Theses supervised by Prof. Dr. Türkan Haliloğlu

16 theses · Boğaziçi University

Master'sOpen AccessEN

Investigation of conformational fluctuations in protein dynamics: Torsional Gaussian network model

Elastic network models based on Normal Mode Analysis (NMA) are very useful in determining the conformational fluctuations that would affect the functions of proteins. Gaussian Network Model (GNM) is proved to be a very efficient method for this purpose. In recent years, coarse-grained elastic network models that use torsional degrees of freedom, have been introduced for aiming to predict the key residues that are responsible for the conformational changes of proteins. In the present thesis, a similar line of approach is applied to the simple one dimensional GNM in order to obtain the mean torsional fluctuations and the correlations between torsional fluctuations of residues by developing a novel method, Torsional Gaussian Network Model (tGNM). The predictions of this new approach is tested by comparing the fluctuations with the results obtained by Molecular Dynamics (MD) simulations for a set of proteins, ASC (Apoptosis-associated Speck-Like Protein Containing CARD), CypA (Cyclophilin A), and CAP (Catabolite Activator Protein). For further analysis of the model predictions, five pairs of protein structures with known open and closed conformations are studied for their conformational transitions. The mean square torsional fluctuations are calculated for the conformations of both states and compared with their torsional angle differences. Some of the peaks of the torsional angle differences are successfully observed in the mean square fluctuations; however, some additional peaks are observed in the model predictions. This suggests the latter fluctuations might be related with the conformations that could be observed on the transition pathway between the two states. The results in general provide that so called tGNM can be used for identifying residues that play an important role in the functional conformational transition of proteins.

Harun Ferit Özbakır
Boğaziçi University · Fen Bilimleri Enstitüsü
2012
00
Master'sOpen AccessEN

Motor proteinlerin dinamik allosterisinin incelenmesi

Allostery is a key concept in regulation of protein function. The communication pathways which are equivalent of the allosteric signaling or energy propagation pathways are the most significant aspect of allosteric mechanism. A Brownian dynamics simulation approach with force distribution analysis is proposed to determine key residues that have a role in allosteric signal transmission. Along, another stochastic approach -Monte Carlo path generation- is also implemented to define likely allosteric pathways through generating an ensemble of maximum probability paths. The force distribution analysis is tested on methionine repressor and purine repressor. The Monte Carlo path generation method is applied on various proteins such as PDZ domain, bovine rhodopsin and three different structures of myosin. The predicted residues that have a significant effect on the allosteric signaling are in agreement with the previous studies. As a case study, the main focus is Kinesin motor protein, which is a key element of cargo transport in intracellular environment. The ATP binding and the motility are directly related and communicated based on allosteric signaling in this molecule. The analyses based on both methodologies are aimed to define allosteric pathways and key sites in signaling. The predicted regions by both methods are successful in determination of the residues that plays a key role in signaling previously determined by other studies. Based on the graph centrality parameters and force distribution analysis, a novel allosteric site proposed (Cys 176). For the validation of algorithms as well as the specific allosteric sites therein, pulling experiments with atomic force microscopy are considered and performed experiments are simulated with some design plans proposed. The change in the rupture force by application of external force from different sites may provide information about the validity of the computational approaches and validate importance of the predicted sites. As a further study, three predicted sites will be tested via different functionalization methods. Overall, the proposed methodologies have potential to identify residues of importance in signaling event.

Cihan Kaya
Boğaziçi University · Institute of Graduate Studies in Science
2013
00
Master'sOpen AccessEN

Dynamic nature of H1N1 influenza virus hemagglutinin

Dynamics of antigenic and host receptor binding behavior of a virus binding protein can be altered via mutations that lead to escape from immune recognition. The main focus of this study is to investigate the dynamic behavior of Influenza A virus Hemagglutinin (HA) of the 2009 human pandemic H1N1 in comparison to other H1N1 strains, by Molecular Dynamics (MD) simulations. HA is the host receptor binding protein that has a key role in virus-host interaction. The dynamic behavior of HA of two Influenza A (H1N1) variants and their mutants at residue position 149 were studied comparatively between the structures to search for the effect of this position between the two variants (K149 and R149) and the mutations (K149R and R149K) with the emphasis on the Receptor Binding Domain (RBD) of HA. The residue 149 is chosen considering its importance due to being a discriminating position between classical (R149) and 2009 pandemic (K149) swine viruses, although it is directly associated with neither antigenic nor sialic binding sites. Residue 149 is also identified as one of the hinge positions in both of the variants within their wild and mutant types, implying an allosteric dynamic affect. The position 149 is found to be highly affecting the dynamics of RBD at 130 loop and 220 loop that includes the receptor binding site (RBS) residues 134-138 and 221-228 as well as antigenic binding sites (Sa and Ca), which is particularly strong for K149. The structures of K149 and R149K displaying more restricted mobility lead to a higher cooperativity and efficient global HA behavior. Overall, the results imply that allosteric dynamic effect of 149 may affect virulence as a result of altering dynamic behavior in RBD of HA.

Zeynep Erge Akbaş
Boğaziçi University · Institute of Graduate Studies in Science
2014
00
Master'sOpen AccessEN

A mechanistic view of cullin neddylation: Allosteric conformational control of E3-ring ligases

Cullin-RING E3 ligases (CRLs), which constitute a superfamily of RING type E3 ligases, regulate ~20% of protein degradation by the ubiquitin-proteasome system in the cells. Cullin neddylation activates CRLs by urging a conformational transition in Cullin C-terminal domain and exposes the Rbx1 RING E3 ligase to facilitate ubiquitination. CRLs are over-activated by Cullin neddylation in myriad of human cancers; thereby the inhibition mechanism of Cullin neddylation is of significant interest for generating novel cancer therapies. To explore how the signals of Cullin neddylation are transmitted through the CRLs to control the Rbx1 conformations, we performed molecular dynamics (MD) simulations combined with Gaussian Network Model (GNM) analysis on the closed-unbound and open-bound to Nedd8 states of Cul5-Rbx1 E3 ligases. Our results display multiple Rbx1 RING domain conformations with neddylation of Cul5 where the rigid dynamics of the unbound-to-Nedd8 state CRLs is abrogated. The Rbx1 RING domain conformations of the neddylated CRLs, which are in agreement with the proposed neddylated state Rbx1 conformations for effective ubiquitin transfer, arise as a result of coupled dynamics between Nedd8 and Rbx1. A dynamic hinge plane of Cul5 hinges at R495-M496, N704-E705 and Rbx1 hinges at D51-L52, T69-V70 coordinates allosteric communication between Nedd8 and Rbx1 on the CRL structure along an H-bond between Cul5 R538, R569 and Rbx1 E67 that restricts the Rbx1 conformational space. Nedd8 predisposes Rbx1 together with Cul5 for an effective ubiquitin transfer by a plausible mechanism explaining the allosteric conformational control of Rbx1. We propose that selective interruptions in the neddylation pathway to disrupt the dynamic hinge plane coordination could shed a light on the allosteric drug design studies on Cullin neddylation inhibition.

Melis Yıldırım
Boğaziçi University · Fen Bilimleri Enstitüsü
2014
00
Master'sOpen AccessEN

Protein komplekslerinin kollektif dinamikleri: Puanlama ve tahmin

Docking algorithms are widely used in determination of protein complex structures. In order to determine which structure is the most similar to native one, ranking algorithms are used in docking studies. Each year both docking and ranking algorithms are tested in a community-wide experiment called CAPRI (Critical Assessment for Prediction of Interactions). With the purpose of creating a ranking scheme, native protein complex structures were characterized according to their dynamic entities using Gaussian and Anistropic Network Models (GNM and ANM, respectively). Dynamic properties such as mean square residue fluctuations (MSF), positions of hinge residues, and the extent of the correlation of the direction of structural motion in unbound and bound states were investigated. Results display that there is a drop in MSF of interface residues and increase in MSF of non-interface residues upon binding. Also, there is a shift in ANM modes upon binding, which tells that dynamic characteristics of a binding partner in unbound state is contained in dynamic characteristics a of complex. In addition, existence of hinge residues at interface is significant for binding. Degree of collectivity, a property which is calculated by ANM modes, increases upon binding. Using these properties, two distinct software were developed. CLAMshell, a dynamics-based clustering tool which is able to cluster native-like structures based on the similarity of the dynamics of protein chains in unbound and bound states; HiPlane, a tool to draw hinge planes and obtain plane normals for dividing the structure into dynamic domains described by global modes for characterizing orientations of binding partners. With the help of the software that were developed, dynamic properties of predicted models from former CAPRI targets and native structures were also investigated in order to build a ranking scheme.

Serdar Özsezen
Boğaziçi University · Fen Bilimleri Enstitüsü
2015
00
DoctorateOpen AccessEN

An integrated computational and experimental approach to allosteric control mechanism of biomolecular processes

Proteinler esnek ve dinamik yapılardır ve işlevsel yapısal değişimler gösterirler. Bu tezin ana motivasyonu, konformasyonel geçişlerin ve moleküllerin bağlanma süreçlerinin alosterik kontrolünü anlamaya yönelik hesaplamalı ve deneysel entegre bir yöntem geliştirmektir. Oncelikle, kararlı yapılar arasındaki geçişler ve serbest enerji yüzeylerini belirlemeye yönelik, Kolektif Modlar Bayas Değişimli Metadinamik (CM-BexMetaD) olarak adlandırılan özgün bir yöntem geliştirilmiştir. Adenilat kinaz enziminin birçok ara yapısı belirlenmiş ve menteşe bölgelerinin alosterik etkisi gösterilmiştir. Ikinci bölümde, ENM, moleküler dinamik (MD), in-vitro görüntüleme ve Atomik Kuvvet Mikroskopu-Dinamik Kuvvet Spektroskopisi (AFM-DFS) yöntemleri ile proteinlerin bağlanma mekanizmalarının alosterik kontrolu üzerine çalışılmıştır. SKEMPI veritabanindaki mutasyonlar ile menteşe bölgeleri arasındaki kuvvetli ilişki, geniş çaplı istatistiksel bir analiz ile gosterilmistir. In-silico ve in-vitro deneyler ile PYD ve ASC proteinlerindeki menteşe bölgelerindeki mutasyonlarin bağlanma davranışı üzerindeki etkisi çalışılmıştır. ASC zerreciklerinin en az iki seviyeli organize yapılar olduğu ve menteşe bölgeleri aracılığı ile kontrol edildiği belirlenmiştir. Bunlara ek olarak, AFM-DFS yöntemini kapsayan hibrit bir yaklaşım geliştirilerek, protein-protein bağlanma mekanizmalarinin menteşe bölgeleri aracılığı ile alosterik olarak kontrol edildiği, kinezin-αβ-tübülin ve Rac1-PAK1 etkileşimleri üzerinden ortaya konulmuştur. Alosterik haberleşme ağındaki değişimin mekanik izahı ise MD simülasyonları ile belirlenmiştir. Elde edilen sonuçlar ile kinezin-αβ-tübülin ve Rac1-PAK1 kompleks yapılarının alternatif ayrılma patikalarına sahip oldukları ve menteşe bölgelerinin bağlanma davranışlarını alosterik olarak kontrol eden anahtar bölgeler olduğu ortaya konulmuştur.

Fidan Sümbül
Boğaziçi University · Fen Bilimleri Enstitüsü
2016
00
DoctorateOpen AccessEN

Computational and experimental investigations into the mechanism of functional motion in biomolecular systems

The seemingly random fluctuations that proteins exhibit in their native states actually harbor major contributions from the global/cooperative motions they undergo during functional transitions. Computationally efficient, robust approaches like elastic network models (ENM) that exploit the intrinsic dynamics of proteins are of great value in building up hybrid approaches bridging experimental and computational studies and extending each application's range. ENM global modes show functional significance and compares well with functional conformational changes. The purpose of this study is to build upon currently available computational techniques that generate transition pathways connecting end states. In this thesis, a practical hybrid approach, anisotropic network model (ANM) guided Langevin dynamics (LD) simulations (ANM-LD) has been developed, where the dominant ANM low-frequency modes are utilized to drive LD simulations. In ANM-LD, the initial conformation is moved along the selected ANM mode at each cycle, taking advantage of the evaluation of the interactions and energetics of the system via short cycles of all-atom implicit LD simulations. A detailed assessment of the method, in terms of creating physically meaningful pathways and approaching the final state, are carried out for a set of proteins. As exemplary cases, the analysis and results for adenylate kinase and maltose transporter is presented. ANM-LD was also applied to heat shock protein 90 (Hsp90) in apo, ATP-bound and Geldanamycin (GDM)-bound states, the latter combined with AFM experiments, to study the effect of temperature on the Hsp90-GDM binding dynamics. Lastly, conventional molecular dynamics simulations, ENM and ANM-LD are utilized to investigate CRP dynamics.

Ayşe Burcu Fas
Boğaziçi University · Fen Bilimleri Enstitüsü
2016
00
DoctorateOpen AccessEN

Unveiling intrinsic dynamics in protein-protein interactions

Binding, fundamental in protein-protein interactions, serves as a gateway to biological function. This PhD thesis explores binding/unbinding phenomena, conformational changes and allosteric interactions, identifying dynamic determinants related to protein function. Elastic Network Models (ENM) are used to analyze protein dynamics in both complex and unbound states, with a focus on conformational changes. The slowest dynamic modes, with hinge sites as key mechanistic points, drive binding/unbinding behavior. Disease-related mutations at these sites reinforce their functional significance. ANM-LD (Anisotropic Network Model-guided Langevin Dynamics) simulations uncover underlying dynamic determinants, revealing specific intermediate steps and latent allosteric pathways, providing ordered sequence of events that elucidate principles controlling functional behavior. We explored its potential as ANM-LD Path webserver with targeted simulations on the GroEL chaperone describing structural changes upon GroES binding alongside Atomic Force Microscopy (AFM) topographic imaging experiments. While being successful on large systems, targeted ANM-LD simulations also capture experimentally observed height increase in the GroEL upper ring while identifying the key residues and correlated domains. These all led to the development of ANM-LD Bind, a platform to tinker and explore the dynamic nature of targeted and untargeted binding/unbinding events, exposing intermediate conformations and allosteric interactions. Finally, intrinsic dynamic behavior of the kinesin-tubulin complex is studied via computational approaches to complement AFM pulling experiments, where disease-related hinge site mutations alter unbinding behavior, increasing binding affinity for D72N and decreasing it for S175A and N332A. These findings collectively enhance our understanding of the dynamic mechanisms governing protein interactions, providing insights for developing therapeutic strategies and designing novel biomolecular functions.

Zeynep Erge Buz
Boğaziçi University · Fen Bilimleri Enstitüsü
2024
00
Master'sOpen AccessEN

Structural dynamics and allosteric mechanisms of gamma-aminobutyric type B (gaba_b) receptor

Gamma-Aminobütirik Asit Tip B (GABA_B) reseptörü, Sınıf C G protein–kenetli reseptör (GPCR) ailesine ait olup, GB1 ve GB2 alt birimlerinin heterodimerik iş birliğine dayanan karmaşık bir allosterik mekanizma aracılığıyla merkezi sinir sisteminde yavaş sinaptik inhibisyonu düzenler. Son yıllarda yapılan yapısal çalışmalar bu reseptörün anlaşılmasını ilerletmiş olsa da, aktivasyonun dinamik temeli hâlâ tam olarak açıklığa kavuşmamıştır. Bu tez, GABA_B reseptörünün aktivasyon süreci ve allosterik ağını yapısal dinamikler açısından karakterize etmeyi amaçlamaktadır. Bu doğrultuda Gaussian Ağ Modeli–Transfer Entropisi (GNM-TE), ANM eş-titreşim modu analizi ve Anizotropik Ağ Modeli–Langevin Dinamiği (ANM-LD) yöntemleri ile kapsamlı bir hesaplamalı yaklaşım uygulanmıştır. Bulgular, alanlar arası uzun menzilli iş birliğinin aktivasyon için belirleyici olduğunu ve izole alt birimlerin tam aktivasyonu sağlayacak esnekliğe sahip olmadığını ortaya koymuştur. GNM-TE ile entropi akış yolları belirlenmiş, işlevsel kalıntılar ve patojenik mutasyonlar küresel allosterik koordinasyondaki rolleriyle öne çıkmıştır. Bu çalışma, GABA_B dinamiklerini daha iyi anlamaya katkı sağlarken, Sınıf C GPCR'ler için aktarılabilir bir model sunmaktadır.

Nida Yaren Yılmaz
Boğaziçi University · Institute of Graduate Studies in Science
2025
00
Master'sOpen AccessEN

Ligand docking to proteins undergoing large conformational changes

Conformational transitions and ligand binding to the enzyme adenylate kinase were investigated by computational docking. Autodock was used to investigate the dynamics of 92 atomistic conformers, which have been previously obtained by a conformational sampling algorithm- ClustENM. First the internal geometry of the atomistic ClustENM conformers were validated by the MolProbity server. Four different ligands,namely AP5, ATP, ADP and AMP, were docked on each conformer to determine the favorable docking poses in terms of low energies and ligand RMSD values. Docking of ATP to the side of the LID domain gave the best results among all ligands, which is accompanied by LID closure to various extents. Minimum RMSD is obtained where ATP ligand docked to LID domain with 0.63 Å. For multiple conformers, successful poses were found for docking of any ligand to the LID side, whereas dockings to the side of the NMP domain did not yield any successful poses for any of the ligands. Results are mostly in compliance with GOLD procedure especially for the ATP case. To analyze the changes in dynamics for the protein-ATP poses, Dynomics web server was used. Moreover, key protein-ligand interactions were determined by Ligplot, which pointed to the interactions around the phosphate groups of the ligand in the closed crystal structure. Intermediate structures with half closed LID were successful for docking of the ligands, indicating a combination of induced fit and conformational search algorithms.

Damla Üçkan
Boğaziçi University · Institute of Graduate Studies in Science
2019
00
Master'sOpen AccessEN

Protein dynamics in deleterious and compensatory mutations

Mutations are associated with many diseases (cancer, Alzheimer's disease etc.). These disease-causing mutations are called deleterious mutations and their effects can be corrected by other mutations, i.e. compensatory mutations. Understanding the underlying dynamics of deleterious and compensatory mutations is of high importance for the treatment of diseases and drug design. To that end, the relationship between evolutionary conserved/reused segments and dynamic domains as well as the dynamic determinants of deleterious and compensatory mutations are investigated in this work. Themes; reused segments (~35-200 amino acids) among proteins with high sequence similarity, are correlated with the dynamic domains unveiled with Gaussian Network Model (GNM) analysis. The correlation between themes and dynamic domains evaluated with Adjusted Mutual Information and Standardized Mutual Information measures is found to be statistically significant. GNM based perturbation analysis revealed that the highest response to perturbations occur at the terminal points of themes. In order to investigate the dynamic properties of deleterious mutations, a dataset consisting of proteins that have been the subject to deep sequencing studies is created. In order to mimic the effect of mutation, a perturbation is placed in the GNM algorithm. The effect of this perturbation on the dynamics of the structures is examined according to the changes in the total fluctuation profiles and eigenvalues. Residues with high mutation sensitivity are found to be the residues that cause distinct change in protein dynamics upon perturbation and this relationship is statistically significant. For compensatory mutation studies, deleterious mutations of tumor suppressor protein p53 and their compensatory mutations are used. The results revealed that deleterious and compensatory mutations are correlated in slow modes of motion and demonstrate the importance of coevolution, hinge points and allosteric interaction in slow modes of motion for these mutations. Furthermore, the dynamic information about deleterious and compensatory mutations reported in this study will be a guide for further studies on the prediction of deleterious and compensatory mutations.

Yiğit Kutlu
Boğaziçi University · Institute of Graduate Studies in Science
2020
00
Master'sOpen AccessEN

How intrinsic dynamics and allostery in G-protein coupled receptors (GPCR) define functional selectivity?

G-protein Coupled Receptors (GPCRs) are the largest family of the of sensory proteins which play an important role in many diseases related to signaling pathways. Thus, understanding GPCR activation mechanism is a corner stone in drug development. However, the transition pathways between inactive/active states remains not fully explored. In this thesis, a new approach, ANM-LD methodology, that combines the Anisotropic Network Model (ANM) and Langevin Dynamics (LD) is performed to decipher the transition pathways in the activation process and determine the key dynamic factors underlying the conformational changes of GPCR family. Towards this goal, three different receptor families (M2R, ß2AR, and NTSR1) are studied. The predicted transition pathways reveal that the intrinsic dynamic modes of the inactive state enable the conformational changes essential for the activation. As a result, the distances between TM3-TM6 and TM3-TM7 helices and the path of some functionally important motifs describe the major pathway that show commonality and differences in all three GPCRs. The agonist binding pocket interactions differ markedly for different subspecies, which are reflected in the selected dynamic modes of motion. When M2R is compared with NTSR1 and ß2AR, the weaker packing interactions result in more flexible segments in the connector of NTSR1 and ß2AR resulting in the TM3-TM6 distance to decay slower. Although a common agonist binding site does not exist, the conserved DRY, and NPxxY motifs are the common drivers of the GP binding site.The hinges of the most dominant dynamic modes coincide functionally important residues, constitute the dynamic infrastructure of allosteric interactions and enables the communication between the extracellular and intracellular sides of the structure.

Özge Duman
Boğaziçi University · Institute of Graduate Studies in Science
2020
00
Master'sOpen AccessEN

Subsets of slow dynamic modes reveal global information sources as allosteric sites

Allostery is a crucial biological regulation mechanism, and dynamic information flow offers a framework to characterize allosteric interactions in causal links. Here, using a novel application of the Transfer Entropy (TE) calculations based on the Gaussian Network Model (GNM), it has been demonstrated how the dissection of dynamic information into subsets of slow dynamic modes reveals various layers of multi-directional allosteric pathways that are intrinsic in a particular protein structure. The degree of collectivity (Col) in the information transfer of residues with their TE values (TECol score) in these subsets of slow modes identifies particular residues as potent effectors and global information sources having a strong dynamic capacity to collectively disseminate information to other residues in the protein structure. These information source residues are linked to known active and allosteric sites, as demonstrated by aspartate transcarbamoylase (ATCase), Na+/K+-adenosine triphosphatase (Na+/K+-ATPase), and human transient receptor potential melastatin 2 (TRPM2), along with a dataset of 20 proteins. These specific residues provide feasible binding sites for structure-based rational drug design since they together affect/control others and direct pathways of allosteric communication.

Information transferCollectivisimCausality+1
Bengi Altıntel
Boğaziçi University · Institute of Graduate Studies in Science
2022
00
Master'sOpen AccessEN

Structural dynamics of ABC transporters

The ATP-binding cassette transporters (ABC transporters) translocate substrates across membranes by ATP binding and hydrolysis. To understand the complex structural dynamics, transporters were investigated on three case studies: heterodimeric multidrug exporter TmrAB, cystic fibrosis transmembrane conductance regulator (CFTR), and manganese ion importer PsaBC. Anisotropic Network Model-Langevin Dynamics (ANM-LD) simulations were performed between four different conformations of TmrAB. The analysis for each transition revealed the allosteric couplings between the ATP binding in nucleotide binding domains (NBD) and extracellular gates on transmembrane domains (TMD). Transfer Entropy (TE) calculations showed that TMDs and NDBs of TmrAB become drivers at different stages during the translocation cycle. The ANM-LD simulations of dephosphorylated to phosphorylated CFTR transition revealed that the initial movements in the NBDs are followed by the movement of the gating areas in TM. The causal relationships between these movements were supported by TE calculations. CFTR potentiator drug Ivacaftor has been shown to have a global effect, mimicking the effects of ATP binding and gating residues. Corrector drug Lumacaftor has a local effect on TM1, 2, 3, and 6. The dynamics of PsaBC was characterized using TE calculations, and testable predictions were made. Four possible gating residues were identified along with an allosteric pocket on TM8. Finally, TE calculations of many ABC transporters showed the differences and similarities between different types. Type I transporters were shown to be controlled by TMDs in the inward facing (IF) conformation, and NBDs and substrate binding protein (SBP) in the outward facing (OF) conformation. Type II transporters were always driven by NBDs, regardless of conformation. Type IV transporters showed more variability, but for most, TMDs are driving for IF and NBDs are driving for OF conformations. Dynamic characterization methods show potential in classifying ABC transporters.

CFTRCarrier proteinsStructural dynamics
Akarun Ayça Ersoy
Boğaziçi University · Fen Bilimleri Enstitüsü
2022
00
DoctorateOpen AccessEN

Dynamics, allostery, and signalling bias in g protein–coupled receptors

G protein-coupled receptors (GPCRs) are central to cellular signal transduction and represent the most intensively targeted protein family in pharmacology. This thesis presents a comprehensive computational analysis of Class A GPCR dynamics, with a focus on activation mechanisms, biased signaling, and allosteric regulation. Conventional MD simulations struggle to capture large-scale, long-timescale motions. To address these limitations, we employed the Anisotropic Network Model–Langevin Dynamics (ANM-LD) framework to model 30 conformational transitions across 25 GPCRs, including dual activation pathways for five receptors. These simulations generated dynamic ensembles of intermediate states, validated using hallmark collective variables such as TM6 tilt and TM7 shift. Covibrational mode analysis revealed distinct dynamical signatures for G protein– and ß-arrestin–bound states, highlighting pathway-specific encoding of collective motions. Mechanistic insights into biased signaling revealed that key hinge residues, including the DRY and NPxxY motifs, were identified as critical mediators of pathway-selective transitions. GNM-based transfer entropy (GNM-TE) analysis demonstrated strong receptor-driven information flow from the orthosteric binding site in both signaling states. A newly developed metric, the Differential Coupling Score, captured structural segments involved in signaling bias. Additionally, ensemble-level comparisons across 300 structures revealed conserved allosteric scaffolds and dynamic diversity associated with ligand type and signaling partner. In particular, Helix 8 and TM7-IC emerged as flexible hubs modulating pathway-specific communication. These findings provide a unified and generalizable framework for decoding GPCR dynamics and guiding the design of biased ligands and allosteric modulators.

G proteinsCellular pathwaysMembrane proteins+1
Özge Duman
Boğaziçi University · Fen Bilimleri Enstitüsü
2025
00
DoctorateOpen AccessEN

Proteın dinamikleri ile evrimsel bağlantıların deşifre edilmesi

Proteins are inherently dynamic, and their functional mechanisms often rely on collective internal motions. While the relationship between structure and dynamics has been widely studied, the connection between intrinsic dynamics and evolutionarily conserved sequence motifs remains poorly understood. This thesis addresses this gap by investigating the dynamic roles of themes—short conserved sequence segments shared across diverse proteins—through elastic network modeling and information-theoretic analysis. Using the Gaussian Network Model (GNM), dynamic elements (DEs) were defined as contiguous, co-fluctuating segments derived from low-frequency collective modes. To assess their correspondence with themes, mutual information (MI) was calculated between DEs and theme-based partitions across a curated dataset of 150 ECOD domains, with statistical validation performed against randomized controls. The results indicate that many themes align significantly with DEs, suggesting that reused segments are embedded in functional dynamic units. In a focused case study, previously identified adenine-binding themes were analyzed using a GNM-based Transfer Entropy (GNM-TE) method. The analysis revealed that these motifs frequently act as dynamic communication hubs, implicating them in long-range allosteric signaling. Building on this, the thesis classifies themes according to their information transfer roles and applies statistical analysis to evaluate enrichment patterns among binding themes. Together, these studies offer the first systematic framework for linking evolutionarily reused sequence segments with intrinsic dynamic architectures and communication properties. By introducing the concept of dynamic elements and integrating them with directional information flow, this work advances our understanding of protein evolution and function from a structure–dynamics perspective.

Biological evolutionMolecular dynamicComputational biology
Yiğit Kutlu
Boğaziçi University · Fen Bilimleri Enstitüsü
2025
10

Other supervisors