; Lu, K. P., The isomerase pin1 controls numerous cancer-driving pathways and is a unique drug target. Nat. Rev. Cancer 2016, 16, 463-478. 18. Hidaka, M.; Kosaka, K.; Tsushima, S.; Uchida, C.; Takahashi, K.; Takahashi, N.; Tsubuki, M.; Hara, Y.; Uchida, T., Food polyphenols targeting peptidyl prolyl cis/trans isomerase pin1. Biochem. Biophys. Res. Commun. 2018, 499, 681-687. 19. Ranganathan, R.; Lu, K. P.; Hunter, T.; Noel, J. P., Structural and functional analysis of the mitotic rotamase pin1 suggests substrate recognition is phosphorylation dependent. Cell 1997, 89, 875-886. 23 ACS Paragon Plus Environment


Page 24 of 26

20. Lu, P. J.; Zhou, X. Z.; Shen, M.; Lu, K. P., Function of ww domains as phosphoserine- or phosphothreonine-binding modules. Science 1999, 283, 1325-1328. 21. Namanja, A. T.; Wang, X. J.; Xu, B.; Mercedes-Camacho, A. Y.; Wilson, K. A.; Etzkorn, F. A.; Peng, J. W., Stereospecific gating of functional motions in pin1. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 12289-12294. 22. Wang, X.; Mahoney, B. J.; Zhang, M.; Zintsmaster, J. S.; Peng, J. W., Negative regulation of peptidyl-prolyl isomerase activity by interdomain contact in human pin1. Structure 2015, 23, 2224-2233. 23. Guo, J.; Pang, X.; Zhou, H. X., Two pathways mediate interdomain allosteric regulation in pin1. Structure 2015, 23, 237-247. 24. Campitelli, P.; Guo, J.; Zhou, H. X.; Ozkan, S. B., Hinge-shift mechanism modulates allosteric regulations in human pin1. J. Phys. Chem. B 2018, acs.jpcb.7b11971. 25. Barman, A.; Hamelberg, D., Coupled dynamics and entropic contribution to the allosteric mechanism of pin1. J. Phys. Chem. B 2016, 120, 8405-8415. 26. Rogals, M. J.; Greenwood, A. I.; Kwon, J.; Lu, K. P.; Nicholson, L. K., Neighboring phosphoser-pro motifs in the undefined domain of irak1 impart bivalent advantage for pin1 binding. FEBS J. 2016, 283, 4528-4548. 27. Eichner, T.; Kutter, S.; Labeikovsky, W.; Buosi, V.; Kern, D., Molecular mechanism of pin1-tau recognition and catalysis. J. Mol. Biol. 2016, 428, 1760-1775. 28. Innes, B. T.; Bailey, M. L.; Brandl, C. J.; Shilton, B. H.; Litchfield, D. W., Non-catalytic participation of the pin1 peptidyl-prolyl isomerase domain in target binding. Front. Physiol. 2013, 4, 18. 29. Wisler, J. W.; Xiao, K.; Thomsen, A. R.; Lefkowitz, R. J., Recent developments in biased agonism. Curr. Opin. Cell Biol. 2014, 27, 18-24. 30. Humphrey, W.; Dalke, A.; Schulten, K., Vmd: Visual molecular dynamics. J. Mol. Graph. 1996, 14, 33-38, 27-38. 31. Case, D. A.; Cerutti, D. S.; Cheatham, T. E., III; Darden, T. A.; Duke, R. E.; Giese, T. J. G., H.; Goetz, A. W.; Greene, D.; Homeyer, N.; Izadi, S., et al., Amber 2017. University of California: San Francisco, 2017. 32. Hornak, V.; Abel, R.; Okur, A.; Strockbine, B.; Roitberg, A.; Simmerling, C., Comparison of multiple amber force fields and development of improved protein backbone parameters. Proteins 2006, 65, 712-725. 33. Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; Merz, K. M.; Ferguson, D. M.; Spellmeyer, D. C.; Fox, T.; Caldwell, J. W.; Kollman, P. A., A second generation force field for the simulation of proteins, nucleic acids, and organic molecules. J. Am. Chem. Soc. 1995, 117, 5179-5197. 34. Doshi, U.; Hamelberg, D., Reoptimization of the amber force field parameters for peptide bond (omega) torsions using accelerated molecular dynamics. J. Phys. Chem. B 2009, 113, 16590-16595. 35. Steinbrecher, T.; Latzer, J.; Case, D. A., Revised amber parameters for bioorganic phosphates. J. Chem. Theory Comput. 2012, 8, 4405-4412. 36. Jorgensen, W. L., Revised tips for simulations of liquid water and aqueous solutions. J. Chem. Phys. 1982, 77, 4156-4163. 37. Mark, P.; Nilsson, L., Structure and dynamics of the tip3p, spc, and spc/e water models at 298 k. J. Phys. Chem. A 2001, 105, 9954-9960.


Page 25 of 26 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


38. Essmann, U.; Perera, L.; Berkowitz, M. L.; Darden, T.; Lee, H.; Pedersen, L. G., A smooth particle mesh ewald method. J. Chem. Phys. 1995, 103, 8577-8593. 39. Kollman, P. A.; Massova, I.; Reyes, C.; Kuhn, B.; Huo, S.; Chong, L.; Lee, M.; Lee, T.; Duan, Y.; Wang, W., et al., Calculating structures and free energies of complex molecules: Combining molecular mechanics and continuum models. Acc. Chem. Res. 2000, 33, 889-897. 40. Tan, C.; Tan, Y. H.; Luo, R., Implicit nonpolar solvent models. J. Phys. Chem. B 2007, 111, 12263-12274. 41. Doshi, U.; Holliday, M. J.; Eisenmesser, E. Z.; Hamelberg, D., Dynamical network of residue-residue contacts reveals coupled allosteric effects in recognition, catalysis, and mutation. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 4735-4740. 42. Grant, B. J.; Rodrigues, A. P.; ElSawy, K. M.; McCammon, J. A.; Caves, L. S., Bio3d: An r package for the comparative analysis of protein structures. Bioinformatics 2006, 22, 26952696. 43. Skjærven, L.; Yao, X.-Q.; Scarabelli, G.; Grant, B. J., Integrating protein structural dynamics and evolutionary analysis with bio3d. BMC Bioinformatics 2014, 15, 399.



Page 26 of 26

TOC Graphic


Substrate Sequence Determines Catalytic ... - ACS Publications

Recommend Documents