Titin kinase ubiquitination aligns autophagy receptors with mechanical signals in the sarcomere.

Striated muscle undergoes remodelling in response to mechanical and physiological stress, but little is known about the integration of such varied signals in the myofibril. The interaction of the elastic kinase region from sarcomeric titin (A168-M1) with the autophagy receptors Nbr1/p62 and MuRF E3 ubiquitin ligases is well suited to link mechanosensing with the trophic response of the myofibril. To investigate the mechanisms of signal cross-talk at this titin node, we elucidated its 3D structure, analysed its response to stretch using steered molecular dynamics simulations and explored its functional relation to MuRF1 and Nbr1/p62 using cellular assays. We found that MuRF1-mediated ubiquitination of titin kinase promotes its scaffolding of Nbr1/p62 and that the process can be dynamically down-regulated by the mechanical unfolding of a linker sequence joining titin kinase with the MuRF1 receptor site in titin. We propose that titin ubiquitination is sensitive to the mechanical state of the sarcomere, the regulation of sarcomere targeting by Nbr1/p62 being a functional outcome. We conclude that MuRF1/Titin Kinase/Nbr1/p62 constitutes a distinct assembly that predictably promotes sarcomere breakdown in inactive muscle.

Reconstruction of mechanical unfolding and refolding pathways of proteins with atomic force spectroscopy and computer simulations.

Most proteins in proteomes are large, typically consist of more than one domain and are structurally complex. This often makes studying their mechanical unfolding pathways challenging. Proteins composed of tandem repeat domains are a subgroup of multi-domain proteins that, when stretched, display a saw-tooth pattern in their mechanical unfolding force extension profiles due to their repetitive structure. However, the assignment of force peaks to specific repeats undergoing mechanical unraveling is complicated because all repeats are similar and they interact with their neighbors and form a contiguous tertiary structure. Here, we describe in detail a combination of experimental and computational single-molecule force spectroscopy methods that proved useful for examining the mechanical unfolding and refolding pathways of ankyrin repeat proteins. Specifically, we explain and delineate the use of atomic force microscope-based single molecule force spectroscopy (SMFS) to record the mechanical unfolding behavior of ankyrin repeat proteins and capture their unusually strong refolding propensity that is responsible for generating impressive refolding force peaks. We also describe Coarse Grain Steered Molecular Dynamic (CG-SMD) simulations which complement the experimental observations and provide insights in understanding the unfolding and refolding of these proteins. In addition, we advocate the use of novel coiled-coils-based mechanical polypeptide probes which we developed to demonstrate the vectorial character of folding and refolding of these repeat proteins. The combination of AFM-based SMFS on native and CC-equipped proteins with CG-SMD simulations is powerful not only for ankyrin repeat polypeptides, but also for other repeat proteins and more generally to various multidomain, non-repetitive proteins with complex topologies.

Exploring the structure characteristics and major channels of cytochrome P450 2A6, 2A13, and 2E1 with pilocarpine.

The majority of cytochromes P450 play a critical role in metabolism of endogenous and exogenous substrates, some of its products are carcinogens. Therefore, inhibition of P450 enzymes activity can promote the detoxification and elimination of chemical carcinogens. In this study, molecular dynamics (MD) simulations and adaptive steered molecular dynamics (ASMD) simulations were performed to explore the structure features and channel dynamics of three P450 isoforms 2A6, 2A13, and 2E1 bound with the common inhibitor pilocarpine. The binding free energy results combined with the PMF calculations give a reasonable ranking of binding affinity, which are consistent with the experimental data. Our results uncover how a sequence divergence of different CYP2 enzymes causes individual variations in major channel selections. On the basis of channel bottleneck and energy decomposition analysis, we propose a gating mechanism of their respective major channels in three enzymes, which may be attributed to a reversal of Phe209 in CYP2A6/2A13, as well as the rotation of Phe116 and Phe298 in CYP2E1. The hydrophobic residues not only make strong hydrophobic interactions with inhibitor, but also act as gatekeeper to regulate the opening of channel. The present study provides important insights into the structure-function relationships of three cytochrome P450s and the molecular basis for development of potent inhibitors.

Designing collagens to shed light on the multi-scale structure-function mapping of matrix disorders.

Collagens are the most abundant structural proteins in the extracellular matrix of animals and play crucial roles in maintaining the structural integrity and mechanical properties of tissues and organs while mediating important biological processes. Fibrillar collagens have a unique triple helix structure with a characteristic repeating sequence of (Gly-X-Y)n. Variations within the repetitive sequence can cause misfolding of the triple helix, resulting in heritable connective tissue disorders. The most common variations are single-point missense mutations that lead to the substitution of a glycine residue with a bulkier amino acid (Gly â†’ X). In this review, we will first discuss the importance of collagen's triple helix structure and how single Gly substitutions can impact its folding, structure, secretion, assembly into higher-order structures, and biological functions. We will review the role of "designer collagens," i.e., synthetic collagen-mimetic peptides and recombinant bacterial collagen as model systems to include Gly â†’ X substitutions observed in collagen disorders and investigate their impact on structure and function utilizing in vitro studies. Lastly, we will explore how computational modeling of collagen peptides, especially molecular and steered molecular dynamics, has been instrumental in probing the effects of Gly substitutions on structure, receptor binding, and mechanical stability across multiple length scales.

Treatment of flexibility of protein backbone in simulations of protein-ligand interactions using steered molecular dynamics.

To ensure that an external force can break the interaction between a protein and a ligand, the steered molecular dynamics simulation requires a harmonic restrained potential applied to the protein backbone. A usual practice is that all or a certain number of protein's heavy atoms or Cα atoms are fixed, being restrained by a small force. This present study reveals that while fixing both either all heavy atoms and or all Cα atoms is not a good approach, while fixing a too small number of few atoms sometimes cannot prevent the protein from rotating under the influence of the bulk water layer, and the pulled molecule may smack into the wall of the active site. We found that restraining the Cα atoms under certain conditions is more relevant. Thus, we would propose an alternative solution in which only the Cα atoms of the protein at a distance larger than 1.2 nm from the ligand are restrained. A more flexible, but not too flexible, protein will be expected to lead to a more natural release of the ligand.

Evaluation of the efficacy of marine natural products against PARP-1/2 proteins in high-grade serous ovarian cancer: insights into MD and SMD simulations.

High-grade serous ovarian cancer (HGSOC) is the most malignant and ubiquitous phenotype of epithelial ovarian cancer. Originating in the fallopian tubes and rapidly spreading to the ovaries, this highly heterogeneous disease is a result of serous tubal intraepithelial carcinoma. The proteins known as poly(ADP-ribose) polymerase (PARP) aid in the development of HGSOC by repairing the cancer cells that proliferate and spread metastatically. By using molecular docking to screen 1100 marine natural products (MNPs) from different marine environments against PARP-1/2 proteins, prominent PARP inhibitors (PARPi) were identified. Four compounds, alisiaquinone A, alisiaquinone C, ascomindone D and (+)-zampanolide referred to as MNP-1, MNP-2, MNP-3 and MNP-4, respectively, were chosen based on their binding affinity towards PARP-1/2 proteins, and their bioavailability and drug-like qualities were accessed using ADMET analysis. To investigate the structural stability and dynamics of these complexes, molecular dynamics simulations were performed for 200 ns. These results were compared with the complexes of olaparib (OLA), a PARPi that has been approved by the FDA for the treatment of advanced ovarian cancer. We determined that MNP-4 exhibited stronger binding energies with PARP-1/2 proteins than OLA by using MM/PBSA calculations. Hotspot residues from PARP-1 (E883, M890, Y896, D899 and Y907) and PARP-2 (Y449, F450, A451, S457 and Y460) showed strong interactions with the compounds. To comprehend the unbinding mechanism of MNP-4 complexed with PARP-1/2, steered molecular dynamics (SMD) simulations were performed. We concluded from the free energy landscape (FEL) map that PARP-1/2 are well-stabilised when the compound MNP-4 is bound rather than being pulled away from its binding pockets. This finding provides significant evidence regarding PARPi, which could potentially be employed in the therapeutic treatment of HGSOC.Communicated by Ramaswamy H. Sarma.

An insight from computational approach to explore novel, high-affinity phosphodiesterase 10A inhibitors for neurological disorders.

The enzyme Phosphodiesterase 10A (PDE10A) plays a regulatory role in the cAMP/protein kinase A (PKA) signaling pathway by means of hydrolyzing cAMP and cGMP. PDE10A emerges as a relevant pharmacological drug target for neurological conditions such as psychosis, schizophrenia, Parkinson's, Huntington's disease, and other memory-related disorders. In the current study, we subjected a set of 1,2,3-triazoles to be explored as PDE10A inhibitors using diverse computational approaches, including molecular docking, classical molecular dynamics (MD) simulations, Molecular Mechanics Poisson-Boltzmann Surface Area (MM-PBSA) calculations, steered MD, and umbrella sampling simulations. Molecular docking of cocrystallized ligands papaverine and PFJ, along with a set of in-house synthesized molecules, suggested that molecule 3i haded the highest binding affinity, followed by 3h and 3j. Furthermore, the structural stability studies using MD and MM-PBSA indicated that the 3h and 3j formed stable complexes with PDE10A. The binding free energy of -240.642 kJ/mol and -201.406 kJ/mol was observed for 3h and 3j, respectively. However, the cocrystallized ligands papaverine and PFJ exhibited comparitively higher binding free energy values of -202.030 kJ/mol and -138.764 kJ/mol, respectively. Additionally, steered MD and umbrella sampling simulations provided conclusive evidence that the molecules 3h and 3j could be exploited as promising candidates to target PDE10A.Communicated by Ramaswamy H. Sarma.

In-silicon studies on hydration in EcoRI-cognate DNA complex.

Restriction endonucleases (REs) cleave DNA at specific site in presence of Mg2+ ion. Experiments further emphasize the role of hydration in metal ion specificity and sequence specificity of DNA cleavage. However, the relation between hydration and specificity has not been understood till date. This leads us to study via all-atom molecular dynamics (MD) simulations how the hydration around the scissile phosphate group changes in presence of Mg2+ and Ca2+ and depend on the DNA sequence. We observe the least number of hydrogen bonds around the scissile phosphate group in presence of Mg2+ ion. We further find that the hydrogen bonds decrease at the scissile phosphate on mutating one base pair in the cleavage region of the DNA in Mg2+ loaded EcoRI-DNA complex. We also perform steered MD simulations and observe that the rate of decrease of fraction of hydrogen bonds is slower in the mutated complex than the unmutated complex.

Molecular Basis of the Recognition of Cholesterol by Cytochrome P450 46A1 along the Major Access Tunnel.

CYP46A1 is an important potential target for the treatment of Alzheimer's disease (AD), which is the most common neurodegenerative disease among older individuals. However, the binding mechanism between CYP46A1 and substrate cholesterol (CH) has not been clarified and will not be conducive to the research of relevant drug molecules. In this study, we integrated molecular docking, molecular dynamics (MD) simulations, and adaptive steered MD simulations to explore the recognition and binding mechanism of CYP46A1 with CH. Two key factors affecting the interaction between CH and CYP46A1 are determined: one is a hydrophobic cavity formed by seven hydrophobic residues (F80, Y109, L112, I222, W368, F371, and T475), which provides nonpolar interactions to stabilize CH, and the other is a hydrogen bond formed by H81 and CH, which ensures the binding direction of CH. In addition, the tunnel analysis results show that tunnel 2a is identified as the primary pathway of CH. The entry of CH induces tunnel 2e to close and tunnel w to open. Our results may provide effective clues for the design of drugs based on the substrate for AD and improve our understanding of the structure-function of CYP46A1.

In silico study to predict potential precursors of human dipeptidyl peptidase-IV inhibitors from hazelnut.

Chinese hazelnut was chosen to become a probable precursor of biological active peptides via computer simulations in this article. There were a large number of bioactive peptides in Chinese hazelnut sequences according to analytical results from the BIOPEP database. The most prominent of these was the inhibitory peptide for dipeptidyl peptidase-IV (DPP-IV; EC 3.4.14.5), which can be used to treat type 2 diabetes, so the theoretical method to obtain DPP-IV inhibitory peptides by hydrolysis with a single or combination of enzymes was studied. Cytotoxicity analysis performed by ToxinPred showed that all of the DPP-IV inhibitory peptides generated from protein hydrolysis were not cytotoxic. Structural interaction fingerprint analysis revealed that Asp663 and Phe357 may be important residues for ligand binding. In order to further understand the inhibitory mechanism of peptide, VR with lowest half maximum inhibitory concentration (IC50) and IPI (inhibitors have been reported) were selected as ligand of DPP-IV to perform steered molecular dynamics simulations and PMF calculations. The results showed that P1 is the preferred (un)binding tunnel for the inhibitors obtained. Our findings help in the development of new DPP-IV inhibitors which were derived from common food.Communicated by Ramaswamy H. Sarma.

Targeting N-Terminal Human Maltase-Glucoamylase to Unravel Possible Inhibitors Using Molecular Docking, Molecular Dynamics Simulations, and Adaptive Steered Molecular Dynamics Simulations.

There are multiple drugs for the treatment of type 2 diabetes, including traditional sulfonylureas biguanides, glinides, thiazolidinediones, α-glucosidase inhibitors, glucagon-like peptide-1 (GLP-1) receptor agonists, dipeptidyl peptidase IV (DPP-4) inhibitors, and sodium-glucose cotransporter 2 (SGLT2) inhibitors. α-Glucosidase inhibitors have been used to control postprandial glucose levels caused by type 2 diabetes since 1990. α-Glucosidases are rather crucial in the human metabolic system and are principally found in families 13 and 31. Maltase-glucoamylase (MGAM) belongs to glycoside hydrolase family 31. The main function of MGAM is to digest terminal starch products left after the enzymatic action of α-amylase; hence, MGAM becomes an efficient drug target for insulin resistance. In order to explore the conformational changes in the active pocket and unbinding pathway for NtMGAM, molecular dynamics (MD) simulations and adaptive steered molecular dynamics (ASMD) simulations were performed for two NtMGAM-inhibitor [de-O-sulfonated kotalanol (DSK) and acarbose] complexes. MD simulations indicated that DSK bound to NtMGAM may influence two domains (inserted loop 1 and inserted loop 2) by interfering with the spiralization of residue 497-499. The flexibility of inserted loop 1 and inserted loop 2 can influence the volume of the active pocket of NtMGAM, which can affect the binding progress for DSK to NtMGAM. ASMD simulations showed that compared to acarbose, DSK escaped from NtMGAM easily with lower energy. Asp542 is an important residue on the bottleneck of the active pocket of NtMGAM and could generate hydrogen bonds with DSK continuously. Our theoretical results may provide some useful clues for designing new α-glucosidase inhibitors to treat type 2 diabetes.

Disruption of conserved polar interactions causes a sequential release of Bim mutants from the canonical binding groove of Mcl1.

Mcl1 is an important anti-apoptotic member of the Bcl2 family proteins that are upregulated in several cancer malignancies. The canonical binding groove (CBG) located at the surface of Mcl1 exhibits a critical role in binding partners selectively via the BH3-domain of pro-apoptotic Bcl2 family members that trigger the downregulation of Mcl1 function. There are several crystal structures of point-mutated pro-apoptotic Bim peptides in complex with Mcl1. However, the mechanistic effects of such point-mutations towards peptide binding and complex stability still remain unexplored. Here, the effects of the reported point mutations in Bim peptides and their binding mechanisms to Mcl1 were computationally evaluated using atomistic-level steered molecular dynamics (SMD) simulations. A range of external-forces and constant-velocities were applied to the Bim peptides to uncover the mechanistic basis of peptide dissociation from the CBG of Mcl1. Although the peptides showed similarities in their dissociation pathways, the peak rupture forces varied significantly. According to simulations results, the disruption of the conserved polar contacts at the complex interface causes a sequential release of the peptides from the CBG of Mcl1. Overall, the results obtained from the current study may provide valuable insights for the development of novel anti-cancer peptide-inhibitors that can downregulate Mcl1's function.

Insights from molecular dynamics simulations and steered molecular dynamics simulations to exploit new trends of the interaction between HIF-1α and p300.

Hypoxia-inducible factor-1 (HIF-1) is a transcription factor that plays an important role in the expression of genes, whose function is exerted through protein-protein interactions (PPIs), such as the transcriptional co-activator (CREB)-binding protein (CBP) and p300. Under hypoxic conditions, HIF-1is stabilized and translocated to CBP or p300, leading to the hypoxic response cascade. Furthermore, the PPI between HIF and p300/CBP is a potential cancer target for their role in the hypoxic response. In this study, molecular dynamics (MD) simulation was used to explore the conformational change for the p300 binding to one subunit of HIF-1, namely HIF-1α. Results indicated that HIF-1α-p300 complex was stable during MD simulation. New H-bonds were made in the intra-chain of p300 with HIF-1α binding. Inhibiting the HIF-1α-p300 interaction modulated the HIF-1α identification of selective molecules, which may indicate the target metabolic and cellular processes that enable the survival and growth of tumors in cancer chemotherapy. CAVER 3.0 results suggested that three main tunnels were present, according to helices 1, 2 and 3 of p300. To explore the unbinding pathway for HIF-1α via p300, we selected helices 1, 2 and 3 on the HIF-1α as a new ligand to explore the unbinding pathway via its own tunnel. For helix 1, R368 in p300 formed a H-bond with E816 in HIF1-α. A345 and D346 in p300 formed H-bonds with N803 in HIF-1α. A H-bond existed between K351(p300) and E789 (Hif1-α). These molecules may be the key residues in the unbinding pathway via its tunnel.Communicated by Ramaswamy H. Sarma.

Three New Scorpion Chloride Channel Toxins as Potential Anti-Cancer Drugs: Computational Prediction of The Interactions With Hmmp-2 by Docking and Steered Molecular Dynamics Simulations.

Scorpion venom is a rich source of toxins which have great potential to develop new therapeutic agents. Scorpion chloride channel toxins (ClTxs), such as Chlorotoxin selectively inhibit human Matrix Methaloproteinase-2 (hMMP-2). The inhibitors of hMMP-2 have potential use in cancer therapy. Three new ClTxs, meuCl14, meuCl15 and meuCl16, derived from the venom transcriptome of Iranian scorpion, M. eupeus (Buthidea family), show high sequence identity (71.4%) with Chlorotoxin. Here, 3-D homology model of new ClTxs were constructed. The models were optimized by Molecular Dynamics simulation based on MDFF (molecular dynamics flexible fitting) method. New ClTxs indicate the presence of CSαß folding of other scorpion toxins. A docking followed by steered molecular dynamics (SMD) simulations to investigate the interactions of meuCl14, meuCl15, and meuCl16 with hMMP-2 was applied. The current study creates a correlation between the unbinding force and the inhibition activities of meuCl14, meuCl15 and meuCl16 to shed some insights as to which toxin may be used as a drug deliverer. To this aim, SMD simulations using Constant Force Pulling method were carried out. The SMD provided useful details related to the changes of electrostatic, van de Waals (vdW), and hydrogen-bonding (H-bonding) interactions between ligands and receptor during the pathway of unbinding. According to SMD results, the interaction of hMMP-2 with meuCl14 is more stable. In addition, Arginine residue was found to contribute significantly in interaction of ClTxs with hMMP-2. All in all, the present study is a dynamical approach whose results are capable of being implemented in structure-based drug design.

Single-Molecule Force Spectroscopy: Experiments, Analysis, and Simulations.

The mechanical properties of cells and of subcellular components are important to obtain a mechanistic molecular understanding of biological processes. The quantification of mechanical resistance of cells and biomolecules using biophysical methods matured thanks to the development of nanotechnologies such as optical and magnetic tweezers, the biomembrane force probe, and atomic force microscopy (AFM). The quantitative nature of force spectroscopy measurements has converted AFM into a valuable tool in biophysics. Force spectroscopy allows the determination of the forces required to unfold protein domains and to disrupt individual receptor/ligand bonds. Molecular simulations as a computational microscope allow investigation of similar biological processes with an atomistic detail. In this chapter, we first provide a step-by-step protocol of force spectroscopy experiments using AFM, including sample preparation, measurements, and analysis and interpretation of the resulting dynamic force spectrum in terms of available theories. Next, we present the background for molecular dynamics (MD) simulations focusing on steered molecular dynamics (SMD) and the importance of bridging computational tools with experimental techniques.

Non-syndromic Mitral Valve Dysplasia Mutation Changes the Force Resilience and Interaction of Human Filamin A.

Filamin A (FLNa), expressed in endocardial endothelia during fetal valve morphogenesis, is key in cardiac development. Missense mutations in FLNa cause non-syndromic mitral valve dysplasia (FLNA-MVD). Here, we aimed to reveal the currently unknown underlying molecular mechanism behind FLNA-MVD caused by the FLNa P637Q mutation. The solved crystal structure of the FLNa3-5 P637Q revealed that this mutation causes only minor structural changes close to mutation site. These changes were observed to significantly affect FLNa's ability to transmit cellular force and to interact with its binding partner. The performed steered molecular dynamics simulations showed that significantly lower forces are needed to split domains 4 and 5 in FLNA-MVD than with wild-type FLNa. The P637Q mutation was also observed to interfere with FLNa's interactions with the protein tyrosine phosphatase PTPN12. Our results provide a crucial step toward understanding the molecular bases behind FLNA-MVD, which is critical for the development of drug-based therapeutics.

Screening of hepatitis C NS5B polymerase inhibitors containing benzothiadiazine core: a steered molecular dynamics.

Hepatic C virus (HCV) is a global health problem, resulting in liver cirrhosis and inflammation that can develop to hepatocellular carcinoma and fatality. The NS5B polymerase of HCV plays an important role in viral RNA replication process, making it an attractive therapeutic target for design and development of anti-HCV drugs. To search new potent compounds against the HCV NS5B polymerase, the molecular docking and the steered molecular dynamics (SMD) simulation techniques were performed. The potential potent inhibitors of the NS5B polymerase were screened out from the ZINC database using structural similarity search and molecular docking technique. Five top-hit compounds (the ZINC compounds 49888724, 49054741, 49777239, 49793673, and 49780355) were then studied by the SMD simulations based on the hypothesis that a high rupture force relates to a high binding efficiency. The results demonstrated that the ZINC compound 49888724 had a greater maximum rupture force, reflecting a good binding strength and inhibitory potency than known inhibitors and the rest four ZINC compounds. Therefore, our finding indicated that the ZINC compound 49888724 is a potential candidate to be a novel NS5B inhibitor for further design. Besides, the van der Waals interaction could be considered as the main contribution for stabilizing the NS5B-ligand complex.

Investigating the effect of different transducer stiffness values on the contactin complex detachment by steered molecular dynamics.

This study investigated the adhesion behavior of Contactin4 (CNTN4), a member of Immunoglobulin Super Family (Ig-SF) of cell adhesion molecules. Contactin4 plays a crucial role in the formation, maintenance, and plasticity of neuronal networks. Contactin in its complex configuration with protein tyrosine phosphatase gamma (PTPRG) was selected for simulation. By utilizing Steered Molecular Dynamics (SMD), the uniaxial force was applied to induce unbinding of the complex, and the force-induced detachment of complex components was probed. Three sets of simulations with three values of transducer stiffness and five pulling speeds were designed. Our results showed the dependence of unbinding force on both accessible parameters of pulling speed and spring stiffness. By increasing the stiffness value and pulling speed the rupture force increased. Accordingly, the dissociation rates due to the Bell's theory based on rupture forces and loading rates were calculated.

In silico study of amphiphilic nanotubes based on cyclic peptides in polar and non-polar solvent.

The stability of cyclic peptide assemblies (CPs) forming a macromolecular nanotube structure was investigated in solvents of different polarity using computational methods. The stability and structure of the complexes were studied using traditional molecular dynamics (MD). Energy of dissociation was estimated from steered MD in combination with umbrella sampling simulations. A cyclic peptide nanotube (CPNT) was constructed by stacking of eight cyclo[(D-Trp-L-Gln-D-Trp-L-Glu)2], and hereafter is referred to as (WQWE)8. Its dissociation was studied by pulling 1, 2, or 3 subunits from the nanotube. The crucial point in the dissociation event of the CP subunit(s) is the breaking of backbone-backbone hydrogen bonds and consecutive annihilation of side chain interactions. Gibbs free energy calculations to estimate the binding affinity of CP subunit(s) reveal that the (WQWE)8 nanotube is significantly more stable in non-polar environments than in polar environments. The presently investigated nanotube, (WQWE)8, displays a higher stability in polar solvent than the previously studied nanotube, (QAEA)8. It appears that tryptophan contributes favorable to the improved stability by forming side chain-side chain hydrogen bonds.

Characterizing the Free-Energy Landscape of MDM2 Protein-Ligand Interactions by Steered Molecular Dynamics Simulations.

Inhibition of p53-MDM2 interaction by small molecules is considered to be a promising approach to re-activate wild-type p53 for tumor suppression. Several inhibitors of the MDM2-p53 interaction were designed and studied by the experimental methods and the molecular dynamics simulation. However, the unbinding mechanism was still unclear. The steered molecular dynamics simulations combined with Brownian dynamics fluctuation-dissipation theorem were employed to obtain the free-energy landscape of unbinding between MDM2 and their four ligands. It was shown that compounds 4 and 8 dissociate faster than compounds 5 and 7. The absolute binding free energies for these four ligands are in close agreement with experimental results. The open movement of helix II and helix IV in the MDM2 protein-binding pocket upon unbinding is also consistent with experimental MDM2-unbound conformation. We further found that different binding mechanisms among different ligands are associated with H-bond with Lys51 and Glu25. These mechanistic results may be useful for improving ligand design.