Aldose reductase (AR) is a key enzyme in the polyol pathway and plays a major role in the progression of secondary complications of diabetes. Despite extensive efforts to develop natural and synthetic aldose reductase inhibitors (ARIs), most candidates have shown limited clinical efficacy, highlighting the need for more potent and selective inhibitors. In this study, we have systematically evaluated the inhibitory potential of vitamin K family members (vitamin K1, vitamin K2, and vitamin K3) using molecular docking, protein-ligand interaction analysis, molecular dynamics simulations, and enzyme kinetics. Docking analysis predicted that vitamin K2 has the highest binding affinity for AR. Subsequent molecular dynamics simulations revealed that both vitamin K1 and vitamin K2 formed stable complexes with the protein, exhibiting comparable RMSD (∼0.5 Å difference), similar RMSF profiles, and reduced radius of gyration, indicating compact and stable binding. Interaction analysis demonstrated that ligand binding is predominantly driven by hydrophobic interactions, with vitamin K2 forming a higher number of hydrophobic contacts, while vitamin K1 exhibited slightly more hydrogen bonding. Molecular Mechanics/Generalized Born Surface Area (MM/GBSA) results further supports stronger binding of vitamin K2 (-56 kcal/mol) compared to vitaminK1 (-51 kcal/mol). Consistent with these findings, enzyme kinetics showed a slightly lower Ki value for vitamin K2 than vitamin K1. In contrast, vitamin K3 failed to maintain stable binding and moved out of the active site during simulation. Overall, the study highlights that hydrophobic interaction-driven stabilization plays a key role in ligand binding, and identifies vitamin K1 and vitamin K2 as promising inhibitors against AR, with vitamin K2 exhibiting more favourable hydrophobic interactions and binding stability.
The phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (p110α) is a key lipid kinase encoded by PIK3CA gene that regulates multiple intracellular signaling pathways. The nsSNPs within PIK3CA can alter protein structure and function, thereby influencing cellular processes and increasing susceptibility to various types of cancers. This study aimed to systematically evaluate the consequences of nsSNPs in PIK3CA, and their potential influence on protein behavior and drug interaction. A comprehensive set of in silico tools-including functional impact predictors, structural stability analyzers, conservation-based algorithms, and protein property assessment tools-was employed to investigate the consequences of nsSNPs. Structural modeling was performed to generate native and mutant p110α structures, followed by molecular docking to evaluate how these variants influence binding affinity with Alpelisib. Ten nsSNPs-G363V, R398C, R555K, C769G, F801C, R808W, R808Q, E849K, E849G, and R992Q-were consistently predicted to be deleterious across all functional prediction platforms and were found to affect evolutionarily conserved residues or domains. Among these, six variants (R398C, C769G, F801C, R808W, R808Q, and R992Q) were further predicted to significantly alter physicochemical protein properties. Docking analyses revealed variant-specific shifts in binding affinity. R555K displaying the highest affinity even higher than native and R398C showing lowest affinity toward Alpelisib. These suggest a potentially altered or dysregulated drug-protein interaction. However, the biological benefit or detriment of this shift warrants further experimental evaluation. The findings of this study may enhances our understanding of how PIK3CA variants contribute to predict disease risk and may support future precision-medicine approaches targeting the PI3K pathway.
Leishmania donovani (L. donovani) causes one of the dreadful diseases called visceral leishmaniasis (VL) affecting millions people in India and some parts of Africa. Several efforts have been made, no effective vaccine is available to date. Immunoinformatics methods play a critical role in predicting potential vaccine candidates. The present study showed extensive analysis of L. donovani proteins for the identification of MHC class II restricted epitopes, cryptic epitopes and B cell epitopes which may help in providing a diverse immune response. A total of 130 epitopes from L. donovani were identified on the basis of their binding towards HLA alleles. These peptides were checked for their hydrophobicity, IFN-γ and IL-10 inducing potentials.A total of 24 peptides were selected for modelling and docking. Molecular dynamics (MD) simulation of the docked complexes with the HLA-DRB1*0101 allele suggests 3 epitopes have the best binding interactions throughout the 100 ns simulation. These epitopes can be considered as potential vaccine candidates and may need further experimental validation before practical application. We also made an attempt to identify cryptic epitopes which remain largely unexplored in the context of L. donovani vaccine design and diagnosis. Out of 24 peptides selected previously, IC50 values of 5 peptides were decreased after the modification with tyrosine residue. Further replacement identified 2-3 modified peptides showing better docking and lesser IC50 values than the original peptides, suggesting cryptic behaviour of the epitopes. Linear and conformation B cell epitopes were also predicted which may suggest importance of these residues for eliciting humoral response.
Conformational B-cell epitopes typically rigidify upon antibody binding, while distal antigen regions often become more flexible, yet the mechanistic basis remains unclear. Here we analyze 17 structurally similar lysozyme-antibody complexes using Anisotropic Network Models, Perturbation Response Scanning, and residue-level network analysis. We find that epitope residues are intrinsically low-mobility and act as dynamically influential sites in the unbound antigen, mechanically coupled to distant regions that increase flexibility upon binding. Normal mode analysis reveals that epitopes behave as rigid blocks within global motions, supporting their mechanically cohesive behavior within the antigen structure. These findings suggest that mobility redistribution upon antibody binding arises from pre-existing long-range dynamic coupling encoded in the antigen fold, providing a mechanistic framework linking epitope rigidity, allosteric communication, and global dynamical regulation in lysozyme-Fab complexes.
A method described in this publication is able (i) to display features related to the general genome organization and packaging; (ii) to identify explicitly motifs responsible for these features and the distribution of motifs over the genome; the most pronounced motifs can be screened against available databases for further studies; (iii) to assess the impact of mutations; the salient mutations can be tried experimentally and used for genome editing; and (iv) to quantify the information that can be insightful for general studies of molecular evolution of living organisms. The method is based on the transitional genome mapping that may be considered as an extension and adaptation of the general correlation technique and is aimed at displaying significant correlations throughout all genome scales. Combining this method with a correlation entropy as the output variables extends significantly its abilities. The relevant theory was developed and tested for the assessment of statistical significance of the striking features in correlation characteristics. The particular applications are illustrated with the genomes of satellite tobacco mosaic virus, coronaviruses SARS-CoV-2 and SARS-CoV, hepatitis delta virus and viroid representatives from two main families. As shown, the genome editing with salient ordering mutations strengthens significantly the features related to the genome packaging and organization. The relevant underlying quasi-periodic patterns can be reconstructed explicitly. The resulting edited viral sequences can potentially be used for the development of vaccines, drug targeting and for the general studies of underlying molecular mechanisms.
Azetidine-functionalized pyrrolo[2,3-d]pyrimidines were designed to examine the comparative influence of methyl-, propyl-, and butylsulfonyl substitution among final synthesized analogues on antiproliferative activity and structure-based behavior. Following in silico ADME filtering and docking-based prioritization, selected analogues were synthesized and characterized. Among the evaluated compounds, BCD17 showed the most favorable predicted docking profile toward 17β-hydroxysteroid dehydrogenase type 1 (3HB5), with a docking score of -9.130 kcal/mol and Glide energy of -69.236 kcal/mol. Molecular dynamics simulation and MM-GBSA analysis further supported the stable predicted interaction profile of BCD17, with a mean binding free energy of -109.49 ± 7.52 kcal/mol. DFT analysis indicated a balanced electronic profile compatible with its predicted structure-based performance. In vitro MTT evaluation confirmed antiproliferative activity in MCF-7 cells, with IC50 values of 0.76, 0.75, and 1.83 µM for BCD1, BCD9, and BCD17, respectively. Although BCD1 and BCD9 showed slightly lower IC50 values, BCD17 was prioritized based on its integrated computational stability, apoptosis-related response, and preliminary selectivity profile. MCF-7 cell-cycle and Annexin V/PI apoptosis analyses further supported the antiproliferative and pro-apoptotic effect of BCD17, although direct 17β-HSD1 target engagement was not experimentally confirmed. In MCF-10A cells, BCD9 and BCD17 showed comparatively lower toxicity than 5-FU. Overall, BCD17 emerged as a promising integrated lead candidate for further optimization and biological validation.
Avian influenza viruses (AIVs) are RNA viruses endemic to wild birds, with highly pathogenic strains such as H5N1 causing major outbreaks. Hemagglutinin (HA) and neuraminidase (NA) proteins are key mediators of viral entry and spread. Cyclotides, plant-derived peptides with stable structures and potent bioactivity, have emerged as novel antiviral candidates. This study explored cyclotide-based drug design against H5N1 using computational approaches. HA and NA protein sequences were retrieved from NCBI and their physicochemical properties analyzed using ProtParam. Cyclotides from CyBase were modeled and refined in PyMOL, followed by molecular docking via ClusPro. Binding energies were further assessed with molecular mechanics generalized born surface area, while toxicity was predicted using ToxinPred. Molecular dynamics simulations supported the stability analyses. Results revealed that Psyle B and Psyle C exhibited strong interactions with HA and NA. Psyle B achieved a docking score of -1001.3 kcal/mol, and Psyle C -1031.6 kcal/mol, both predicted as non-toxic. These findings highlight Psyle B and Psyle C as promising antiviral leads. Nonetheless, in vitro and in vivo validation remains essential to confirm their therapeutic potential.
Targeting the immune checkpoint programmed cell death-ligand 1 (PD-L1) faces diverse challenges against pancreatic ductal adenocarcinoma (PDAC), the most common type of cancer-related death. Shedding of PD-L1 from the tumor cell surface by A Disintegrin And Metalloprotease 10 (ADAM10) may impede the function of antitumor immune cells. Understanding the molecular mechanisms behind this interaction is urgently needed. This study elucidates the mechanism of ADAM10/PD-L1 interaction in the pancreatic adenocarcinoma cell line, PANC-1, using cellular assays, molecular docking, and molecular dynamics (MD) simulations. Cellular assays validated that ADAM10 interacts with PD-L1 and sheds the PD-L1 from the cell surface, generating soluble PD-L1 (sPD-L1). Concurrently, the docking results identified critical residues and potential contact points, providing strong evidence for a stable and specific interaction between ADAM10 and PD-L1. The molecular dynamics (MD) simulations further confirmed the structural integrity of the ADAM10/PD-L1 complex, predicting its compactness and stability. These results elucidate how ADAM10 recognizes and interacts with PD-L1, facilitating its cleavage. Future work can leverage these findings to develop potent cancer immunotherapies that inhibit the ADAM10/PD-L1 interaction and the generation of soluble molecules.
HIV-2 protease (PR2) is a homodimeric protein essential for viral maturation and represents a key therapeutic target. Structural studies of PR2 report a semi-open apo form and closed ligand-bound conformations. However, these static structural observations do not fully capture the intrinsic dynamics of PR2. Here, we use molecular dynamics simulations to investigate the conformational landscape sampled by the monoprotonated PR2 in the absence of ligand. The analysis of three independent trajectories reveals a reproducible two-stage dynamical behavior. An initial stage corresponds to a transition from semi-open conformations through transient opening, followed by a second phase characterized by flap closure leading to predominantly closed conformations. These results suggest that PR2 can both access open conformations and spontaneously reach closed states in the absence of ligand. Using complementary geometric descriptors, we show that PR2 closure involves a coordinated reorientation of the flap regions, in which flap B moves from a position in front of flap A to a position behind it, consistently preceding closure. During the late stage, PR2 samples multiple structurally distinct closed conformations, including extended, bent, and inward-bent states. Some of these conformations are consistent with those observed in ligand-bound systems, suggesting that they pre-exist in the apo enzyme. These results provide mechanistic insights into flap rearrangements in PR2. They are also consistent with a conformational selection mechanism in which ligand binding preferentially stabilizes conformations already sampled by the apo enzyme. This findings provide new insights into the intrinsic dynamics of apo PR2 and may assist future structure-based inhibitor design.
Methylenetetrahydrofolate reductase 2 (MTHFR2) plays a vital role in the one-carbon (1C) pathway, mediating plant immunity in Arabidopsis against rice blast, a role conserved across the plant kingdom. To understand the structural, conformational dynamics, evolutionary modifications and molecular activity of MTHFR2, we identified that the A55V substitution in an EMS mutant of Arabidopsis which diminishes disease resistance against rice blast. We hypothesized about its key role in regulating the enzyme activities involved in the 1C metabolic pathway. Molecular dynamic simulations, demonstrated that the A55V mutation induces structural alterations and instability at the enzyme active site, affecting its function. Ramachandran plot analysis revealed that the mutant had a slight reduction in favored conformations (94.097% vs. 94.662%). The root mean square deviation analysis exhibited 48.7% and 183% increase in structural deviation and conformational variability, respectively, in the mutant suggesting lower stability. Root mean square fluctuation analysis showed a 39.8% increase in the flexibility of residues in the NADH binding pocket, indicating impaired ligand recognition, altering the resultant product (5-CH3-THF). Differences in global compactness (radius of gyration) and solvent exposure (solvent-accessible surface area) were minimal, yet localized instability was apparent. Evolutionary analysis revealed that MTHFR2 was highly conserved among plant species and soybean orthologs had 81% similarity, therefore exhibit nonhost resistance. In vitro functional assays showed that soybean extracts suppressed Magnaporthe oryzae conidia germination and development, implying a conserved metabolite-based defense mechanism. Collectively, our findings establish a structural and functional framework for understanding MTHFR2's conserved role and lay the groundwork for future research aiming at connecting enzyme function to metabolite-based immunity in plants.
The VWF A1-GPIbα interaction mediates platelet tethering under flow and exhibits a force-dependent mechanical response. Previous simulations identified R1334 within a multi-residue electrostatic region of the interface, but its specific contribution to equilibrium interfacial stability and tensile resistance has been less directly examined through matched WT and R1334A comparisons. Here, replicated equilibrium and multi-rate steered molecular dynamics simulations were used to further examine how R1334A alters these previously reported interactions and the mechanical response of the complex. R1334A decreased the mean interfacial heavy-atom contact count from 306.9±15.5 to 244.3±40.1 and left only sparse residual contacts at residue 1334. Under tensile loading, the mutation lowered the maximum observed resistance by 423-482 pN across all tested rates. Residue-resolved analysis showed that WT R1334 repeatedly engaged GPIbα D18, H37, S39 and N61 during maintenance and tensile separation of the bound interface, whereas these coordinated interactions were largely absent after mutation. Notably, R1334A could reach sustained interfacial contact loss at greater extension despite its lower tensile resistance, showing that delayed contact loss does not necessarily indicate stronger binding but can instead arise from the persistence of weaker residual interactions. These results refine the established role of R1334 by directly linking its interactions with neighbouring GPIbα residues to equilibrium interfacial stability and tensile resistance across multiple loading rates.
Alternative oxidase (AOX) is cyanide insensitive terminal oxidase of plant mitochondrial electron transport chain serving as important bypass from classical cytochrome pathway for reduction in generation of reactive oxygen species (ROS) and for maintenance of redox homeostasis under diverse stress conditions. While many studies have emphasised on AOX structure in premature form, the mitochondria localised mature form which is biologically relevant for substrate interaction is not much explored. In the current study, we have performed comparative structural and molecular docking analysis of mature AOX isoforms from stress-sensitive Arabidopsis thaliana and extremophile Eutrema salsugineum to elucidate molecular determinants of ubiquinol binding and stress tolerance. Homology models of AOX isoforms were constructed and validated through SWISS-MODEL, GalaxyWEB, and Ramachandran plot. Docking of activators (pyruvate, glyoxylate, oxaloacetate, 2-oxoglutarate) and substrate (ubiquinol) revealed strong hydrophobic interactions, predominantly Pi-Pi alkyl bonds, within conserved catalytic domains. Among all isoforms, AtAOX1a and EsAOX1a exhibited highest binding affinities with ubiquinol. In-silico site-directed mutagenesis and molecular dynamics simulations demonstrated that substitution of conserved hydrophobic residues, Val184 in AtAOX1a and Val118 in EsAOX1a, with aspartic acid (V→D) did not alter structural stability but significantly disrupted ubiquinol-binding pocket and reduced binding affinity indicating conserved valine residues as hydrophobic anchors crucial for substrate stabilisation. The study underscores hydrophobic interactions as key determinant of AOX function proposing valine as potential site for in-vitro mutagenesis to modulate AOX-mediated stress responses. The findings serve as foundation for future experimental studies addressing AOX mediated stress responses and exploring strategies for improving plant stress tolerance.
Leucyl-tRNA synthetase 1 (LARS1) is a multi-domain enzyme essential for translational fidelity and amino acid sensing, with its anticodon-binding domain (ABD) playing a critical role in tRNA recognition. Missense variants in LARS1 have been associated with infantile liver failure syndrome type 1 (ILFS1), yet the structural mechanisms underlying their functional impact remain poorly understood. In this study, we investigated the structural and dynamical effects of the patient derived D794G mutation located in the anticodon-binding domain of LARS1. To evaluate its potential pathogenicity, comparative molecular dynamics (MD) simulations were performed for wild-type LARS1, the D794G variant, and the ClinVar-reported pathogenic variant Y835C. Structural analysis revealed that Asp794 contributes to a stabilizing interaction network involving hydrogen bonding with Tyr835 and aromatic stacking between Tyr835 and Phe797, linking adjacent α-helices within the ABD. Substitution with glycine abolished the Asp794-Tyr835 hydrogen bond and altered the local hydration and contact environment. Although RMSD and radius-of-gyration analyses indicated preservation of the global fold and compactness, RMSF, hydrogen-bond, SASA, contact, and principal-component analyses identified variant-dependent changes in local flexibility and conformational sampling, most prominently for D794G. These effects extend beyond the mutation site, suggesting propagation of structural perturbations to neighboring domains. Overall, these findings provide mechanistic insight into variant-associated changes in LARS1 structural dynamics and generate testable hypotheses regarding the importance of the 794-797-835 interaction network. The results highlight the importance of local interaction networks in maintaining domain stability and suggest that disruption of these interactions may have functional consequences.
Calorie restriction (CR) exerts its anti-aging beneficial effects by influencing key metabolic and stress-response regulators, including AMPK, SIRT1, COX-2, and catalase. A dietary flavonoid called Chrysin is thought to replicate some of the benefits of CR, although its multi-target molecular interactions are still unclear. This study employs molecular docking, MM-GBSA calculations, ADMET screening, and 100-ns molecular dynamics (MD) simulations to investigate the binding behavior, energy stability, and dynamic characteristics of Chrysin with respect to key CR-related proteins. According to docking studies, chrysin exhibits substantial affinity for all proteins examined, with AMPK showing the most favorable interaction profile. Catalase exhibited comparatively decreased binding but retained its structural integrity, while SIRT1 and COX-2 showed moderate affinity. MD simulations revealed that the COX-2 complex exhibited significant instability, while catalase provided modest stabilization, and AMPK and SIRT1 formed extremely stable complexes. The key catalytic domains of these proteins are not disrupted by chrysin binding, according to RMSF analysis. High oral absorption capacity and good drug-likeness were predicted by ADMET. All things considered, chrysin seems to function as a promising calorie-restriction mimetic by interacting with several metabolic regulators, with strong and consistent interactions seen between AMPK and SIRT1. These findings support the potential for additional research as a multi-target therapeutic anti-aging alternative.
Diabetes mellitus (DM) is a chronic metabolic condition characterized by persistently high blood glucose levels. The emergence of drug resistance in DM therapy presents significant clinical challenges, highlighting the need for the development of new and effective therapeutic agents. The primary objective of this study is to investigate the potential anti-diabetic activity of a newly designed Schiff base molecule, 2-[(2,4-dichloro-benzylidene)-amino]-4-methyl-phenol [compound (I)]. Compound (I) was synthesized, and its structure was confirmed using spectroscopic methods such as UV-Vis, FT-IR, and NMR spectroscopy. Single-crystal X-ray diffraction (SC-XRD) revealed that compound (I) crystallizes in the orthorhombic P212121 space group and is stabilized by a combination of intra- and intermolecular hydrogen bonds (O-H···N, O-H···O, C-H···O, and C-H···Cl), forming a robust 3D supramolecular framework. The compound (I) was evaluated for its anti-diabetic potential using In-silico α-amylase inhibition assays. The compound's cytocompatibility was assessed using the MTT assay. The interaction pattern and binding orientation of compound (I) within the catalytic site of human pancreatic α-amylase (HPA) was explored through molecular docking studies, supporting the biological results. Furthermore, the stability of the docked complex was evaluated through 100 ns molecular dynamics (MD) simulations. In addition, the quantum computational analysis was investigated using Density Functional Theory (DFT) at the B3LYP/6-31++G(d,p) level were conducted to explore the electronic properties of compound (I), including electrophilicity, nucleophilicity, hardness, and softness. Moreover, the binding thermodynamics of compound (I) with α-amylase were examined by Isothermal Titration Calorimetry (ITC), which elucidated the thermodynamic parameters and interaction mechanisms.
Post Translational Modifications significantly influence conformational preferences and activity of proteins. In Protein Kinase-B (Akt2), numerous residues undergo phosphorylation and O-GlcNAcylation, in same region. This results in a crosstalk mechanism modulating Akt2 activity, and is dysregulated in metabolic diseases such as type-II diabetes. However, atomic level details of how such modifications influence Akt2 dynamics and activity is poorly understood. In this work, we employed in silico approaches to explore the conformational preferences and functional consequences of Akt2 in O-GlcNAcylated, phosphorylated and unmodified states. Multi-microsecond simulations, resulting in total sampling of 7.6μs were performed on Akt2 in all the above states, and the structural effects of such modifications were analyzed. We found that such diverse modifications result in considerable difference in Akt2 dynamics. In phosphorylated state, the activation loop of Akt2 is highly flexible and has high solvent accessibility, in contrast with the buried nature of O-GlcNAcylated Akt2. This involved contacts between the O-GlcNAc residues and the pleckstrein homology domain, resulting in its burial. In the phosphorylated state, an elaborate hydrogen bonded network was observed, where phosphorylated threonine residues and active site residues formed hydrogen bonds with the phosphatidylinositol (3,4,5)-trisphosphate (PIP3) binding residues, forming the structural basis of autoinhibition and phosphatase shielding by Akt2. We further sampled for first time, the conformational changes associated with the PIP3 induced Akt2 activation from the Akt2 simulations in presence of PIP3, ATP and Mg2+ cofactor. This work demonstrated the structural effects of varied post-translational modifications, which form a crosstalk mechanism regulating Akt2 activity.
Porcupines (Hystrix spp.) are widely distributed rodents known not only for their ecological roles but also for their traditional medicinal applications. In Indonesia, quills from species like Hystrix javanica, Hystrix sumatraensis, and Hystrix brachyura are traditionally used to relieve pain and promote wound healing. Scientific studies have revealed that porcupine quills are rich in bioactive compounds, including keratin, peptides, flavonoids, and triterpenoids. Keratin, a structural protein abundant in epithelial cells extracted from porcupine quill, has been found to support various wound-healing processes, having antibacterial properties, and the ability to induce apoptosis in breast cancer cells. Meanwhile, the vascular endothelial growth factor receptor 2 (VEGFR-2) plays a crucial role in angiogenesis and tissue regeneration during the wound-healing process. This study employed exploratory structural bioinformatics to study the potential interaction between porcupine quill-derived homolog proteins and VEGFR-2 from our previous published work. Through protein-protein docking and molecular dynamics simulation, we predicted molecular recognition and binding affinity computationally. Keratin type II (P50446), Krt2 (B2RTP7), and Keratin isoform X1 (A0A6P5R3S8) demonstrated the most robust and stable interactions with VEGFR-2. Contrarily, Keratin type II (P04264) displayed comparatively greater structural fluctuations and weaker stability, suggesting less optimal and more flexible binding interactions with the receptor. Thus, this research provides preliminary hypothesis evidence of potential interactions between porcupine quill-derived proteins and VEGFR-2, supporting further experimental studies to determine whether these interactions have biological relevance in wound-healing processes.
Cadmium ion (Cd2+) is a toxic environmental pollutant, and DNA aptamers are promising recognition elements for its detection. However, the dynamic response of aptamer-metal complexes to temperature cycling remains largely unexplored. Here, six independent all-atom molecular dynamics simulations (500 ns each) were performed to investigate a Cd2+-specific DNA aptamer under temperature cycling (300 K → 375 K → 300 K). A robust "conformational memory" effect was discovered. In the initial binding phase, Cd2+ coordinated exclusively with the O2 atom of residue DC12 (Cd-O ≈ 0.206 nm). Upon heating to 323-360 K, the ion fully dissociated. Notably, Cd2+ rebound at even higher temperatures (still during the heating ramp), but the coordination site irreversibly migrated to phosphate backbone oxygens (O1P/O2P) of different residues (Cd-O 0.200-0.202 nm). The average contacts per frame increased from 2.28 to 3.31. Dissociation and rebinding followed different free-energy branches, yielding clear temperature hysteresis loops (area 0.84-5.31 K·nm). Transition temperatures followed a normal distribution (mean 338.8 K, SD 10.3 K; Shapiro-Wilk p = 0.250), explained by a random-barrier model with activation energy fluctuations (Ea = 33.7 ± 1.0 kJ/mol). After thermal cycling, the DNA backbone became more flexible (RMSF increased from 0.207 nm to 0.422 nm), and the free-energy basin along the distance axis broadened from 0.19 nm to 0.39 nm, indicating a more expanded, entropically favored ensemble. These findings reveal an irreversible switch from nucleobase to backbone phosphate coordination, demonstrate conformational memory in a metal-aptamer system, and provide design principles for intelligent nucleic acid devices.
