Revealing the Molecular Mechanisms of PIP2 Binding and Regulating KCNQ1: Twists, Links, and Binding-Site Transfers via the Developed SIMDA Strategy

Revealing the Molecular Mechanisms of PIP2 Binding and Regulating KCNQ1: Twists, Links, and Binding-Site Transfers via the Developed SIMDA Strategy | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Revealing the Molecular Mechanisms of PIP2 Binding and Regulating KCNQ1: Twists, Links, and Binding-Site Transfers via the Developed SIMDA Strategy Huanxiang Liu, LingLing Wang, Shu Li, Yunsen Zhang, Huiyong Sun, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6059518/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Voltage-gated potassium channel KCNQ1 (Kv7.1) is essential for various physiological processes, including cardiac repolarization, epithelial ion transport, and inner ear function. Its functional versatility arises from interactions with auxiliary KCNE proteins, calmodulin (CaM), and the lipid phosphatidylinositol 4,5-bisphosphate (PIP2), which modulate its gating properties, trafficking, and activity in a tissue-specific manner. Despite advancements in structural and functional studies, the precise molecular mechanisms underlying PIP2's role in KCNQ1 activation, as well as the contribution of KCNE3 and CaM to PIP2-KCNQ1 binding, remain unclear. Here, we present the Stepwise Integrated Multi-scale Dynamics and Advanced Analysis (SIMDA) framework, which incrementally integrates coarse-grained and all-atom molecular dynamics, quantum mechanics, and well-tempered metadynamics, along with advanced clustering and energy analysis techniques. Over 280 µs multi-scale simulations revealed eight PIP2-binding sites, including new regions on the S0 segment and the S6-HA junction. We also observed KCNE3 enhances the “twist” effect at KCNQ1’s C-terminal, promoting PIP2 binding. Furthermore, eight PIP2 dissociation pathways revealed transitions across binding sites, which highlight its dynamic transfer behavior. These findings provide a comprehensive understanding of PIP2-mediated regulation of KCNQ1 and establish SIMDA as a robust tool for studying lipid-protein dynamics. Biological sciences/Biophysics/Membrane biophysics Biological sciences/Computational biology and bioinformatics Biological sciences/Molecular biology Stepwise Integrated Multi-scale Dynamics and Advanced Analysis Protein-lipid Interaction PIP2-KCNQ1 binding Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 INTRODUCTION Voltage-gated potassium channels are essential for regulating cellular excitability, ionic homeostasis, and signal transduction across diverse tissues. Among these, the KCNQ1 (Kv7.1) channel stands out for its functional versatility, playing pivotal roles in processes such as cardiac repolarization, epithelial ion transport, and inner ear function 1 – 4 . This versatility arises from KCNQ1’s ability to interact with auxiliary KCNE proteins (e.g., KCNE1, KCNE2, and KCNE3) and regulatory molecules like calmodulin (CaM). These interactions enable KCNQ1 to exhibit tissue-specific gating properties, trafficking behavior, and activity 5 – 8 . For instance, the binding of KCNE3 suppresses the channel’s voltage dependence, converting KCNQ1 into a constitutively open potassium channel at physiological membrane potentials 9 – 12 . Similarly, CaM, a small, highly conserved calcium-binding protein, binds to the intracellular C-terminal domain of KCNQ1 to regulate channel gating, folding, and trafficking 13 – 15 . In addition to these protein interactions, KCNQ1 channel activity is tightly regulated by phosphatidylinositol 4,5-bisphosphate (PIP2), a signaling lipid that plays a central role in controlling channel function 16 – 18 . For example, PIP2 plays a crucial role in mediating the interaction between the voltage-sensing domains (VSDs) and pore domains (PDs), thereby facilitating the opening of KCNQ channels (Fig. 1 a, b) 19 – 23 . PIP2 specifically binds to positively charged residues within KCNQ1’s C-terminal domain, stabilizing the channel’s open state and enabling constitutive activity at resting membrane potentials 9 , 24 – 26 . Importantly, this PIP2-binding region is located near the CaM-binding site. Previous studies suggest that CaM regions can stabilize KCNQ1 and potentially compete with PIP2 for binding to KCNQ1 9, 25 . This cooperative interaction between CaM and PIP2 significantly increases the channel’s sensitivity to PIP2, ensuring that PIP2 can efficiently activate KCNQ1 and maintain its function 9 . In the KCNQ1/KCNE complex, PIP2 also plays an important regulatory role. KCNE3 proteins alter the conformation of KCNQ1’s VSDs, locking it in an inactivated state and abolishing voltage dependence 27 – 30 . This conformational shift enhances PIP2-binding efficiency and promotes constitutive channel activity. Experimental evidence has shown that PIP2 binding is essential for the function of the KCNQ1/KCNE complex 9 , 31 . Moreover, KCNE1 increases PIP2 sensitivity by 100-fold compared to channels formed by KCNQ1 alone, effectively amplifying channel current 17 , 32 , 33 . These findings underscore the critical role of PIP2 in maintaining the channel’s open state and highlight its functional synergy with KCNE proteins. Moreover, recent structural studies have revealed the spatial organization of PIP2-binding sites within KCNQ1 and their structural interplay with CaM and KCNE3 9, 25 . For example, in the KCNQ1-KCNE3-PIP2 complex of the channel-open state, a key PIP2 binding site has been identified near the loop connecting the S4 transmembrane segment and the S4-S5 linker (Fig. 1 a) 9 . Additionally, a range of biochemical, mutational, and computational studies have shown that PIP2 interacts with multiple binding sites across the KCNQ channel family, including the S0 segment, the S2-S3 linker, the S4-S5 linker, the HA helix, the junction between S6 and HA, and the HA-HB linker 13 , 34 – 39 . Despite these advances, the precise molecular mechanisms underlying PIP2-binding dynamics and its synergistic interactions with KCNE and CaM to activate KCNQ1, remain poorly understood. To address these limitations and provide a more comprehensive understanding of PIP2-KCNQ1 interactions, here, we proposed a novel Stepwise Integrated Multi-scale Dynamics and Advanced Analysis (SIMDA) framework. SIMDA integrates multiple methodologies to address the challenges in studying lipid-channel interactions. By combining coarse-grained molecular dynamics (CG-MD), all-atom molecular dynamics (AA-MD), quantum mechanics (QM), and well-tempered metadynamics (Wt-MetaD), SIMDA enables rapid identification of lipid-binding sites, detailed characterization of lipid-protein interactions, and exploration of lipid association and dissociation pathways. Furthermore, SIMDA employs advanced analytical tools, such as the Uniform Manifold Approximation and Projection (UMAP)-Leiden clustering algorithm, for consistent identification of functionally equivalent binding sites across multimeric assemblies and systems. The incorporation of QM-based Independent Gradient Model based on Hirshfeld partitioning and Atoms in Molecules (IGMH-AIM) analysis offers quantitative insights into binding interactions and thermodynamic properties, providing a holistic view of lipid-channel dynamics. In this study, we utilized our developed SIMDA method to perform over 280 µs of multi-scale molecular dynamics (MD) simulations to explore the binding mechanisms of PIP2 to the KCNQ1 channel in both inactive and active states, with and/or without the auxiliary proteins KCNE3 and CaM. Our results identified eight distinct PIP2-binding sites on KCNQ1. These sites included previously reported regions such as the S2-S3 linker, as well as newly identified sites on the S0 segment, the S4-S5 linker, the HA helix, and the junction between the S6 and HA helix. The presence of KCNE3 and CaM was found to significantly modulate PIP2 binding at these regions. Specifically, KCNE3 enhances the “twist” effect at the C-terminal of KCNQ1 in its active state, increasing PIP2’s binding affinity. Additionally, KCNE3 and CaM facilitate PIP2 binding to a deeper region at the junction between S6 and the HA helix, triggering the tightening of “loose links”. Finally, our simulations identified eight dissociation pathways of PIP2 from KCNQ1. Along these pathways, PIP2 undergoes transitions across the identified binding sites, revealing the transfer dynamics of PIP2 between KCNQ1 binding sites and offering critical insights into its interaction mechanisms with KCNQ1. RESULTS Binding of PIP2 to KCNQ1 via multi-scale MD simulation CG-MD simulations provide an efficient approach to identifying potential lipid-binding sites by simplifying molecular complexity while retaining key interactions 40 – 43 . To comprehensively explore the possible PIP2 binding sites on the KCNQ1 channel in different functional states-including both the inactive and active states, with or without the auxiliary proteins KCNE3 and CaM, which modulate channel function, we performed a total of 260 µs CG-MD simulations. The simulations encompassed eight KCNQ1 systems in both inactive (KCNQ1, KCNQ1-KCNE3, KCNQ1-CaM, KCNQ1-KCNE3-CaM) and active states. Each system was simulated in at least five parallel runs, each lasting 5 µs. The simulations were performed on a fluid POPC membrane with 5% PIP2. Root-mean-square-deviation (RMSD) calculations were performed for each simulation confirming that the systems reached equilibrium after 1 µs of simulation (Fig. S1 ). The final 4 µs trajectories of each simulation were merged for analysis. To identify the interactions between PIP2 and KCNQ1, we calculated 2D-density maps in the plane of the bilayer (XY) to visualize the distribution of PIP2 related to the different KCNQ1 states. As shown in Fig. 1 c, d, these density maps revealed PIP2 binding hotspots around the KCNQ1, particularly near regions S3-S4 and S4-S5 linker, which were consistently observed in cryo-EM structures of KCNQ1 in complex with PIP2 9 . Moreover, the presence of KCNE3 and CaM significantly enhanced the PIP2 binding patterns, enhancing hotspots toward regions around S4-S5 linker, S5 and S0-S1 segment, while in the absence of KCNE3 and CaM, PIP2 predominantly interacted with S2-S3 linker. These results highlight the modulatory effects of KCNE3 and CaM on PIP2 binding, suggesting that these auxiliary proteins may stabilize specific binding sites and thereby influence the functional dynamics of the KCNQ1 channel. To identify key PIP2-binding sites relevant to different functional states of KCNQ1, clustering and ranking analyses were performed. The PIP2 binding sites in each system were screened based on several key parameters, including occupancy, surface area, and duration (Fig.S2). As shown in Fig.S2, each KCNQ1 functional state ultimately contained 10–12 categories of PIP2 binding sites, reflecting the diversity of PIP2 interactions within the KCNQ1 channel. In order to precisely assess the binding stability of the identified PIP2 binding sites, we extracted representative CG structures for each binding site in each KCNQ1 functional state system, converted them into atomistic models, and performed 100 ns AA-MD simulations. This resulted in 79 PIP2 binding site simulations, totaling 7.9 µs of AA-MD simulation. As shown in Fig. S3/S4, most binding sites exhibited minimal RMSD fluctuations, with lipid and protein RMSD values below 3.5Å and 8Å, respectively, indicating stable PIP2 binding and good protein structural stability. However, some PIP2 binding sites, such as Bsid9 in the active KCNQ1-KCNE3-CaM system, showed significant RMSD fluctuations (Fig. 1 e), suggesting unstable PIP2 binding. Therefore, to screen out unstable binding sites, we excluded binding sites with RMSD fluctuations of more than 1 Å in the last 10 ns. Through this AA-MD simulation-based screening, we identified a total of 68 stable PIP2 binding sites across the different KCNQ1 functional states. The identified binding sites within each KCNQ1 functional state system reflected distinct binding locations but did not account for differences between systems. Additionally, since KCNQ1 is a homotetramer composed of four identical subunits, the binding of PIP2 to one subunit is equivalent to its binding to the others. To avoid misinterpreting symmetric binding sites as distinct regions, we performed UMAP-Leiden dimensionality reduction and clustering analysis. UMAP-Leiden cluster is a combined workflow where UMAP reduces high-dimensional data into a low-dimensional space for intuitive visualization, and the Leiden algorithm clusters the data points into well-connected groups, ensuring faster, more accurate, and robust community detection 44 , 45 . This method enabled the classification and consolidation of binding sites across systems, ultimately clustering 8 distinct binding sites (namely C0 to C7) shared among all systems from a total of 68 all-atom screened PIP2-binding sites. The distribution of these shared binding sites across different systems is presented in Fig. 2 a. These binding sites interact with various regions of KCNQ1 and/or KCNE3 and CaM (Fig. 2 b), and can be classified into two distinct group based on their spatial distribution with the C-terminal domain of KCNQ1. The first group including sites C0, C1, C2 and C7, is located on the one side of the C-terminal domain, near the S2-S3 and S3-S4 linker regions. Within this group, C0, C1, and C2 interact with the S3 segment, S2-S3 linker, and S4-S5 linker, while C1 also binds to the C-terminal of KCNE3. Notably, C7 primarily associates with the S2-S3 linker and the S3 segment. The second group, comprising site C3, C4, C5 and C6, is distributed on the opposite side of the C-terminal domain, near the S2-S3 and S3-S4 linker regions.C4, C5, and C6 interact with the S0-S1 loop and the HA/HB regions, with C5 and C6 also interacting with CaM. C3 primarily interacts with the S1 segment and the S4-S5 linker. We further analyzed the distribution characteristics of these binding sites in each system (Fig. 2 c). In the inactive and active KCNQ1-only systems, PIP2 binds to the same sites on both sides of the C-terminal domain, specifically C0, C4, and C7. However, upon the binding of KCNE3, significant redistribution of PIP2 binding sites is observed. On the opposite side of the C-terminal domain, C4 remains unchanged in both states, while a new binding site, C3, emerges. On the C-terminal side, the binding sites shift from C0 to C1, and C7 disappears. When CaM binds to KCNQ1, further state-dependent changes in PIP2 binding sites occur. On the C-terminal side, the binding site shifts from C0 to C2 in the inactive state, whereas C0 is retained in the active state. On the opposite side, C4 shifts to C5 in the inactive state, while in the active state, part of the binding remains at C4, but the majority shifts to C6. Additionally, a new binding site, C3, appears on the opposite side in both states. In systems where both KCNE3 and CaM bind to KCNQ1, PIP2 binding sites on both sides of the C-terminal domain exhibit further redistribution. In the inactive state, PIP2 interacts with C2 on the C-terminal side and with C5 and C3 on the opposite side. In the active state, the binding sites shift to C1 on the C-terminal side and to C6 and C3 on the opposite side. These findings suggest that KCNE3 and CaM regulate PIP2 interactions in distinct but complementary ways. KCNE3 primarily modifies PIP2 binding on the C-terminal side, favoring C1 over C0. In contrast, CaM modifies PIP2 binding in a state-dependent manner: it favors C2 over C0 in the inactive state and influences PIP2 redistribution on the opposite side of the C-terminal domain, stabilizing interactions at C5 in the inactive state and at C6 in the active state. Additionally, C7 is unique to the KCNQ1-only system and is eliminated in the presence of KCNE3 or CaM, which instead contribute to the emergence of C3 as a new PIP2 binding site absent in the KCNQ1-only system. Together, KCNE3 and CaM fine-tune PIP2 interactions, likely contributing to the dynamic regulation of KCNQ1 channel function in a state-dependent and cooperative manner. Qualitative and Quantitative Interaction Analysis of PIP2 through QM Calculations To investigate the underlying determinants of the 8 identified PIP2 binding sites and their distinct distributions, we first quantified PIP2-protein contact number and frequencies across all binding site clusters using AA-MD simulations (Fig. S5/S6 & Fig. 3 ). This analysis provided an overview of PIP2 interaction profiles at each binding site. We then selected the top 10 interacting residues from each cluster for quantum mechanical (QM) calculations, which offer greater precision than AA-MD and CG-MD simulations. Following convergence, we applied the Independent Gradient Model based on Hirshfeld partitioning and Atoms in Molecules (IGMH-AIM) analysis to identify key interaction hotspots (Fig. 3 ). In C0, residues such as ARG13, LYS93, and ARG89 formed strong hydrogen bonds with PIP2, with bond critical points (BCPs) between 0.033–0.046 (Fig. 3 a). In C1, KCNE3’s LYS35 formed the strongest hydrogen bond (BCP: 0.062), reducing the interaction between ARG13 and PIP2 in the S1 region (Fig. 3 b). In C2, ARG89 exhibited the strongest interaction (BCP: 0.121), followed by LYS93 and ARG92 (Fig. 3 c). These interaction patterns highlight ARG146, LYS93, ARG92, and ARG89 as consistently critical residues for PIP2 binding across C0, C1, and C2, corroborating previous experimental findings 9 , 46 . The observed differences in binding modes and interaction strengths across these clusters reflect the unique roles of these residues within distinct structural environments. Notably, C1 and C2 show enhanced interactions between PIP2 and residues in the S3/S2-S3 linker region, suggesting that the presence of KCNE3 or CaM promotes closer and more stable binding interactions in this region. Moreover, in C2 (Fig. 3 d), key PIP2-binding residues are concentrated in the S4-S5 linker and S1 regions. ARG156 and PHE153 in the S4-S5 linker form the strongest hydrogen bonds with PIP2 (BCPs: 0.045 and 0.026), while ARG31 and ARG29 contribute weaker bonds (BCPs: 0.026 and 0.020). These findings highlight the importance of both the S4-S5 linker and S1 regions in stabilizing PIP2 interactions within this cluster. In C7, PIP2 binding is primarily concentrated in the S2-S3 linker/S3 region, with key residues LYS80, ARG89, TYR81, and LYS93 contributing to interactions (Fig. 3 h). LYS80 forms the strongest hydrogen bond with PIP2 (BCP: 0.062), highlighting its critical role in binding. Although C7 shows fewer interacting regions compared to other clusters, the strong interactions within the S2-S3 linker emphasize its importance in stabilizing PIP2 binding. In C3, ARG156 (BCP: 0.045) and PHE153 (BCP: 0.026) in the S4-S5 linker region are identified as the most critical residues, forming strong hydrogen bonds with PIP2. Additionally, the involvement of ARG29 (BCP: 0.020) and ARG31(BCP: 0.026) further underscores the importance of both the S1 and S4-S5 linker regions in PIP2 binding within PIP2 at C3 binding sites (Fig. 3 d). Binding sites C4, C5, and C6 displayed distinct patterns. In C4, residues in the HA and S0-S1 regions (e.g., ARG156, ARG6) formed neutral hydrogen bonds (BCP: 0.015–0.034) (Fig. 3 e). In C5, interactions focused on the S0-S2 region, with LYS71 forming the strongest bond (BCP: 0.048) (Fig. 3 f). In C6, residues in the CaM region, such as ARG31 and ARG33, exhibited strong interactions with PIP2 (BCP: 0.028–0.049) (Fig. 3 g). Finally, in C7, LYS80 formed the strongest bond (BCP: 0.062), followed by ARG89 and TYR81 (Fig. 3 h). Across clusters C4, C5, and C6, ARG6, LYS18, and TRP17 are consistently critical for PIP2 binding. Notably, In C6, ARG31 and ARG33 in the CaM region play crucial roles in PIP2 interactions, emphasizing the importance of CaM in regulating binding dynamics. While C5 shows stronger overall binding than C4, its interactions are more localized in the S1 region, whereas C4 features broader interactions across the S0, S1, and HA regions. These findings highlight the regulatory role of CaM in PIP2 binding, offering insights into how lipid-protein interactions impact membrane protein function and providing a basis for further research and therapeutic development. The Impact of KCNE3 and CaM on the “Twist and Loose Links” Mechanism Unbiased CG-MD simulations are limited in their ability to capture detailed atomic-level interactions 47 – 51 , while AA-MD simulations, despite providing accurate descriptions of such interactions, are insufficient for comprehensively characterizing the kinetics and thermodynamics of lipid-protein binding and unbinding processes 43 , 52 , 53 . To overcome these limitations, we utilized well-tempered metadynamics (Wt-metaD) simulations based on AA-MD simulation results, to systematically investigate the dissociation pathways and binding preferences across distinct PIP2 binding sites. A total of > 10 µs of Wt-metaD simulations (> 100 ns per binding site) were conducted, focusing on the 8 distinct PIP2 binding sites identified through UMAP-Leiden dimensionality reduction and clustering analysis. In each Wt-metaD simulation, two collective variables (CVs) were defined to enhance sampling. Specifically, the X and Y components of the distances between each lipid and surrounding residues within 10 Å were selected as CVs. This choice of CVs enhances precision in capturing local interactions, simplifies the representation of complex systems, and improves the exploration of binding landscapes, providing detailed insights into lipid-protein interaction dynamics. For each binding site, PIP2 dissociates along its respective CV, following similar or distinct pathways. This process enables the sampling of unique Free Energy Surfaces (FES), which represent the energy landscape of the system as a function of the CVs, highlighting the energetic barriers and stable states during dissociation. Additionally, the Minimum Free Energy Path (MFEP), which traces the most favorable energetic route across the FES, provides a detailed view of the preferred dissociation pathway for PIP2 at each binding site. To explore the preferences among different dissociation paths, we performed kinetic and thermodynamic analyses. Specifically, kinetic trajectory analysis from Wt-metaD simulations revealed that binding sites C0, C1, and C2 follow the same dissociation pathway (Path1) (Fig.S7). By identifying the Minimum Free Energy Path (MFEP) from the corresponding Free Energy Surfaces (FES) (Fig. 4 a, d, g), the free energy profiles were obtained, with dissociation free energy values of 58.01 kJ/mol, 110.27 kJ/mol, and 77.72 kJ/mol for C0, C1, and C2, respectively. To identify the source of these free energy differences, we conducted dynamic analyses of C0, C1, and C2 (for example, Fig. 4 b,e). Each dissociation trajectory was classified into four states (A-D), corresponding to the minimal energy states on the FES. These states illustrate the proportions of each binding state along the dissociation trajectory. Interaction analysis revealed significant conformational changes in the S2-S3 linker across C0, C1, and C2. Notably, C1 exhibited prominent changes in the S2 and S3 regions, as well as in KCNE3. Although C2 showed smaller fluctuations compared to C1, they were more pronounced than in C0. The presence of KCNE3 enhanced the affinity of PIP2 for C1, with the conformational changes of KCNE3 closely tied to the binding and dissociation process. This may be due to the strong hydrogen bond interactions formed between LYS35 and ARG36 at the KCNE3 tail and PIP2. Here, we clearly observed the C-terminal “twist effect” induced by PIP2 binding, along with significant conformational changes in the S2-S3 linker region. The presence of KCNE3 intensifies this effect, causing larger twists in other regions, such as S2 and S3. The higher free energy required for PIP2 binding in C2 compared to C0 may result from the tighter hydrogen bond interactions between PIP2 and the residues in the S3/S2-S3 linker region in C2. These stronger interactions lead to greater fluctuations in the C-terminal region of C2, primarily involving the S2-S3 linker, S2, and S3. The FES analyses of the dissociation pathways of PIP2 at the binding sites C4, C5, and C6 revealed that the primary dissociation paths were Path 7 and Path 8, with Path 8 being dominant (Fig.S7 & Fig.S8e-h). The free energy calculations showed values of approximately 100 kJ/mol at C4, 49.16 kJ/mol at C5, and 79.96 kJ/mol for C6. Notably, C5 was only present in the inactive state of KCNQ1 with CaM, while C6 appeared exclusively in the active state of KCNQ1 with CaM. These findings suggest that CaM in the active state of KCNQ1 exerts a competitive attraction, lowering the free energy data comparing with non-CaM system. This phenomenon is consistent with the interaction analysis results. In C6, the CaM in the active KCNQ1 systems forms hydrogen bond interactions with PIP2 through ARG31 and ARG33, attracting PIP2 binding and competing with the PIP2 bound in the S1 region of KCNQ1, which may explain the decrease in PIP2 binding affinity. In C5, although the CaM in the inactive KCNQ1 systems makes contact with PIP2, it does not form hydrogen bond interactions. As a result, while CaM attracts PIP2, it fails to establish more stable interactions, leading to a further reduction in binding affinity. C3 serves as a highly dynamic binding site for PIP2. Due to its deeply buried location within the protein structure (Fig. 2 b), PIP2 tends to rebind to other sites-specifically C0, C4, and C7 during its dissociation along Path 5 and Path 6 (Fig. 5 a). This propensity for rebinding induces substantial conformational rearrangements in the VSD region. Moreover, FES analyses, as illustrated in Fig.S9, reveal that the dissociation of PIP2 from C3 requires a high free energy input exceeding 95 kJ/mol. These complex dissociation characteristics activate multiple other binding sites, including C0, C4, and C7, thereby influencing the protein’s structural dynamics (Fig. 5 ). Community network analysis of key stable binding sites further elucidates this phenomenon. As PIP2 gradually dissociates from C3, the community coupling network between the VSD and PD regions evolves significantly. Initially, the network comprises 3 communities (Fig. S9a); it then transitions to 3 communities with altered coupling (Fig. S9b), progresses to 4 communities (Fig. S9c), and ultimately expands to 5 distinct communities once PIP2 completely exits the binding pocket communities (Fig. S9d). This evolution indicates that PIP2 binding at C3 brings these regions into closer proximity, intensifying the “loose links” effect and enhancing inter-domain communication. In contrast, C7 functions as a shallow binding site through which several binding pathways converge, exhibiting a relatively straightforward dissociation pathway primarily along Path 3 (Fig. 5 a). Notably, the strong hydrogen bond interactions between PIP2 and residues in the S2-S3 linker/S3 region at C7 lead to significant fluctuations in the S2-S3 linker. These fluctuations are critical as they trigger the C-terminal “twist” effect, a conformational change that may influence the protein’s gating mechanisms. Dissociation Kinetic Pathways and Binding Site Transfers of PIP2 in KCNQ1 To further investigate the kinetic behavior of binding and dissociation at these sites, we analyzed the dissociation trajectories of PIP2 (Fig. 5 b) and identified 8 representative pathways. PIP2 primarily binds to two sides of the KCNQ1 C-terminal domain. The binding sites on the C-terminal domain can be categorized into deeper binding sites (C0-C2) and a more outward site (C7) on one side, while the opposite side includes another deeper site (C3) and outward sites (C4-C6). We found that PIP2 can directly dissociate from the binding regions C0-C2 (Path 5), C4-C6 (Path 8), and C7 (Path 3). However, dissociation from C3 requires a transfer to either C0-C2 (Path 5, involving the S5-S6, HA, and the C-terminal S0-S4, S4-S5 linker, and KCNE3 regions) or C4-C6 (Path 6, involving the S5-S6, HA/HB, CaM, and KCNE3 regions) before release (Fig. 6 b). This indicates that the dissociation of PIP2 from C3 (S5-S6) is kinetically dependent on intermediate transfer steps, highlighting a unique mechanism whereby C3 serves as a transitional binding site rather than a direct dissociation site. In addition to direct dissociation, PIP2 can also dissociate from C0-C2, C4-C6, and C7 by transitioning through other binding sites. For example, PIP2 can transfer from C0-C2 to C7 before dissociation (Path 1, involving the C-terminal S0-S4, S4-S5 linker, and the KCNE3 region) or to C4-C6 before dissociation (Path 2, involving the C-terminal S0-S4, S4-S5 linker, CaM, and the HA-HB regions). Furthermore, PIP2 can transition between the C4-C6 and C7 binding sites before eventually dissociating from KCNQ1 (Path 4 and Path 5). These findings reveal the complex kinetic pathways and inter-site dynamics of PIP2 binding and dissociation, underscoring the critical role of intermediate states and site transitions in regulating PIP2 interactions with KCNQ1. Building upon this trajectory analysis, we propose that PIP2 binding is primarily associated with five distinct structural domains: the VSD at the C-terminal (S0-S4, S4-S5 linker), the PD (S5-S6), KCNE3, CaM, and the HA-HB region near the membrane (Fig. 5 ). Among these, the VSD and PD regions are closely linked to electromechanical coupling, playing a critical role in PIP2 binding. Specifically, binding at deeper sites, such as C2 and C3, results in the tightening of “loose links”, which enhances the coupling between the voltage-sensor and pore domains. PIP2 binding at these sites is strongly associated with KCNE3, whose presence facilitates PIP2 binding. Furthermore, CaM modulates the binding preference of PIP2, shifting it toward shallower sites like C5/C6, which prevents some PIP2 molecules from reaching deeper binding sites. Overall, PIP2 binding dynamics-whether through direct interactions or transitions between binding sites are significantly influenced by the presence of KCNE3 and CaM. These factors regulate both the accessibility and binding preferences of PIP2 at deep and shallow binding sites, ultimately affecting the overall dynamics of PIP2 interactions within the system. DISCUSSION In this study, we developed SIMDA, an advanced computational framework designed to systematically analyze the complex interactions between membrane proteins and lipids. By integrating stepwise multiscale MD simulations and sophisticated advanced analysis, SIMDA enables precise identification of lipid binding sites, detailed calculation of residue interactions, and comprehensive analysis of dynamics and thermodynamic properties of lipid binding. The framework employs a three-tiered strategy: multiscale simulations from CG-MD to AA-MD simulations and to QM calculations/Wt-metaD simulations; systematic identification and classification of lipid binding sites based on CG-MD and AA-MD simulations; and residue interaction quantification via AA-MD and IGMH-AIM calculations. The incorporation of Umap-Leiden clustering for dimensionality reduction facilitates robust handling of high-dimensional lipid-protein interaction data, ensuring accurate and scalable analyses. PIP2 directly modulates the activity of a wide range of ion channels, including inward rectifier potassium (K + ) channels, voltage-gated calcium channels, transient receptor potential channels, and particularly voltage-gated KCNQ potassium channels (KCNQ1-5, also known as KV7.1-5) 19 , 20 , 22 , 54 – 57 . The binding sites and mechanisms of PIP2 interaction with various KCNQ family members (KCNQ2-4) have been extensively investigated through both experimental approaches and molecular dynamics simulations 58 – 61 . For KCNQ2, Zhang et al. 62 demonstrated that in the closed state, PIP2 is anchored at the S2–S3 loop, while upon channel activation, it interacts with the S4–S5 linker, thereby participating in channel gating. Pant et al. 41 characterized PIP2 binding in human Kv7.2 channels, revealing that in the closed state, PIP2 is localized at the periphery of the voltage-sensing domain (VSD), and in the open state, it binds to four distinct interfaces formed by the cytoplasmic ends of the VSD, the gate, intracellular helices A and B, and their linkers. PIP2 binding induces a bilayer-interacting conformation of helices A and B, as well as coordinated movement between the VSD and the pore domain. Additionally, Zheng et al. 24 identified two PIP2 molecules in the open-state structure of KCNQ4, which function as a bridge to couple the VSD and pore domain, facilitating channel opening. While substantial evidence supports the critical role of PIP2 in the activation and inactivation processes of KCNQ1, only a single PIP2 binding site has been identified 9 , 25 , 63 – 65 . A comprehensive understanding of the full binding mechanism in KCNQ1 remains lacking. Using the SIMDA framework, we explored the interactions between the voltage-gated ion channel KCNQ1 and PIP2, a lipid essential for channel activation. Our analysis covered eight distinct systems involving KCNQ1 in both inactive and active states, with or without the regulatory proteins KCNE3 and CaM. Through CG-MD and AA-MD simulations, combined with Umap-Leiden clustering, we identified eight distinct PIP2 binding clusters (C0-7). These clusters exhibited consistent kinetic and thermodynamic properties across different systems, revealing a conserved binding pattern despite varying regulatory conditions. Our results show that the presence of KCNE3 significantly enhances the C-terminal “twist” effect of KCNQ1, increasing the affinity of PIP2. Key residues LYS35 and ARG36 in in C2 form strong hydrogen bonds with PIP2, inducing significant fluctuations in the KCNE3 tail during PIP2 dissociation. These interactions propagate conformational changes across the S2, S3, and S4-S5 linker regions, creating a higher energy barrier for dissociation. The structural rearrangement suggests that KCNE3 accelerates the C-terminal “twist” effect, thereby stabilizing PIP2 binding and increasing its affinity. In contrast, CaM plays a regulatory role by competing with PIP2 for binding, thereby reducing PIP2 affinity in a state-dependent manner. Our analysis shows that CaM inhibits PIP2 binding at C4, regardless of whether KCNQ1 is in the active or inactive state. However, CaM in the active state of KCNQ1 enhances the strength and complexity of interactions at C6 compared to that in the inactive state of KCNQ1. Further IGMH-AIM calculations confirm that ARG31 and ARG33 in the CaM region play critical roles in PIP2 binding, emphasizing CaM’s significant influence on the binding dynamics of KCNQ1. Our investigation also reveals the unique role of C3 in reflecting the “loose links” effect within the binding network. Key residues ARG156 and PHE153 in C3 form strong hydrogen bonds with PIP2, and the dissociation of PIP2 from C2 often passes through other clusters, such as C0, C1, C5, and C6. This migration gradually tightens previously loose regions of the protein, demonstrating the transition from “loose links” to a more stable structure, which may be critical for the activation of the VSD. Furthermore, C7 plays a crucial role in the C-terminal “twist” effect. Closely associated with the S2-S3 linker, C7 acts as an intermediate site during PIP2 dissociation and facilitates the transition between multiple binding sites, highlighting its previously overlooked regulatory importance. Additionally, our analysis of the dissociation kinetics and binding site transfers of PIP2 in KCNQ1 reveals complex, multi-step dynamics involving both direct dissociation and intermediate site transitions. We identified eight representative dissociation pathways, which involve transitions between deeper and more outward binding sites, highlighting the intricate nature of PIP2 interactions with KCNQ1. Notably, dissociation from certain sites, such as C3, requires intermediate transitions, indicating that these transitional binding sites play a critical role in facilitating dissociation. We propose a model wherein PIP2 binding occurs at five distinct structural domains-namely, the voltage-sensor domain (VSD), the pore domain (PD), KCNE3, CaM, and the HA-HB region near the membrane. This model underscores the electromechanical coupling between these domains, with deeper binding sites such as C2 and C3 facilitating the tightening of “loose links” and thereby enhancing the coupling between the voltage-sensor and pore domains. These findings provide new insights into the molecular mechanisms underlying PIP2 regulation of KCNQ1, revealing the complex interplay between binding and dissociation processes that are essential for the functional dynamics of this ion channel. In summary, our study provides a comprehensive understanding of the complex interactions between KCNQ1 and PIP2, capturing the intricate dynamics we refer to as the “twists, links, and binding site transfers” effect. We find that KCNE3 in the active state of KCNQ1 accelerates the C-terminal twist effect and significantly enhances PIP2 binding affinity, while CaM in the active state of KCNQ1 competitively reduces PIP2 binding by modulating interaction strength and complexity. The observed migration of PIP2 between multiple binding sites suggests that dynamic transitions play a vital role in its regulatory mechanism. Although we captured several intermediate states of PIP2 during these transitions, we did not observe a complete migration of a single PIP2 molecule across all possible binding sites. Future experimental and computational studies can explore these transitions in greater detail to construct a comprehensive landscape of PIP2-mediated regulation of KCNQ1. MD simulations have become essential tools for investigating protein-lipid interactions, providing valuable insights into lipid-binding mechanisms at various scales 40 , 43 , 52 , 66 – 69 . CG-MD simulations, in particular, are widely used due to their ability to efficiently identify potential lipid-binding sites 69 – 71 . For example, Bartocci et al. 72 employed millisecond-scale CG-MD simulations combined with reverse-mapping to all-atom representations, uncovering multiple cannabinoid-binding sites on GlyR and offering an effective strategy for identifying allosteric binding sites. However, CG-MD simulations are limited by their resolution, which is insufficient to capture secondary structural changes or detailed atomic-level interactions, a key drawback when trying to understand the precise mechanisms underlying protein-lipid binding 51 , 52 , 73 . In contrast, AA-MD simulations provide higher-resolution insights into atomic-level conformational changes, enabling a deeper understanding of lipid-protein interactions 41 . Nevertheless, unbiased AA-MD simulations are constrained by time scale limitations, often capturing only equilibrium states close to lipid-binding sites, and are unable to fully sample rare, high-energy events that are critical for understanding dynamic binding processes. To address this, enhanced sampling techniques such as well-tempered metadynamics, umbrella sampling, and Gaussian accelerated molecular dynamics are employed. These techniques allow for the exploration of large-scale conformational transitions and lipid-binding events. For instance, Chan et al. 74 used umbrella sampling simulations to examine binding pockets on the PH domain and generated potential of mean force profiles, elucidating their preference for PIP2 lipids. This approach, combined with orientation analysis and binding pocket studies, provided valuable molecular insights into protein-lipid interactions during membrane remodeling. While CG-MD, AA-MD, and enhanced sampling techniques each have their strengths, they also have limitations, particularly in capturing the full complexity of protein-lipid interactions. To overcome these limitations, we propose the SIMDA method, which integrates these techniques in a stepwise manner. The SIMDA method begins with CG-MD simulations to identify potential lipid-binding sites, followed by AA-MD simulations to capture detailed atomic interactions, and culminates with QM or well-tempered Wt-metaD simulations to probe rare events and electronic interactions with high precision. This progressive multi-scale approach provides a comprehensive view of lipid-protein interactions, enabling the detailed characterization of lipid binding at different time and length scales. Additionally, several tools have been developed to assist with the analysis of protein-lipid binding structures and MD trajectories. Tools like PyLipID, LipIDens, and ProLint are particularly useful for identifying, ranking, and refining lipid binding poses, enhancing the accuracy of lipid-protein interaction studies 75 – 77 . For instance, PyLipID can be used to calculate residence times of lipids at binding sites, further advancing our understanding of the dynamics of protein-lipid interactions 75 . We believe these tools have significantly aided our research. However, these analytical tools do not address the consistent identification of functionally equivalent binding sites across multimeric assemblies and systems, nor do they provide more accurate quantitative interactions. Therefore, the SIMDA method introduces new analysis tools, such as the UMAP-Leiden clustering algorithm, which enables consistent identification of functionally equivalent binding sites across multimeric assemblies and complex systems. This method improves the resolution of binding site patterns in larger, more intricate systems, facilitating more accurate predictions of functional behaviors. The integration of quantum mechanical-based IGMH-AIM analysis further enhances SIMDA by providing quantitative insights into binding interactions and thermodynamic properties, allowing for a holistic view of lipid-channel dynamics that bridges multiple scales of simulation. In conclusion, the SIMDA method systematically captures detailed data on lipid-protein interactions through stepwise multiscale dynamic simulations, employing a progressive strategy that transitions from CG-MD to AA-MD and further to QM or Wt-metaD simulations. This multilayered framework not only comprehensively elucidates complex membrane protein-lipid interactions but also establishes a solid foundation for in-depth analysis. The key advantage of SIMDA lies in its high flexibility and precision, enabling accurate identification of binding sites through its progressive approach while using dimensionality reduction and clustering analysis to uncover functional properties and relationships among binding sites. Furthermore, SIMDA facilitates a seamless transition from qualitative assessments to quantitative studies, allowing detailed investigations of binding strength, interaction modes, and their dynamic changes under various conditions. It also analyzes the thermodynamics and kinetics of binding and dissociation processes, revealing energy barriers, stability, and microscopic regulatory mechanisms. This macro-to-micro analysis provides theoretical support for the design of targeted drugs based on binding sites, while offering insights into membrane protein mechanisms such as lipid binding, structural dynamics, and interaction regulation. In the future, SIMDA can be applied beyond ion channels to investigate the intricate interactions of G protein-coupled receptors (GPCRs) and membrane transporters, uncovering critical regulatory sites and energy transition pathways. These capabilities make SIMDA a powerful tool for studying membrane protein-lipid interactions, providing essential theoretical foundations for developing novel regulatory mechanisms and therapeutic targets, with broad potential applications in biomedicine and drug development. METHODS To obtain comprehensive, precise, and reliable insights into protein-lipid binding sites, as well as the mechanisms by which cofactor modulation influences intricate protein-lipid systems in different states, we conducted extensive molecular dynamics simulations and analyses. The Stepwise Integrated Multi-scale Dynamics and Advanced Analysis (SIMDA) strategy consists of two primary components: the implementation of stepwise multiscale molecular dynamics (MD) simulations and the integration of advanced analytical methods (Fig. 6 ). First, during the stepwise multiscale MD simulation process, our strategy begun by using coarse-grained molecular dynamics (CG-MD) simulations to efficiently and comprehensively determine how lipids bind to proteins 52 , 78 . From the first round of screening based on CG-MD simulations, lipid-protein complexes for each system were saved and converted into all-atom models. In the second round of screening, all-atom molecular dynamics (AA-MD) simulations was employed to analyze the binding details at each site on the atomic level 51 , 79 , 80 . Subsequently, we used the X and Y components of the distances between each lipid and the surrounding residues within 10 Å as collective variables (CVs) for enhanced sampling via well-tempered metadynamics (Wt-metaD), based on AA-MD results, to explore lipid binding and dissociation processes at each site 81 – 83 . Key residues involved in lipid binding are identified from AA-MD simulations, followed by quantum mechanics (QM) simulations for a deeper analysis 84 – 86 . This stepwise multiscale MD simulation approach is designed to reveal critical binding information across scales, from large to small, from coarse to fine, and from qualitative to quantitative, all in a highly efficient and precise manner. Moreover, this method maximizes the strengths of CG-MD, AA-MD/Wt-metaD, and QM while mitigating their individual limitations, providing a comprehensive and systematic approach to studying lipid-protein interactions (Fig. 6 b). Next, based on the above stepwise multiscale MD simulation results, we conducted a series of advanced analyses. First, we performed a two-step screening process for the binding sites. The initial screening was based on CG-MD results, using the PyLipID package to identify all potential binding sites across the eight systems 87 . For each binding site, we performed analyses of occupancy, surface area and duration, followed by ranking to select the top-ranked binding modes. These binding modes were then converted into all-atom models for AA-MD simulations, where we calculated the RMSD of the lipids at each binding site. Using these RMSD values, we performed a second screening to identify the stably bound sites (Fig. 6 c). While the previous two screenings focused on vertical analyses within each system, we also conducted a horizontal comparison to explore the differences in lipid binding across systems and assess the effects of different states and cofactors. To achieve this, we applied dimensionality reduction and clustering analysis on the interaction data from all systems’ binding sites. Specifically, we used Uniform Manifold Approximation and Projection (UMAP) for dimensionality reduction and Leiden clustering for grouping 88 . UMAP is a nonlinear dimensionality reduction technique that maps high-dimensional data into low-dimensional space, usually in two or three dimensions, while preserving local data structures for more intuitive visualization 44 , 45 . Leiden is a community detection algorithm that optimizes modularity to cluster similar data points into the same group 44 , 45 . Compared to other clustering algorithms, Leiden is faster, more accurate, and ensures the connectivity of each cluster, avoiding fragmentation 89 . First, we applied UMAP to map the interaction data of binding site residues from all systems into a low-dimensional space, allowing the similarities between binding sites to be visualized more intuitively. We then applied Leiden clustering to the reduced data, grouping similar binding sites into distinct clusters. This approach effectively identified subgroups of binding sites, capturing biological heterogeneity in the data and providing insights into the types and states of binding sites (Fig. 6 c). To further investigate the key interaction patterns of the screened binding sites, we performed a stepwise analysis of protein-lipid interactions of contact number analysis based on AA-MD to Hirshfeld partitioning and Atoms in Molecules (IGMH-AIM) analysis analysis based on QM simulations, transitioning from qualitative to quantitative insights 90 . This comprehensive multiscale approach provided reliable interaction information, revealing key residues closely associated with lipid binding. Finally, based on the Wt-metaD trajectories of lipid binding and dissociation, we examined the thermodynamic and kinetic differences of lipid binding at different sites, uncovering the dynamic effects of protein states and cofactors on lipid binding (Fig. 6 c) 91 , 92 . In summary, the SIMDA method systematically investigates lipid-protein interactions through a progressive multiscale MD simulation framework, transitioning from CG-MD to AA-MD, and further to QM or Wt-metaD simulations. This approach captures both macro-level structural features and micro-level interaction dynamics, providing a robust foundation for in-depth analysis. It follows four key steps: identifying and screening the most stable and functional binding sites using progressively refined simulations for precision and reliability; applying dimensionality reduction and clustering analysis to classify binding site characteristics and reveal functional relationships; transitioning from qualitative to quantitative analysis to explore binding strength, interaction modes, and dynamic changes under varying conditions; and analyzing thermodynamics and kinetics to uncover energy barriers, stability, regulatory mechanisms, and microscopic kinetic behaviors. By integrating multiscale MD simulations, binding site identification, and interaction analysis, SIMDA comprehensively elucidates PIP2 binding sites, state-dependent interactions, and cofactor modulation in complex membrane proteins. Beyond structural analysis, it quantifies thermodynamic and kinetic properties, including energy transformations, interaction patterns, and dissociation pathways, while identifying regulatory sites and demonstrating how cofactors like PIP2 modulate membrane protein function. SIMDA’s flexibility across simulation scales ensures seamless integration of macro-structural insights with micro-dynamic analyses, offering a holistic understanding of membrane protein-lipid interactions. This strategy not only reveals energy distributions and kinetic pathways but also provides actionable insights for designing targeted drugs and exploring lipid-binding mechanisms, membrane protein dynamics, and interaction regulation, establishing SIMDA as a powerful tool for biomedical research and therapeutic development. Declarations Data Availability The initial coordinates for the active and inactive structures of KCNQ1 were retrieved from the RCSB Protein Data Bank with accession codes 6v01 and 6v00, respectively. All CG-MD, AA-MD, and WT-metaD were performed using Gromacs 2021.4. QM calculations and IGMH-AIM analysis were conducted using ORCA software. PIP2 binding site identification, screening, classification, and UMAP-Leiden clustering were carried out using Python (version 3.10). Structural visualization was done with PyMOL (version 2.5) and VMD (version 1.9.4). The source data and parameter settings for the multi-scale molecular dynamics simulations have been deposited in Zenodo (https://zenodo.org/records/14885900). Detailed methods, along with additional data analysis and figures, are provided in the Supplementary Information. Code Availability All code of SIMDA framework are freely available on GitHub at https://github.com/xiaoxiaowang1201/SIMDA-framework and on Zenodo at https://zenodo.org/records/14885900 with an MIT license. Acknowledgments This work was supported by the Science and Technology Development Fund, Macau, SAR (No. 0091/2022/A2) and Macao Polytechnic University (No. RP/FCA-02/2023) Author Contributions L. W. contributed to the main ideas, coding, data analysis, and writing of the manuscript. S. L. contributed to the initial modeling, data analysis, and manuscript revisions. Y. Z. and H. S. helped review the paper and provided suggestions for revisions. Q. L., W. Z., X. L., X. Y., and H. T. participated in discussions on the method implementation. H. Liu provided computational resources and the initial concept for the paper. Competing Interests The authors declare no competing interest. References Barhanin J, Lesage F, Guillemare E, Fink M, Lazdunski M, Romey GJ (1996) N., KvLQT1 and IsK (minK) proteins associate to form the I Ks cardiac potassium current. 384:78–80 Sanguinetti MC, Curran ME, Zou A, Shen J, Specter P, Atkinson D, Keating MJN (1996) Coassembly of KvLQT1 and minK (IsK) proteins to form cardiac I Ks potassium channel. 384:80–83 Warth R, Barhanin JJT (2003) J. o. m. b., Function of K + channels in the intestinal epithelium. 193, 67–78 Zheng W, Verlander JW, Lynch IJ, Cash M, Shao J, Stow LR, Cain BD, Weiner ID, Wall SM, Wingo CS (2007) J. A. J. o. P.-R. P., Cellular distribution of the potassium channel KCNQ1 in normal mouse kidney. 292:F456–F466 Al-Hazza A, Linley J, Aziz Q, Hunter M, Sandle GJB (2016) communications, b. r., Upregulation of basolateral small conductance potassium channels (KCNQ1/KCNE3) in ulcerative colitis. 470:473–478 Abbott GW, Tai K-K, Neverisky DL, Hansler A, Hu Z, Roepke TK, Lerner DJ, Chen Q, Liu L, Zupan BJS (2014) s., KCNQ1, KCNE2, and Na+-coupled solute transporters form reciprocally regulating complexes that affect neuronal excitability. 7:ra22–ra22 McCrossan ZA, Abbott GW (2004) The MinK-related peptides. Neuropharmacology 47:787–821 Kirchhofer A, Helma J, Schmidthals K, Frauer C, Cui S, Karcher A, Pellis M, Muyldermans S, Casas-Delucchi CS, Cardoso MC, Leonhardt H, Hopfner K-P, Rothbauer U (2010) Modulation of protein properties in living cells using nanobodies. Nat Struct Mol Biol 17:133–138 Sun J, MacKinnon R (2020) Structural Basis of Human KCNQ1 Modulation and Gating. Cell 180:340–347e9 Barro-Soria R, Perez ME, Larsson HP (2015) KCNE3 acts by promoting voltage sensor activation in KCNQ1. Proc Natl Acad Sci U S A 112:E7286–E7292 Schroeder BC, Waldegger S, Fehr S, Bleich M, Warth R, Greger R, Jentsch TJ (2000) A constitutively open potassium channel formed by KCNQ1 and KCNE3. Nature 403:196–199 Barro-Soria R, Ramentol R, Liin SI, Perez ME, Kass RS, Larsson H (2017) P., KCNE1 and KCNE3 modulate KCNQ1 channels by affecting different gating transitions. Proc Natl Acad Sci U S A 114:E7367–E7376 Sun J, MacKinnon R (2017) Cryo-EM Structure of a KCNQ1/CaM Complex Reveals Insights into Congenital Long QT Syndrome. Cell 169:1042–1050e9 Abbott GW, Tai KK, Neverisky DL, Hansler A, Hu Z, Roepke TK, Lerner DJ, Chen Q, Liu L, Zupan B, Toth M, Haynes R, Huang X, Demirbas D, Buccafusca R, Gross SS, Kanda VA, Berry GT (2014) KCNQ1, KCNE2, and Na+-coupled solute transporters form reciprocally regulating complexes that affect neuronal excitability. Sci Signal 7:ra22 Ghosh S, Nunziato DA, Pitt GS (2006) KCNQ1 assembly and function is blocked by long-QT syndrome mutations that disrupt interaction with calmodulin. Circ Res 98:1048–1054 Kasimova MA, Zaydman MA, Cui J, Tarek M (2015) PIP₂-dependent coupling is prominent in Kv7.1 due to weakened interactions between S4-S5 and S6. Sci Rep 5:7474 Li Y, Zaydman MA, Wu D, Shi J, Guan M, Virgin-Downey B, Cui J (2011) KCNE1 enhances phosphatidylinositol 4,5-bisphosphate (PIP2) sensitivity of IKs to modulate channel activity. Proc Natl Acad Sci U S A 108:9095–9100 Loussouarn G, Park KH, Bellocq C, Baró I, Charpentier F, Escande D (2003) Phosphatidylinositol-4,5-bisphosphate, PIP2, controls KCNQ1/KCNE1 voltage-gated potassium channels: a functional homology between voltage-gated and inward rectifier K + channels. Embo j 22:5412–5421 Zhang H, He C, Yan X, Mirshahi T, Logothetis DE (1999) Activation of inwardly rectifying K + channels by distinct PtdIns(4,5)P2 interactions. Nat Cell Biol 1:183–188 Chuang HH, Prescott ED, Kong H, Shields S, Jordt SE, Basbaum AI, Chao MV, Julius D (2001) Bradykinin and nerve growth factor release the capsaicin receptor from PtdIns(4,5)P2-mediated inhibition. Nature 411:957–962 Rodríguez-Menchaca AA, Adney SK, Zhou L, Logothetis DE (2012) Dual Regulation of Voltage-Sensitive Ion Channels by PIP(2). Front Pharmacol 3:170 Delmas P, Coste B, Gamper N, Shapiro MS (2005) Phosphoinositide lipid second messengers: new paradigms for calcium channel modulation. Neuron 47:179–182 Suh BC, Leal K, Hille B (2010) Modulation of high-voltage activated Ca(2+) channels by membrane phosphatidylinositol 4,5-bisphosphate. Neuron 67:224–238 Zheng Y, Liu H, Chen Y, Dong S, Wang F, Wang S, Li GL, Shu Y, Xu F (2022) Structural insights into the lipid and ligand regulation of a human neuronal KCNQ channel. Neuron 110:237–247e4 Willegems K, Eldstrom J, Kyriakis E, Ataei F, Sahakyan H, Dou Y, Russo S, Van Petegem F, Fedida D (2022) Structural and electrophysiological basis for the modulation of KCNQ1 channel currents by ML277. Nat Commun 13:3760 Mandala VS, MacKinnon R (2023) The membrane electric field regulates the PIP(2)-binding site to gate the KCNQ1 channel. Proc Natl Acad Sci U S A 120:e2301985120 Sun J, MacKinnon R (2020) Structural Basis of Human KCNQ1 Modulation and Gating. Cell 180:340–347e9 Chen H, Kim LA, Rajan S, Xu S, Goldstein SAJ (2003) N., Charybdotoxin binding in the IKs pore demonstrates two MinK subunits in each channel complex. 40:15–23 Morin TJ, Kobertz WR (2008) J. P. o. t. N. A. o. S., Counting membrane-embedded KCNE β-subunits in functioning K + channel complexes. 105, 1478–1482 Wu X, Perez ME, Noskov SY, Larsson HP (2021) A general mechanism of KCNE1 modulation of KCNQ1 channels involving non-canonical VSD-PD coupling. Commun Biol 4:887 Kuenze G, Vanoye CG, Desai RR, Adusumilli S, Brewer KR, Woods H, McDonald EF, Sanders CR, George AL, Jr.;, Meiler J (2020) Allosteric mechanism for KCNE1 modulation of KCNQ1 potassium channel activation. Elife 9 Rodriguez-Menchaca AA, Adney SK, Tang QY, Meng XY, Rosenhouse-Dantsker A, Cui M, Logothetis DE (2012) PIP2 controls voltage-sensor movement and pore opening of Kv channels through the S4-S5 linker. Proc Natl Acad Sci U S A 109:E2399–E2408 Xiao J, Zhen XG, Yang J (2003) Localization of PIP2 activation gate in inward rectifier K + channels. Nat Neurosci 6:811–818 Zhang H, Craciun LC, Mirshahi T, Rohács T, Lopes CM, Jin T, Logothetis DE (2003) PIP(2) activates KCNQ channels, and its hydrolysis underlies receptor-mediated inhibition of M currents. Neuron 37:963–975 Robbins J (2001) KCNQ potassium channels: physiology, pathophysiology, and pharmacology. Pharmacol Ther 90:1–19 Choveau FS, De la Rosa V, Bierbower SM, Hernandez CC, Shapiro MS (2018) Phosphatidylinositol 4,5-bisphosphate (PIP(2)) regulates KCNQ3 K(+) channels by interacting with four cytoplasmic channel domains. J Biol Chem 293:19411–19428 Tobelaim WS, Dvir M, Lebel G, Cui M, Buki T, Peretz A, Marom M, Haitin Y, Logothetis DE, Hirsch JA, Attali B (2017) Competition of calcified calmodulin N lobe and PIP2 to an LQT mutation site in Kv7.1 channel. Proc Natl Acad Sci U S A 114:E869–e878 Liu Y, Xu X, Gao J, Naffaa MM, Liang H, Shi J, Wang HZ, Yang ND, Hou P, Zhao W, White KM, Kong W, Dou A, Cui A, Zhang G, Cohen IS, Zou X, Cui J (2020) A PIP(2) substitute mediates voltage sensor-pore coupling in KCNQ activation. Commun Biol 3:385 Zhang Q, Zhou P, Chen Z, Li M, Jiang H, Gao Z, Yang H (2013) Dynamic PIP2 interactions with voltage sensor elements contribute to KCNQ2 channel gating. 110:20093–20098 Wang Q, Corey RA, Hedger G, Aryal P, Grieben M, Nasrallah C, Baronina A, Pike ACW, Shi J, Carpenter EP, Sansom MS (2020) P., Lipid Interactions of a Ciliary Membrane TRP Channel: Simulation and Structural Studies of Polycystin-2. Structure 28, 169–184 e5 Pant S, Zhang J, Kim EC, Lam K, Chung HJ, Tajkhorshid E (2021) PIP2-dependent coupling of voltage sensor and pore domains in Kv7.2 channel. Commun Biol 4:1189 Naughton FB, Kalli AC, Sansom MS (2016) Association of Peripheral Membrane Proteins with Membranes: Free Energy of Binding of GRP1 PH Domain with Phosphatidylinositol Phosphate-Containing Model Bilayers. J Phys Chem Lett 7:1219–1224 Muller MP, Jiang T, Sun C, Lihan M, Pant S, Mahinthichaichan P, Trifan A, Tajkhorshid E (2019) Characterization of Lipid-Protein Interactions and Lipid-Mediated Modulation of Membrane Protein Function through Molecular Simulation. Chem Rev 119:6086–6161 McInnes L, Healy J, Melville J (2018) J. a. p. a., Umap: Uniform manifold approximation and projection for dimension reduction Ghojogh B, Ghodsi A, Karray F, Crowley M (2021) J. a. p. a., Uniform manifold approximation and projection (UMAP) and its variants: tutorial and survey Kasimova MA, Zaydman MA, Cui J, Tarek M (2015) PIP(2)-dependent coupling is prominent in Kv7.1 due to weakened interactions between S4-S5 and S6. Sci Rep 5:7474 Ingólfsson HI, Lopez CA, Uusitalo JJ, de Jong DH, Gopal SM, Periole X, Marrink SJ (2014) The power of coarse graining in biomolecular simulations. 4:225–248 Wang L, Zhang Q, Tong HHY, Yao X, Liu H, Li G (2024) Computational methods for unlocking the secrets of potassium channels: Structure, mechanism, and drug design. 14, e1704 Marrink SJ, de Vries AH, Mark AE (2004) Coarse Grained Model for Semiquantitative Lipid Simulations. J Phys Chem B 108:750–760 Marrink SJ, Risselada HJ, Yefimov S, Tieleman DP, de Vries AH (2007) The MARTINI Force Field: Coarse Grained Model for Biomolecular Simulations. J Phys Chem B 111:7812–7824 Enkavi G, Javanainen M, Kulig W, Rog T, Vattulainen I (2019) Multiscale Simulations of Biological Membranes: The Challenge To Understand Biological Phenomena in a Living Substance. Chem Rev 119:5607–5774 Corradi V, Sejdiu BI, Mesa-Galloso H, Abdizadeh H, Noskov SY, Marrink SJ, Tieleman DP (2019) Emerging Diversity in Lipid-Protein Interactions. Chem Rev 119:5775–5848 Marrink SJ, Corradi V, Souza PCT, Ingolfsson HI, Tieleman DP, Sansom MS (2019) P., Computational Modeling of Realistic Cell Membranes. Chem Rev 119:6184–6226 Suh BC, Hille B (2008) PIP2 is a necessary cofactor for ion channel function: how and why? Annu Rev Biophys 37:175–195 Czech MP (2000) PIP2 and PIP3: complex roles at the cell surface. Cell 100:603–606 Rodríguez Menchaca AA, Adney SK, Zhou L, Logothetis DE (2012) Dual Regulation of Voltage-Sensitive Ion Channels by PIP2. 3 Suh B-C, Leal K, Hille B (2010) Modulation of High-Voltage Activated Ca 2+ Channels by Membrane Phosphatidylinositol 4,5-Bisphosphate. Neuron 67:224–238 Kim RY, Pless SA, Kurata HT (2017) PIP2 mediates functional coupling and pharmacology of neuronal KCNQ channels. 114:E9702–E9711 Mandala VS, MacKinnon R (2023) The membrane electric field regulates the PIP 2 -binding site to gate the KCNQ1 channel. 120, e2301985120 Kim H-J, Jeong M-H, Kim K-R, Jung C-Y, Lee S-Y, Kim H, Koh J, Vuong TA, Jung S, Yang H, Park S-K, Choi D, Kim SH, Kang K, Sohn J-W, Park JM, Jeon D, Koo S-H, Ho W-K, Kang J-S, Kim S-T, Cho H (2016) Protein arginine methylation facilitates KCNQ channel-PIP2 interaction leading to seizure suppression. eLife 5:e17159 Cooper EC, Abreo T, Tran B (2022) KCNQ channel PIP2 modulation: Two loose links, three rings, and a twist. Neuron 110:178–180 Zhang Q, Zhou P, Chen Z, Li M, Jiang H, Gao Z, Yang H (2013) Dynamic PIP2 interactions with voltage sensor elements contribute to KCNQ2 channel gating. Proc Natl Acad Sci U S A 110:20093–20098 Pipatpolkai T, Mitrovic D, Cui J, Delemotte L (2023) PIP2 binding at the voltage sensor domain facilitates KCNQ1 VSD activation and gating. Biophys J 122:29a Loussouarn G, Park KH, Bellocq C, Baró I, Charpentier F, Escande D (2003) P PIP2 controls KCNQ1/KCNE1 voltagae gated potassium channels: a functional homology between voltage gated and inward rectifier K + channels. EMBO J 22:5412–5421 Kasimova MA, Zaydman MA, Cui J, Tarek M (2015) PIP2-dependent coupling is prominent in Kv7.1 due to weakened interactions between S4-S5 and S6. Sci Rep 5:7474 Corradi V, Mendez-Villuendas E, Ingolfsson HI, Gu RX, Siuda I, Melo MN, Moussatova A, DeGagne LJ, Sejdiu BI, Singh G, Wassenaar TA, Delgado Magnero K, Marrink SJ, Tieleman DP (2018) Lipid-Protein Interactions Are Unique Fingerprints for Membrane Proteins. ACS Cent Sci 4:709–717 Corey RA, Vickery ON, Sansom MSP, Stansfeld PJ (2019) Insights into Membrane Protein-Lipid Interactions from Free Energy Calculations. J Chem Theory Comput 15:5727–5736 Corey RA, Stansfeld PJ, Sansom MSP (2020) The energetics of protein-lipid interactions as viewed by molecular simulations. Biochem Soc Trans 48:25–37 Kefauver JM, Hakala M, Zou L, Alba J, Espadas J, Tettamanti MG, Gajić J, Gabus C, Campomanes P, Estrozi LF, Sen NE, Vanni S, Roux A, Desfosses A, Loewith R (2024) Cryo-EM architecture of a near-native stretch-sensitive membrane microdomain. Nature 632:664–671 Ansell TB, Corey RA, Viti LV, Kinnebrew M, Rohatgi R, Siebold C, Sansom MSP (2023) The energetics and ion coupling of cholesterol transport through Patched1. Sci Adv 9:eadh1609 Arnold WR, Mancino A, Moss FR, Frost A, Julius D, Cheng Y (2024) Structural basis of TRPV1 modulation by endogenous bioactive lipids. Nat Struct Mol Biol 31:1377–1385 Bartocci A, Grazzi A, Awad N, Corringer P-J, Souza PCT, Cecchini M (2024) A millisecond coarse-grained simulation approach to decipher allosteric cannabinoid binding at the glycine receptor α1. Nat Commun 15:9040 Kmiecik S, Gront D, Kolinski M, Wieteska L, Dawid AE, Kolinski A (2016) Coarse-Grained Protein Models and Their Applications. Chem Rev 116:7898–7936 Chan KC, Lu L, Sun F, Fan J (2017) Molecular Details of the PH Domain of ACAP1(BAR-PH) Protein Binding to PIP-Containing Membrane. J Phys Chem B 121:3586–3596 Song W, Corey RA, Ansell TB, Cassidy CK, Horrell MR, Duncan AL, Stansfeld PJ, Sansom MS (2022) P., PyLipID: A Python Package for Analysis of Protein–Lipid Interactions from Molecular Dynamics Simulations. J Chem Theory Comput 18:1188–1201 Ansell TB, Song W, Coupland CE, Carrique L, Corey RA, Duncan AL, Cassidy CK, Geurts MMG, Rasmussen T, Ward AB, Siebold C, Stansfeld PJ, Sansom MS (2023) P., LipIDens: simulation assisted interpretation of lipid densities in cryo-EM structures of membrane proteins. Nat Commun 14:7774 Sejdiu BI, Tieleman DP (2021) ProLint: a web-based framework for the automated data analysis and visualization of lipid–protein interactions. Nucleic Acids Res 49:W544–W550 Abraham MJ, Murtola T, Schulz R, Páll S, Smith JC, Hess B, Lindahl E (2015) GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1–2, 19–25 Barducci A, Bonomi M, Parrinello M, Metadynamics (2011) Wiley Interdisciplinary Reviews-Computational Mol Sci 1:826–843 Bennett WF, Tieleman DP (2013) Computer simulations of lipid membrane domains. Biochim Biophys Acta 1828:1765–1776 Barducci A, Bussi G, Parrinello M (2008) Well-Tempered Metadynamics: A Smoothly Converging and Tunable Free-Energy Method. Phys Rev Lett 100:020603 Wang L, Li S, Xiang S, Liu H, Sun H (2024) Elucidating the Selective Mechanism of Drugs Targeting Cyclin-Dependent Kinases with Integrated MetaD-US Simulation. J Chem Inf Model 64:6899–6911 Laio A, Parrinello M (2002) Escaping free-energy minima. Proc Natl Acad Sci U S A 99:12562–12566 Grimme S, Bannwarth C, Shushkov P, Robust A (2017) Accurate Tight-Binding Quantum Chemical Method for Structures, Vibrational Frequencies, and Noncovalent Interactions of Large Molecular Systems Parametrized for All spd-Block Elements (Z = 1–86). J Chem Theory Comput 13:1989–2009 Lu T, Chen Q (2022) Independent gradient model based on Hirshfeld partition: A new method for visual study of interactions in chemical systems. J Comput Chem 43:539–555 Groenhof GJ (2013) B. s. m.; protocols, Introduction to QM/MM simulations. 43–66 Song W, Corey RA, Ansell TB, Cassidy CK, Horrell MR, Duncan AL, Stansfeld PJ, Sansom MS (2022) P., PyLipID: A Python Package for Analysis of Protein-Lipid Interactions from Molecular Dynamics Simulations. J Chem Theory Comput 18:1188–1201 Traag VA, Waltman L, Van Eck NJ (2019) J. S. r., From Louvain to Leiden: guaranteeing well-connected communities. 9, 1–12 Anuar SHH, Abas ZA, Yunos NM, Zaki NHM, Hashim NA, Mokhtar MF, Asmai SA, Abidin ZZ, Nizam AF (2021) Comparison between Louvain and Leiden algorithm for network structure: a review. In Journal of Physics: Conference Series, ; IOP Publishing: 2021; Vol. 2129; p 012028 Bader RF (1985) J. A. o. c. r., Atoms in molecules. 18, 9–15 Lu H, Marti J (2020) Cellular absorption of small molecules: free energy landscapes of melatonin binding at phospholipid membranes. Sci Rep 10:9235 Lelimousin M, Limongelli V, Sansom MSP (2016) Conformational Changes in the Epidermal Growth Factor Receptor: Role of the Transmembrane Domain Investigated by Coarse-Grained MetaDynamics Free Energy Calculations. J Am Chem Soc 138:10611–10622 Additional Declarations There is NO Competing Interest. Supplementary Files SISIMDA20250218.docx Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6059518","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":425142254,"identity":"7de5cf01-f346-4d20-9e5e-40f9e592b034","order_by":0,"name":"Huanxiang Liu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAsklEQVRIiWNgGAWjYBACAwbmhgMMFRCOBJFaGIFazpCqhYGxjRQt5hKJjYd5523LMzjAfPA2D8M2OYJaLHsONhzm3Xa72OAAW7I1D8NtY8IOO94I1pK44QCPmTRQS2IDQS2HGYFa5oC08H8jUgvYlgawLWxEajlzsOHgnGO3iyUPsxlbzjEgxi83kg9/eFNzO4/vePPDG28qbhMOMRhIYGAGm0C0BpCWUTAKRsEoGAW4AADA80IV1ME8bgAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-9284-3667","institution":"Macao Polytechnic University","correspondingAuthor":true,"prefix":"","firstName":"Huanxiang","middleName":"","lastName":"Liu","suffix":""},{"id":425142255,"identity":"ba8ab688-f268-49c9-9c3d-e7a885bb110c","order_by":1,"name":"LingLing Wang","email":"","orcid":"","institution":"Macao Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"LingLing","middleName":"","lastName":"Wang","suffix":""},{"id":425142256,"identity":"1b06d865-1d20-42a2-a78f-f23a19832fe5","order_by":2,"name":"Shu Li","email":"","orcid":"","institution":"Macao Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Shu","middleName":"","lastName":"Li","suffix":""},{"id":425142257,"identity":"88a7bba2-d5e5-407b-9c76-8867ff79899d","order_by":3,"name":"Yunsen Zhang","email":"","orcid":"","institution":"University of Illinois Urbana-Champaign","correspondingAuthor":false,"prefix":"","firstName":"Yunsen","middleName":"","lastName":"Zhang","suffix":""},{"id":425142258,"identity":"4f97b8bc-1228-40f4-9f9d-4a855305455d","order_by":4,"name":"Huiyong Sun","email":"","orcid":"https://orcid.org/0000-0002-7107-7481","institution":"China Pharmaceutical University","correspondingAuthor":false,"prefix":"","firstName":"Huiyong","middleName":"","lastName":"Sun","suffix":""},{"id":425142259,"identity":"d5223159-7ec9-4566-bf94-8c13c069412e","order_by":5,"name":"Qin Li","email":"","orcid":"","institution":"Macao Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Qin","middleName":"","lastName":"Li","suffix":""},{"id":425142260,"identity":"0513476e-353e-439f-8b84-25d5e937d1ae","order_by":6,"name":"Wei Zhao","email":"","orcid":"","institution":"Macao Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Zhao","suffix":""},{"id":425142261,"identity":"965c3d92-3f52-40e2-9ec7-b27875bcca38","order_by":7,"name":"Xiaomeng Liu","email":"","orcid":"","institution":"Macao Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Xiaomeng","middleName":"","lastName":"Liu","suffix":""},{"id":425142262,"identity":"e195d23d-60c2-4eb9-9814-7e3a67ad4ac1","order_by":8,"name":"Xiao Yan","email":"","orcid":"","institution":"Macao Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Xiao","middleName":"","lastName":"Yan","suffix":""},{"id":425142263,"identity":"77e1f7d9-3fe8-4380-a877-0a06be283edf","order_by":9,"name":"Henry Tong","email":"","orcid":"https://orcid.org/0000-0003-2687-741X","institution":"Centre for Artificial Intelligence Driven Drug Discovery, Faculty of Applied Sciences, Macao Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Henry","middleName":"","lastName":"Tong","suffix":""},{"id":425142264,"identity":"05dc2800-38f9-495b-9d29-269d033b4e31","order_by":10,"name":"Xiaojun Yao","email":"","orcid":"","institution":"Macao Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Xiaojun","middleName":"","lastName":"Yao","suffix":""}],"badges":[],"createdAt":"2025-02-19 01:20:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6059518/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6059518/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":77960413,"identity":"3958020b-c492-4139-a520-81f5f56ad540","added_by":"auto","created_at":"2025-03-07 09:02:08","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2537880,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification and Screening of PIP2 binding sites based on CG -MD and AA-MD simulation.\u003cstrong\u003ea\u003c/strong\u003e Structural representation of the voltage-sensing domain (VSD, S0-S4) and the S5-S6 region of KCNQ1, along with CaM and KCNE3 in the Inactive and Active states. Structures are derived from the Protein Data Bank (PDB) entries 6v00 (Inactive) and 6v01 (Active). \u003cstrong\u003eb\u003c/strong\u003e Top and side views of the KCNQ1-KCNE3-CaM complex based on the 6v01 crystal structure. KCNQ1 is shown as a tetramer, with KCNE3 and CaM highlighted as essential regulatory components. \u003cstrong\u003ec\u003c/strong\u003e 2D-density maps in the plane of the bilayer (XY) of the distribution of PIP2 relative to the four inactive state systems: KCNQ1, KCNQ1-KCNE3, KCNQ1-CaM, and KCNQ1-KCNE3-CaM.\u003cstrong\u003e d\u003c/strong\u003e Density maps of PIP2 binding across the Active state of KCNQ1, KCNQ1-KCNE3, KCNQ1-CaM, and KCNQ1-KCNE3-CaM. \u003cstrong\u003ee\u003c/strong\u003e Binding sites identified and screening from the AA-MD simulations in the Active KCNQ1-KCNE3-CaM system were subjected to the second screening selection. Moreover, the binding of PIP2 at the excluded Bsid9 binding site based on 100 ns AA-MD simulation (a total of 1000 frames).\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6059518/v1/576c8319e4222e9c3a50f7d2.jpeg"},{"id":77960411,"identity":"c790bc6f-d967-4f23-8cb4-32bdead20017","added_by":"auto","created_at":"2025-03-07 09:02:08","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1692874,"visible":true,"origin":"","legend":"\u003cp\u003e8 PIP2 Binding Sites on KCNQ1. \u003cstrong\u003ea \u003c/strong\u003eThe classifications of binding modes of PIP2 at the 8 binding sites.\u003cstrong\u003e b \u003c/strong\u003eBinding modes of PIP2 at the 8 binding sites. \u003cstrong\u003ed-e\u003c/strong\u003e Distribution of PIP2 binding sites in the Inactive and Active KCNQ1 systems, respectively. The numbers in red represent different binding sites (C0-7).\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6059518/v1/eb1d1d78ddd05d5ecc9caf93.jpeg"},{"id":77960418,"identity":"362135b7-ea49-4b3f-862c-83f03dcfe9db","added_by":"auto","created_at":"2025-03-07 09:02:08","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":2526990,"visible":true,"origin":"","legend":"\u003cp\u003eThe hotspot residue contact frequency analysis (radial bar chart) based on AA-MD and the qualitative and quantitative interaction analysis (ball-and-stick model) based on QM-calculated IGMH-AIM analysis for different categories of binding sites (C0-C7). In ball-and-stick model, each circle represents a key weak interaction, with BCP values indicating the strength of these interactions. The color at the center of each circle corresponds to the color on the colorbar, reflecting the interaction type.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6059518/v1/6eef853286bb8834d5cdbf83.png"},{"id":77961841,"identity":"51f765dc-df64-45e5-a575-8bebb7b641f4","added_by":"auto","created_at":"2025-03-07 09:10:08","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":2764566,"visible":true,"origin":"","legend":"\u003cp\u003eThe dissociation pathways of PIP2 at the binding sites C0, C1, and C2 based on Wt-MetaD simulations. \u003cstrong\u003ea,d,g\u003c/strong\u003e Free energy surfaces (FES) during the dissociation process of PIP2 at C0, C1, and C2 binding sites, respectively. The lines represent the Minimum Free Energy Paths (MFEPs), with white areas indicating key metastable states, and black numbers denoting different binding sites. \u003cstrong\u003eb,e,h\u003c/strong\u003eDistance fluctuations between the S2-S3 linker, S2, S3, and S4 regions near the C-terminal and their positions in the initial simulation frame for C0, C1, and C2. \u003cstrong\u003ec,f,i\u003c/strong\u003e Conformational changes near the C-terminal region, including the S2-S3 linker, S2, S3, and S4, during transitions between key stable and metastable states for C0, C1, and C2.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6059518/v1/0f321d32e043226956db0d2f.png"},{"id":77961840,"identity":"63c52928-ef36-4d76-a1b5-a6e468dd618e","added_by":"auto","created_at":"2025-03-07 09:10:08","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":3877747,"visible":true,"origin":"","legend":"\u003cp\u003eThe dissociation pathways of PIP2 across different binding sites consist of a total of 8 paths, labeled Path1-Path8.\u003cstrong\u003e a\u003c/strong\u003e The transparent points represent different types of binding sites, in each subplot (Path1-Path8), the spheres transition from blue to red, representing the positions of PIP2 as dissociation progresses over time. \u003cstrong\u003eb\u003c/strong\u003e. Top-down view integrating all 8 paths.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-6059518/v1/99b51f331d8af0def2ffbf77.png"},{"id":77960415,"identity":"6ca1bd87-0cf5-430e-bb43-22c5cfae25be","added_by":"auto","created_at":"2025-03-07 09:02:08","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":2485060,"visible":true,"origin":"","legend":"\u003cp\u003eThe SIMDA pipeline. \u003cstrong\u003ea\u003c/strong\u003e This paper primarily focuses on the computational simulation methods in the series of studies on the binding and interactions between proteins and lipids.\u003cstrong\u003eb \u003c/strong\u003eStepwise integration of multiscale molecular dynamics (MD) simulations, starting from structure preparation to coarse-grained molecular dynamics (CG-MD), then backmapping each CG-based protein-lipid complex structure at the binding site to all-atom molecular dynamics (AA-MD). A 100-ns AA-MD simulation is performed for each protein-lipid complex structure based on the transformed structure. Enhanced sampling and IGMH calculations are conducted based on AA-MD simulations.\u003cstrong\u003e c \u003c/strong\u003eThe advanced analysis based on multi-scale MD simulations. Firstly, Stepwise screening of reasonable binding sites. All lipid binding sites are identified from CG-MD using the Pylipid package. The initially identified sites are screened based on occupancy, surface area, duration, and lipid count. A second round of screening is performed based on the lipid RMSD from AA-MD trajectories to select stable sites. Secondly, all binding sites are classified using the Leiden clustering algorithm, and Umap is applied for dimensionality reduction to obtain the site classification map and the distribution Sankey diagram of these sites across different systems. Thirdly, from AA-MD to QM, the key interactions of critical binding sites are analyzed from qualitative to quantitative perspectives. Finally, based on well-tempered metadynamics (Wt-metaD) analysis for each site, the thermodynamic and kinetic properties of lipid binding sites are examined.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-6059518/v1/111e404f8a7f39f561aa7a7e.png"},{"id":90941778,"identity":"b505cc00-576f-4cc1-a802-705e60904403","added_by":"auto","created_at":"2025-09-09 18:33:42","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":17137145,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6059518/v1/7dcb5464-dad5-4e34-aa72-2c037a681fb8.pdf"},{"id":77960417,"identity":"fcd172f1-29af-4301-a2cb-c50118c9a0ca","added_by":"auto","created_at":"2025-03-07 09:02:08","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":6037037,"visible":true,"origin":"","legend":"","description":"","filename":"SISIMDA20250218.docx","url":"https://assets-eu.researchsquare.com/files/rs-6059518/v1/36859cf91b7a3fd689df9d11.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Revealing the Molecular Mechanisms of PIP2 Binding and Regulating KCNQ1: Twists, Links, and Binding-Site Transfers via the Developed SIMDA Strategy","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eVoltage-gated potassium channels are essential for regulating cellular excitability, ionic homeostasis, and signal transduction across diverse tissues. Among these, the KCNQ1 (Kv7.1) channel stands out for its functional versatility, playing pivotal roles in processes such as cardiac repolarization, epithelial ion transport, and inner ear function\u003csup\u003e\u003cspan additionalcitationids=\"CR2 CR3\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. This versatility arises from KCNQ1\u0026rsquo;s ability to interact with auxiliary KCNE proteins (e.g., KCNE1, KCNE2, and KCNE3) and regulatory molecules like calmodulin (CaM). These interactions enable KCNQ1 to exhibit tissue-specific gating properties, trafficking behavior, and activity\u003csup\u003e\u003cspan additionalcitationids=\"CR6 CR7\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. For instance, the binding of KCNE3 suppresses the channel\u0026rsquo;s voltage dependence, converting KCNQ1 into a constitutively open potassium channel at physiological membrane potentials\u003csup\u003e\u003cspan additionalcitationids=\"CR10 CR11\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Similarly, CaM, a small, highly conserved calcium-binding protein, binds to the intracellular C-terminal domain of KCNQ1 to regulate channel gating, folding, and trafficking\u003csup\u003e\u003cspan additionalcitationids=\"CR14\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn addition to these protein interactions, KCNQ1 channel activity is tightly regulated by phosphatidylinositol 4,5-bisphosphate (PIP2), a signaling lipid that plays a central role in controlling channel function\u003csup\u003e\u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. For example, PIP2 plays a crucial role in mediating the interaction between the voltage-sensing domains (VSDs) and pore domains (PDs), thereby facilitating the opening of KCNQ channels (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea, b) \u003csup\u003e\u003cspan additionalcitationids=\"CR20 CR21 CR22\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. PIP2 specifically binds to positively charged residues within KCNQ1\u0026rsquo;s C-terminal domain, stabilizing the channel\u0026rsquo;s open state and enabling constitutive activity at resting membrane potentials\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan additionalcitationids=\"CR25\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Importantly, this PIP2-binding region is located near the CaM-binding site. Previous studies suggest that CaM regions can stabilize KCNQ1 and potentially compete with PIP2 for binding to KCNQ1\u003csup\u003e9, 25\u003c/sup\u003e. This cooperative interaction between CaM and PIP2 significantly increases the channel\u0026rsquo;s sensitivity to PIP2, ensuring that PIP2 can efficiently activate KCNQ1 and maintain its function\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn the KCNQ1/KCNE complex, PIP2 also plays an important regulatory role. KCNE3 proteins alter the conformation of KCNQ1\u0026rsquo;s VSDs, locking it in an inactivated state and abolishing voltage dependence\u003csup\u003e\u003cspan additionalcitationids=\"CR28 CR29\" citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. This conformational shift enhances PIP2-binding efficiency and promotes constitutive channel activity. Experimental evidence has shown that PIP2 binding is essential for the function of the KCNQ1/KCNE complex\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. Moreover, KCNE1 increases PIP2 sensitivity by 100-fold compared to channels formed by KCNQ1 alone, effectively amplifying channel current\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. These findings underscore the critical role of PIP2 in maintaining the channel\u0026rsquo;s open state and highlight its functional synergy with KCNE proteins.\u003c/p\u003e \u003cp\u003eMoreover, recent structural studies have revealed the spatial organization of PIP2-binding sites within KCNQ1 and their structural interplay with CaM and KCNE3\u003csup\u003e9, 25\u003c/sup\u003e. For example, in the KCNQ1-KCNE3-PIP2 complex of the channel-open state, a key PIP2 binding site has been identified near the loop connecting the S4 transmembrane segment and the S4-S5 linker (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea)\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Additionally, a range of biochemical, mutational, and computational studies have shown that PIP2 interacts with multiple binding sites across the KCNQ channel family, including the S0 segment, the S2-S3 linker, the S4-S5 linker, the HA helix, the junction between S6 and HA, and the HA-HB linker\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan additionalcitationids=\"CR35 CR36 CR37 CR38\" citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. Despite these advances, the precise molecular mechanisms underlying PIP2-binding dynamics and its synergistic interactions with KCNE and CaM to activate KCNQ1, remain poorly understood.\u003c/p\u003e \u003cp\u003eTo address these limitations and provide a more comprehensive understanding of PIP2-KCNQ1 interactions, here, we proposed a novel Stepwise Integrated Multi-scale Dynamics and Advanced Analysis (SIMDA) framework. SIMDA integrates multiple methodologies to address the challenges in studying lipid-channel interactions. By combining coarse-grained molecular dynamics (CG-MD), all-atom molecular dynamics (AA-MD), quantum mechanics (QM), and well-tempered metadynamics (Wt-MetaD), SIMDA enables rapid identification of lipid-binding sites, detailed characterization of lipid-protein interactions, and exploration of lipid association and dissociation pathways. Furthermore, SIMDA employs advanced analytical tools, such as the Uniform Manifold Approximation and Projection (UMAP)-Leiden clustering algorithm, for consistent identification of functionally equivalent binding sites across multimeric assemblies and systems. The incorporation of QM-based Independent Gradient Model based on Hirshfeld partitioning and Atoms in Molecules (IGMH-AIM) analysis offers quantitative insights into binding interactions and thermodynamic properties, providing a holistic view of lipid-channel dynamics.\u003c/p\u003e \u003cp\u003eIn this study, we utilized our developed SIMDA method to perform over 280 \u0026micro;s of multi-scale molecular dynamics (MD) simulations to explore the binding mechanisms of PIP2 to the KCNQ1 channel in both inactive and active states, with and/or without the auxiliary proteins KCNE3 and CaM. Our results identified eight distinct PIP2-binding sites on KCNQ1. These sites included previously reported regions such as the S2-S3 linker, as well as newly identified sites on the S0 segment, the S4-S5 linker, the HA helix, and the junction between the S6 and HA helix. The presence of KCNE3 and CaM was found to significantly modulate PIP2 binding at these regions. Specifically, KCNE3 enhances the \u0026ldquo;twist\u0026rdquo; effect at the C-terminal of KCNQ1 in its active state, increasing PIP2\u0026rsquo;s binding affinity. Additionally, KCNE3 and CaM facilitate PIP2 binding to a deeper region at the junction between S6 and the HA helix, triggering the tightening of \u0026ldquo;loose links\u0026rdquo;. Finally, our simulations identified eight dissociation pathways of PIP2 from KCNQ1. Along these pathways, PIP2 undergoes transitions across the identified binding sites, revealing the transfer dynamics of PIP2 between KCNQ1 binding sites and offering critical insights into its interaction mechanisms with KCNQ1.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eBinding of PIP2 to KCNQ1 via multi-scale MD simulation\u003c/h2\u003e \u003cp\u003eCG-MD simulations provide an efficient approach to identifying potential lipid-binding sites by simplifying molecular complexity while retaining key interactions\u003csup\u003e\u003cspan additionalcitationids=\"CR41 CR42\" citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. To comprehensively explore the possible PIP2 binding sites on the KCNQ1 channel in different functional states-including both the inactive and active states, with or without the auxiliary proteins KCNE3 and CaM, which modulate channel function, we performed a total of 260 \u0026micro;s CG-MD simulations. The simulations encompassed eight KCNQ1 systems in both inactive (KCNQ1, KCNQ1-KCNE3, KCNQ1-CaM, KCNQ1-KCNE3-CaM) and active states. Each system was simulated in at least five parallel runs, each lasting 5 \u0026micro;s. The simulations were performed on a fluid POPC membrane with 5% PIP2.\u003c/p\u003e \u003cp\u003eRoot-mean-square-deviation (RMSD) calculations were performed for each simulation confirming that the systems reached equilibrium after 1 \u0026micro;s of simulation (Fig.\u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). The final 4 \u0026micro;s trajectories of each simulation were merged for analysis. To identify the interactions between PIP2 and KCNQ1, we calculated 2D-density maps in the plane of the bilayer (XY) to visualize the distribution of PIP2 related to the different KCNQ1 states. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec, d, these density maps revealed PIP2 binding hotspots around the KCNQ1, particularly near regions S3-S4 and S4-S5 linker, which were consistently observed in cryo-EM structures of KCNQ1 in complex with PIP2\u003csup\u003e9\u003c/sup\u003e. Moreover, the presence of KCNE3 and CaM significantly enhanced the PIP2 binding patterns, enhancing hotspots toward regions around S4-S5 linker, S5 and S0-S1 segment, while in the absence of KCNE3 and CaM, PIP2 predominantly interacted with S2-S3 linker. These results highlight the modulatory effects of KCNE3 and CaM on PIP2 binding, suggesting that these auxiliary proteins may stabilize specific binding sites and thereby influence the functional dynamics of the KCNQ1 channel.\u003c/p\u003e \u003cp\u003eTo identify key PIP2-binding sites relevant to different functional states of KCNQ1, clustering and ranking analyses were performed. The PIP2 binding sites in each system were screened based on several key parameters, including occupancy, surface area, and duration (Fig.S2). As shown in Fig.S2, each KCNQ1 functional state ultimately contained 10\u0026ndash;12 categories of PIP2 binding sites, reflecting the diversity of PIP2 interactions within the KCNQ1 channel.\u003c/p\u003e \u003cp\u003eIn order to precisely assess the binding stability of the identified PIP2 binding sites, we extracted representative CG structures for each binding site in each KCNQ1 functional state system, converted them into atomistic models, and performed 100 ns AA-MD simulations. This resulted in 79 PIP2 binding site simulations, totaling 7.9 \u0026micro;s of AA-MD simulation.\u003c/p\u003e \u003cp\u003eAs shown in Fig. S3/S4, most binding sites exhibited minimal RMSD fluctuations, with lipid and protein RMSD values below 3.5\u0026Aring; and 8\u0026Aring;, respectively, indicating stable PIP2 binding and good protein structural stability. However, some PIP2 binding sites, such as Bsid9 in the active KCNQ1-KCNE3-CaM system, showed significant RMSD fluctuations (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ee), suggesting unstable PIP2 binding. Therefore, to screen out unstable binding sites, we excluded binding sites with RMSD fluctuations of more than 1 \u0026Aring; in the last 10 ns. Through this AA-MD simulation-based screening, we identified a total of 68 stable PIP2 binding sites across the different KCNQ1 functional states.\u003c/p\u003e \u003cp\u003eThe identified binding sites within each KCNQ1 functional state system reflected distinct binding locations but did not account for differences between systems. Additionally, since KCNQ1 is a homotetramer composed of four identical subunits, the binding of PIP2 to one subunit is equivalent to its binding to the others. To avoid misinterpreting symmetric binding sites as distinct regions, we performed UMAP-Leiden dimensionality reduction and clustering analysis. UMAP-Leiden cluster is a combined workflow where UMAP reduces high-dimensional data into a low-dimensional space for intuitive visualization, and the Leiden algorithm clusters the data points into well-connected groups, ensuring faster, more accurate, and robust community detection\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. This method enabled the classification and consolidation of binding sites across systems, ultimately clustering 8 distinct binding sites (namely C0 to C7) shared among all systems from a total of 68 all-atom screened PIP2-binding sites. The distribution of these shared binding sites across different systems is presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea.\u003c/p\u003e \u003cp\u003eThese binding sites interact with various regions of KCNQ1 and/or KCNE3 and CaM (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb), and can be classified into two distinct group based on their spatial distribution with the C-terminal domain of KCNQ1. The first group including sites C0, C1, C2 and C7, is located on the one side of the C-terminal domain, near the S2-S3 and S3-S4 linker regions. Within this group, C0, C1, and C2 interact with the S3 segment, S2-S3 linker, and S4-S5 linker, while C1 also binds to the C-terminal of KCNE3. Notably, C7 primarily associates with the S2-S3 linker and the S3 segment. The second group, comprising site C3, C4, C5 and C6, is distributed on the opposite side of the C-terminal domain, near the S2-S3 and S3-S4 linker regions.C4, C5, and C6 interact with the S0-S1 loop and the HA/HB regions, with C5 and C6 also interacting with CaM. C3 primarily interacts with the S1 segment and the S4-S5 linker.\u003c/p\u003e \u003cp\u003eWe further analyzed the distribution characteristics of these binding sites in each system (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec). In the inactive and active KCNQ1-only systems, PIP2 binds to the same sites on both sides of the C-terminal domain, specifically C0, C4, and C7. However, upon the binding of KCNE3, significant redistribution of PIP2 binding sites is observed. On the opposite side of the C-terminal domain, C4 remains unchanged in both states, while a new binding site, C3, emerges. On the C-terminal side, the binding sites shift from C0 to C1, and C7 disappears.\u003c/p\u003e \u003cp\u003eWhen CaM binds to KCNQ1, further state-dependent changes in PIP2 binding sites occur. On the C-terminal side, the binding site shifts from C0 to C2 in the inactive state, whereas C0 is retained in the active state. On the opposite side, C4 shifts to C5 in the inactive state, while in the active state, part of the binding remains at C4, but the majority shifts to C6. Additionally, a new binding site, C3, appears on the opposite side in both states.\u003c/p\u003e \u003cp\u003eIn systems where both KCNE3 and CaM bind to KCNQ1, PIP2 binding sites on both sides of the C-terminal domain exhibit further redistribution. In the inactive state, PIP2 interacts with C2 on the C-terminal side and with C5 and C3 on the opposite side. In the active state, the binding sites shift to C1 on the C-terminal side and to C6 and C3 on the opposite side.\u003c/p\u003e \u003cp\u003eThese findings suggest that KCNE3 and CaM regulate PIP2 interactions in distinct but complementary ways. KCNE3 primarily modifies PIP2 binding on the C-terminal side, favoring C1 over C0. In contrast, CaM modifies PIP2 binding in a state-dependent manner: it favors C2 over C0 in the inactive state and influences PIP2 redistribution on the opposite side of the C-terminal domain, stabilizing interactions at C5 in the inactive state and at C6 in the active state. Additionally, C7 is unique to the KCNQ1-only system and is eliminated in the presence of KCNE3 or CaM, which instead contribute to the emergence of C3 as a new PIP2 binding site absent in the KCNQ1-only system. Together, KCNE3 and CaM fine-tune PIP2 interactions, likely contributing to the dynamic regulation of KCNQ1 channel function in a state-dependent and cooperative manner.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eQualitative and Quantitative Interaction Analysis of PIP2 through QM Calculations\u003c/h3\u003e\n\u003cp\u003eTo investigate the underlying determinants of the 8 identified PIP2 binding sites and their distinct distributions, we first quantified PIP2-protein contact number and frequencies across all binding site clusters using AA-MD simulations (Fig. S5/S6 \u0026amp; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). This analysis provided an overview of PIP2 interaction profiles at each binding site. We then selected the top 10 interacting residues from each cluster for quantum mechanical (QM) calculations, which offer greater precision than AA-MD and CG-MD simulations. Following convergence, we applied the Independent Gradient Model based on Hirshfeld partitioning and Atoms in Molecules (IGMH-AIM) analysis to identify key interaction hotspots (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn C0, residues such as ARG13, LYS93, and ARG89 formed strong hydrogen bonds with PIP2, with bond critical points (BCPs) between 0.033\u0026ndash;0.046 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea). In C1, KCNE3\u0026rsquo;s LYS35 formed the strongest hydrogen bond (BCP: 0.062), reducing the interaction between ARG13 and PIP2 in the S1 region (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb). In C2, ARG89 exhibited the strongest interaction (BCP: 0.121), followed by LYS93 and ARG92 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec). These interaction patterns highlight ARG146, LYS93, ARG92, and ARG89 as consistently critical residues for PIP2 binding across C0, C1, and C2, corroborating previous experimental findings\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. The observed differences in binding modes and interaction strengths across these clusters reflect the unique roles of these residues within distinct structural environments. Notably, C1 and C2 show enhanced interactions between PIP2 and residues in the S3/S2-S3 linker region, suggesting that the presence of KCNE3 or CaM promotes closer and more stable binding interactions in this region. Moreover, in C2 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ed), key PIP2-binding residues are concentrated in the S4-S5 linker and S1 regions. ARG156 and PHE153 in the S4-S5 linker form the strongest hydrogen bonds with PIP2 (BCPs: 0.045 and 0.026), while ARG31 and ARG29 contribute weaker bonds (BCPs: 0.026 and 0.020). These findings highlight the importance of both the S4-S5 linker and S1 regions in stabilizing PIP2 interactions within this cluster. In C7, PIP2 binding is primarily concentrated in the S2-S3 linker/S3 region, with key residues LYS80, ARG89, TYR81, and LYS93 contributing to interactions (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eh). LYS80 forms the strongest hydrogen bond with PIP2 (BCP: 0.062), highlighting its critical role in binding. Although C7 shows fewer interacting regions compared to other clusters, the strong interactions within the S2-S3 linker emphasize its importance in stabilizing PIP2 binding.\u003c/p\u003e \u003cp\u003eIn C3, ARG156 (BCP: 0.045) and PHE153 (BCP: 0.026) in the S4-S5 linker region are identified as the most critical residues, forming strong hydrogen bonds with PIP2. Additionally, the involvement of ARG29 (BCP: 0.020) and ARG31(BCP: 0.026) further underscores the importance of both the S1 and S4-S5 linker regions in PIP2 binding within PIP2 at C3 binding sites (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ed).\u003c/p\u003e \u003cp\u003eBinding sites C4, C5, and C6 displayed distinct patterns. In C4, residues in the HA and S0-S1 regions (e.g., ARG156, ARG6) formed neutral hydrogen bonds (BCP: 0.015\u0026ndash;0.034) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee). In C5, interactions focused on the S0-S2 region, with LYS71 forming the strongest bond (BCP: 0.048) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ef). In C6, residues in the CaM region, such as ARG31 and ARG33, exhibited strong interactions with PIP2 (BCP: 0.028\u0026ndash;0.049) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eg). Finally, in C7, LYS80 formed the strongest bond (BCP: 0.062), followed by ARG89 and TYR81 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eh). Across clusters C4, C5, and C6, ARG6, LYS18, and TRP17 are consistently critical for PIP2 binding. Notably, In C6, ARG31 and ARG33 in the CaM region play crucial roles in PIP2 interactions, emphasizing the importance of CaM in regulating binding dynamics. While C5 shows stronger overall binding than C4, its interactions are more localized in the S1 region, whereas C4 features broader interactions across the S0, S1, and HA regions. These findings highlight the regulatory role of CaM in PIP2 binding, offering insights into how lipid-protein interactions impact membrane protein function and providing a basis for further research and therapeutic development.\u003c/p\u003e\n\u003ch3\u003eThe Impact of KCNE3 and CaM on the “Twist and Loose Links” Mechanism\u003c/h3\u003e\n\u003cp\u003eUnbiased CG-MD simulations are limited in their ability to capture detailed atomic-level interactions\u003csup\u003e\u003cspan additionalcitationids=\"CR48 CR49 CR50\" citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e, while AA-MD simulations, despite providing accurate descriptions of such interactions, are insufficient for comprehensively characterizing the kinetics and thermodynamics of lipid-protein binding and unbinding processes\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e. To overcome these limitations, we utilized well-tempered metadynamics (Wt-metaD) simulations based on AA-MD simulation results, to systematically investigate the dissociation pathways and binding preferences across distinct PIP2 binding sites. A total of \u0026gt;\u0026thinsp;10 \u0026micro;s of Wt-metaD simulations (\u0026gt;\u0026thinsp;100 ns per binding site) were conducted, focusing on the 8 distinct PIP2 binding sites identified through UMAP-Leiden dimensionality reduction and clustering analysis. In each Wt-metaD simulation, two collective variables (CVs) were defined to enhance sampling. Specifically, the X and Y components of the distances between each lipid and surrounding residues within 10 \u0026Aring; were selected as CVs. This choice of CVs enhances precision in capturing local interactions, simplifies the representation of complex systems, and improves the exploration of binding landscapes, providing detailed insights into lipid-protein interaction dynamics.\u003c/p\u003e \u003cp\u003eFor each binding site, PIP2 dissociates along its respective CV, following similar or distinct pathways. This process enables the sampling of unique Free Energy Surfaces (FES), which represent the energy landscape of the system as a function of the CVs, highlighting the energetic barriers and stable states during dissociation. Additionally, the Minimum Free Energy Path (MFEP), which traces the most favorable energetic route across the FES, provides a detailed view of the preferred dissociation pathway for PIP2 at each binding site. To explore the preferences among different dissociation paths, we performed kinetic and thermodynamic analyses. Specifically, kinetic trajectory analysis from Wt-metaD simulations revealed that binding sites C0, C1, and C2 follow the same dissociation pathway (Path1) (Fig.S7). By identifying the Minimum Free Energy Path (MFEP) from the corresponding Free Energy Surfaces (FES) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea, d, g), the free energy profiles were obtained, with dissociation free energy values of 58.01 kJ/mol, 110.27 kJ/mol, and 77.72 kJ/mol for C0, C1, and C2, respectively. To identify the source of these free energy differences, we conducted dynamic analyses of C0, C1, and C2 (for example, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb,e). Each dissociation trajectory was classified into four states (A-D), corresponding to the minimal energy states on the FES. These states illustrate the proportions of each binding state along the dissociation trajectory. Interaction analysis revealed significant conformational changes in the S2-S3 linker across C0, C1, and C2. Notably, C1 exhibited prominent changes in the S2 and S3 regions, as well as in KCNE3. Although C2 showed smaller fluctuations compared to C1, they were more pronounced than in C0. The presence of KCNE3 enhanced the affinity of PIP2 for C1, with the conformational changes of KCNE3 closely tied to the binding and dissociation process. This may be due to the strong hydrogen bond interactions formed between LYS35 and ARG36 at the KCNE3 tail and PIP2. Here, we clearly observed the C-terminal \u0026ldquo;twist effect\u0026rdquo; induced by PIP2 binding, along with significant conformational changes in the S2-S3 linker region. The presence of KCNE3 intensifies this effect, causing larger twists in other regions, such as S2 and S3. The higher free energy required for PIP2 binding in C2 compared to C0 may result from the tighter hydrogen bond interactions between PIP2 and the residues in the S3/S2-S3 linker region in C2. These stronger interactions lead to greater fluctuations in the C-terminal region of C2, primarily involving the S2-S3 linker, S2, and S3.\u003c/p\u003e \u003cp\u003eThe FES analyses of the dissociation pathways of PIP2 at the binding sites C4, C5, and C6 revealed that the primary dissociation paths were Path 7 and Path 8, with Path 8 being dominant (Fig.S7 \u0026amp; Fig.S8e-h). The free energy calculations showed values of approximately 100 kJ/mol at C4, 49.16 kJ/mol at C5, and 79.96 kJ/mol for C6. Notably, C5 was only present in the inactive state of KCNQ1 with CaM, while C6 appeared exclusively in the active state of KCNQ1 with CaM. These findings suggest that CaM in the active state of KCNQ1 exerts a competitive attraction, lowering the free energy data comparing with non-CaM system. This phenomenon is consistent with the interaction analysis results. In C6, the CaM in the active KCNQ1 systems forms hydrogen bond interactions with PIP2 through ARG31 and ARG33, attracting PIP2 binding and competing with the PIP2 bound in the S1 region of KCNQ1, which may explain the decrease in PIP2 binding affinity. In C5, although the CaM in the inactive KCNQ1 systems makes contact with PIP2, it does not form hydrogen bond interactions. As a result, while CaM attracts PIP2, it fails to establish more stable interactions, leading to a further reduction in binding affinity.\u003c/p\u003e \u003cp\u003eC3 serves as a highly dynamic binding site for PIP2. Due to its deeply buried location within the protein structure (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb), PIP2 tends to rebind to other sites-specifically C0, C4, and C7 during its dissociation along Path 5 and Path 6 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea). This propensity for rebinding induces substantial conformational rearrangements in the VSD region. Moreover, FES analyses, as illustrated in Fig.S9, reveal that the dissociation of PIP2 from C3 requires a high free energy input exceeding 95 kJ/mol. These complex dissociation characteristics activate multiple other binding sites, including C0, C4, and C7, thereby influencing the protein\u0026rsquo;s structural dynamics (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Community network analysis of key stable binding sites further elucidates this phenomenon. As PIP2 gradually dissociates from C3, the community coupling network between the VSD and PD regions evolves significantly. Initially, the network comprises 3 communities (Fig. S9a); it then transitions to 3 communities with altered coupling (Fig. S9b), progresses to 4 communities (Fig. S9c), and ultimately expands to 5 distinct communities once PIP2 completely exits the binding pocket communities (Fig. S9d). This evolution indicates that PIP2 binding at C3 brings these regions into closer proximity, intensifying the \u0026ldquo;loose links\u0026rdquo; effect and enhancing inter-domain communication. In contrast, C7 functions as a shallow binding site through which several binding pathways converge, exhibiting a relatively straightforward dissociation pathway primarily along Path 3 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea). Notably, the strong hydrogen bond interactions between PIP2 and residues in the S2-S3 linker/S3 region at C7 lead to significant fluctuations in the S2-S3 linker. These fluctuations are critical as they trigger the C-terminal \u0026ldquo;twist\u0026rdquo; effect, a conformational change that may influence the protein\u0026rsquo;s gating mechanisms.\u003c/p\u003e \n\u003ch3\u003eDissociation Kinetic Pathways and Binding Site Transfers of PIP2 in KCNQ1\u003c/h3\u003e\n\u003cp\u003eTo further investigate the kinetic behavior of binding and dissociation at these sites, we analyzed the dissociation trajectories of PIP2 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb) and identified 8 representative pathways. PIP2 primarily binds to two sides of the KCNQ1 C-terminal domain. The binding sites on the C-terminal domain can be categorized into deeper binding sites (C0-C2) and a more outward site (C7) on one side, while the opposite side includes another deeper site (C3) and outward sites (C4-C6). We found that PIP2 can directly dissociate from the binding regions C0-C2 (Path 5), C4-C6 (Path 8), and C7 (Path 3). However, dissociation from C3 requires a transfer to either C0-C2 (Path 5, involving the S5-S6, HA, and the C-terminal S0-S4, S4-S5 linker, and KCNE3 regions) or C4-C6 (Path 6, involving the S5-S6, HA/HB, CaM, and KCNE3 regions) before release (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb). This indicates that the dissociation of PIP2 from C3 (S5-S6) is kinetically dependent on intermediate transfer steps, highlighting a unique mechanism whereby C3 serves as a transitional binding site rather than a direct dissociation site.\u003c/p\u003e \u003cp\u003eIn addition to direct dissociation, PIP2 can also dissociate from C0-C2, C4-C6, and C7 by transitioning through other binding sites. For example, PIP2 can transfer from C0-C2 to C7 before dissociation (Path 1, involving the C-terminal S0-S4, S4-S5 linker, and the KCNE3 region) or to C4-C6 before dissociation (Path 2, involving the C-terminal S0-S4, S4-S5 linker, CaM, and the HA-HB regions). Furthermore, PIP2 can transition between the C4-C6 and C7 binding sites before eventually dissociating from KCNQ1 (Path 4 and Path 5). These findings reveal the complex kinetic pathways and inter-site dynamics of PIP2 binding and dissociation, underscoring the critical role of intermediate states and site transitions in regulating PIP2 interactions with KCNQ1.\u003c/p\u003e \u003cp\u003eBuilding upon this trajectory analysis, we propose that PIP2 binding is primarily associated with five distinct structural domains: the VSD at the C-terminal (S0-S4, S4-S5 linker), the PD (S5-S6), KCNE3, CaM, and the HA-HB region near the membrane (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Among these, the VSD and PD regions are closely linked to electromechanical coupling, playing a critical role in PIP2 binding. Specifically, binding at deeper sites, such as C2 and C3, results in the tightening of \u0026ldquo;loose links\u0026rdquo;, which enhances the coupling between the voltage-sensor and pore domains. PIP2 binding at these sites is strongly associated with KCNE3, whose presence facilitates PIP2 binding. Furthermore, CaM modulates the binding preference of PIP2, shifting it toward shallower sites like C5/C6, which prevents some PIP2 molecules from reaching deeper binding sites. Overall, PIP2 binding dynamics-whether through direct interactions or transitions between binding sites are significantly influenced by the presence of KCNE3 and CaM. These factors regulate both the accessibility and binding preferences of PIP2 at deep and shallow binding sites, ultimately affecting the overall dynamics of PIP2 interactions within the system.\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eIn this study, we developed SIMDA, an advanced computational framework designed to systematically analyze the complex interactions between membrane proteins and lipids. By integrating stepwise multiscale MD simulations and sophisticated advanced analysis, SIMDA enables precise identification of lipid binding sites, detailed calculation of residue interactions, and comprehensive analysis of dynamics and thermodynamic properties of lipid binding. The framework employs a three-tiered strategy: multiscale simulations from CG-MD to AA-MD simulations and to QM calculations/Wt-metaD simulations; systematic identification and classification of lipid binding sites based on CG-MD and AA-MD simulations; and residue interaction quantification via AA-MD and IGMH-AIM calculations. The incorporation of Umap-Leiden clustering for dimensionality reduction facilitates robust handling of high-dimensional lipid-protein interaction data, ensuring accurate and scalable analyses.\u003c/p\u003e \u003cp\u003ePIP2 directly modulates the activity of a wide range of ion channels, including inward rectifier potassium (K\u003csup\u003e+\u003c/sup\u003e) channels, voltage-gated calcium channels, transient receptor potential channels, and particularly voltage-gated KCNQ potassium channels (KCNQ1-5, also known as KV7.1-5)\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan additionalcitationids=\"CR55 CR56\" citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e–\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e. The binding sites and mechanisms of PIP2 interaction with various KCNQ family members (KCNQ2-4) have been extensively investigated through both experimental approaches and molecular dynamics simulations\u003csup\u003e\u003cspan additionalcitationids=\"CR59 CR60\" citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e–\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e. For KCNQ2, Zhang et al.\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e demonstrated that in the closed state, PIP2 is anchored at the S2–S3 loop, while upon channel activation, it interacts with the S4–S5 linker, thereby participating in channel gating. Pant et al.\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e characterized PIP2 binding in human Kv7.2 channels, revealing that in the closed state, PIP2 is localized at the periphery of the voltage-sensing domain (VSD), and in the open state, it binds to four distinct interfaces formed by the cytoplasmic ends of the VSD, the gate, intracellular helices A and B, and their linkers. PIP2 binding induces a bilayer-interacting conformation of helices A and B, as well as coordinated movement between the VSD and the pore domain. Additionally, Zheng et al.\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e identified two PIP2 molecules in the open-state structure of KCNQ4, which function as a bridge to couple the VSD and pore domain, facilitating channel opening. While substantial evidence supports the critical role of PIP2 in the activation and inactivation processes of KCNQ1, only a single PIP2 binding site has been identified\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan additionalcitationids=\"CR64\" citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e–\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e. A comprehensive understanding of the full binding mechanism in KCNQ1 remains lacking.\u003c/p\u003e \u003cp\u003eUsing the SIMDA framework, we explored the interactions between the voltage-gated ion channel KCNQ1 and PIP2, a lipid essential for channel activation. Our analysis covered eight distinct systems involving KCNQ1 in both inactive and active states, with or without the regulatory proteins KCNE3 and CaM. Through CG-MD and AA-MD simulations, combined with Umap-Leiden clustering, we identified eight distinct PIP2 binding clusters (C0-7). These clusters exhibited consistent kinetic and thermodynamic properties across different systems, revealing a conserved binding pattern despite varying regulatory conditions.\u003c/p\u003e \u003cp\u003eOur results show that the presence of KCNE3 significantly enhances the C-terminal “twist” effect of KCNQ1, increasing the affinity of PIP2. Key residues LYS35 and ARG36 in in C2 form strong hydrogen bonds with PIP2, inducing significant fluctuations in the KCNE3 tail during PIP2 dissociation. These interactions propagate conformational changes across the S2, S3, and S4-S5 linker regions, creating a higher energy barrier for dissociation. The structural rearrangement suggests that KCNE3 accelerates the C-terminal “twist” effect, thereby stabilizing PIP2 binding and increasing its affinity.\u003c/p\u003e \u003cp\u003eIn contrast, CaM plays a regulatory role by competing with PIP2 for binding, thereby reducing PIP2 affinity in a state-dependent manner. Our analysis shows that CaM inhibits PIP2 binding at C4, regardless of whether KCNQ1 is in the active or inactive state. However, CaM in the active state of KCNQ1 enhances the strength and complexity of interactions at C6 compared to that in the inactive state of KCNQ1. Further IGMH-AIM calculations confirm that ARG31 and ARG33 in the CaM region play critical roles in PIP2 binding, emphasizing CaM’s significant influence on the binding dynamics of KCNQ1.\u003c/p\u003e \u003cp\u003eOur investigation also reveals the unique role of C3 in reflecting the “loose links” effect within the binding network. Key residues ARG156 and PHE153 in C3 form strong hydrogen bonds with PIP2, and the dissociation of PIP2 from C2 often passes through other clusters, such as C0, C1, C5, and C6. This migration gradually tightens previously loose regions of the protein, demonstrating the transition from “loose links” to a more stable structure, which may be critical for the activation of the VSD. Furthermore, C7 plays a crucial role in the C-terminal “twist” effect. Closely associated with the S2-S3 linker, C7 acts as an intermediate site during PIP2 dissociation and facilitates the transition between multiple binding sites, highlighting its previously overlooked regulatory importance.\u003c/p\u003e \u003cp\u003eAdditionally, our analysis of the dissociation kinetics and binding site transfers of PIP2 in KCNQ1 reveals complex, multi-step dynamics involving both direct dissociation and intermediate site transitions. We identified eight representative dissociation pathways, which involve transitions between deeper and more outward binding sites, highlighting the intricate nature of PIP2 interactions with KCNQ1. Notably, dissociation from certain sites, such as C3, requires intermediate transitions, indicating that these transitional binding sites play a critical role in facilitating dissociation. We propose a model wherein PIP2 binding occurs at five distinct structural domains-namely, the voltage-sensor domain (VSD), the pore domain (PD), KCNE3, CaM, and the HA-HB region near the membrane. This model underscores the electromechanical coupling between these domains, with deeper binding sites such as C2 and C3 facilitating the tightening of “loose links” and thereby enhancing the coupling between the voltage-sensor and pore domains. These findings provide new insights into the molecular mechanisms underlying PIP2 regulation of KCNQ1, revealing the complex interplay between binding and dissociation processes that are essential for the functional dynamics of this ion channel.\u003c/p\u003e \u003cp\u003eIn summary, our study provides a comprehensive understanding of the complex interactions between KCNQ1 and PIP2, capturing the intricate dynamics we refer to as the “twists, links, and binding site transfers” effect. We find that KCNE3 in the active state of KCNQ1 accelerates the C-terminal twist effect and significantly enhances PIP2 binding affinity, while CaM in the active state of KCNQ1 competitively reduces PIP2 binding by modulating interaction strength and complexity. The observed migration of PIP2 between multiple binding sites suggests that dynamic transitions play a vital role in its regulatory mechanism. Although we captured several intermediate states of PIP2 during these transitions, we did not observe a complete migration of a single PIP2 molecule across all possible binding sites. Future experimental and computational studies can explore these transitions in greater detail to construct a comprehensive landscape of PIP2-mediated regulation of KCNQ1.\u003c/p\u003e \u003cp\u003eMD simulations have become essential tools for investigating protein-lipid interactions, providing valuable insights into lipid-binding mechanisms at various scales\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e, \u003cspan additionalcitationids=\"CR67 CR68\" citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e–\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e. CG-MD simulations, in particular, are widely used due to their ability to efficiently identify potential lipid-binding sites\u003csup\u003e\u003cspan additionalcitationids=\"CR70\" citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e–\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e. For example, Bartocci et al.\u003csup\u003e\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e\u003c/sup\u003e employed millisecond-scale CG-MD simulations combined with reverse-mapping to all-atom representations, uncovering multiple cannabinoid-binding sites on GlyR and offering an effective strategy for identifying allosteric binding sites. However, CG-MD simulations are limited by their resolution, which is insufficient to capture secondary structural changes or detailed atomic-level interactions, a key drawback when trying to understand the precise mechanisms underlying protein-lipid binding\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e, \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn contrast, AA-MD simulations provide higher-resolution insights into atomic-level conformational changes, enabling a deeper understanding of lipid-protein interactions\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. Nevertheless, unbiased AA-MD simulations are constrained by time scale limitations, often capturing only equilibrium states close to lipid-binding sites, and are unable to fully sample rare, high-energy events that are critical for understanding dynamic binding processes. To address this, enhanced sampling techniques such as well-tempered metadynamics, umbrella sampling, and Gaussian accelerated molecular dynamics are employed. These techniques allow for the exploration of large-scale conformational transitions and lipid-binding events. For instance, Chan et al.\u003csup\u003e\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e\u003c/sup\u003e used umbrella sampling simulations to examine binding pockets on the PH domain and generated potential of mean force profiles, elucidating their preference for PIP2 lipids. This approach, combined with orientation analysis and binding pocket studies, provided valuable molecular insights into protein-lipid interactions during membrane remodeling. While CG-MD, AA-MD, and enhanced sampling techniques each have their strengths, they also have limitations, particularly in capturing the full complexity of protein-lipid interactions. To overcome these limitations, we propose the SIMDA method, which integrates these techniques in a stepwise manner. The SIMDA method begins with CG-MD simulations to identify potential lipid-binding sites, followed by AA-MD simulations to capture detailed atomic interactions, and culminates with QM or well-tempered Wt-metaD simulations to probe rare events and electronic interactions with high precision. This progressive multi-scale approach provides a comprehensive view of lipid-protein interactions, enabling the detailed characterization of lipid binding at different time and length scales.\u003c/p\u003e \u003cp\u003eAdditionally, several tools have been developed to assist with the analysis of protein-lipid binding structures and MD trajectories. Tools like PyLipID, LipIDens, and ProLint are particularly useful for identifying, ranking, and refining lipid binding poses, enhancing the accuracy of lipid-protein interaction studies\u003csup\u003e\u003cspan additionalcitationids=\"CR76\" citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e–\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e\u003c/sup\u003e. For instance, PyLipID can be used to calculate residence times of lipids at binding sites, further advancing our understanding of the dynamics of protein-lipid interactions\u003csup\u003e\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e\u003c/sup\u003e. We believe these tools have significantly aided our research. However, these analytical tools do not address the consistent identification of functionally equivalent binding sites across multimeric assemblies and systems, nor do they provide more accurate quantitative interactions. Therefore, the SIMDA method introduces new analysis tools, such as the UMAP-Leiden clustering algorithm, which enables consistent identification of functionally equivalent binding sites across multimeric assemblies and complex systems. This method improves the resolution of binding site patterns in larger, more intricate systems, facilitating more accurate predictions of functional behaviors. The integration of quantum mechanical-based IGMH-AIM analysis further enhances SIMDA by providing quantitative insights into binding interactions and thermodynamic properties, allowing for a holistic view of lipid-channel dynamics that bridges multiple scales of simulation.\u003c/p\u003e \u003cp\u003eIn conclusion, the SIMDA method systematically captures detailed data on lipid-protein interactions through stepwise multiscale dynamic simulations, employing a progressive strategy that transitions from CG-MD to AA-MD and further to QM or Wt-metaD simulations. This multilayered framework not only comprehensively elucidates complex membrane protein-lipid interactions but also establishes a solid foundation for in-depth analysis. The key advantage of SIMDA lies in its high flexibility and precision, enabling accurate identification of binding sites through its progressive approach while using dimensionality reduction and clustering analysis to uncover functional properties and relationships among binding sites. Furthermore, SIMDA facilitates a seamless transition from qualitative assessments to quantitative studies, allowing detailed investigations of binding strength, interaction modes, and their dynamic changes under various conditions. It also analyzes the thermodynamics and kinetics of binding and dissociation processes, revealing energy barriers, stability, and microscopic regulatory mechanisms. This macro-to-micro analysis provides theoretical support for the design of targeted drugs based on binding sites, while offering insights into membrane protein mechanisms such as lipid binding, structural dynamics, and interaction regulation. In the future, SIMDA can be applied beyond ion channels to investigate the intricate interactions of G protein-coupled receptors (GPCRs) and membrane transporters, uncovering critical regulatory sites and energy transition pathways. These capabilities make SIMDA a powerful tool for studying membrane protein-lipid interactions, providing essential theoretical foundations for developing novel regulatory mechanisms and therapeutic targets, with broad potential applications in biomedicine and drug development.\u003c/p\u003e "},{"header":"METHODS","content":"\u003cp\u003eTo obtain comprehensive, precise, and reliable insights into protein-lipid binding sites, as well as the mechanisms by which cofactor modulation influences intricate protein-lipid systems in different states, we conducted extensive molecular dynamics simulations and analyses. The Stepwise Integrated Multi-scale Dynamics and Advanced Analysis (SIMDA) strategy consists of two primary components: the implementation of stepwise multiscale molecular dynamics (MD) simulations and the integration of advanced analytical methods (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eFirst, during the stepwise multiscale MD simulation process, our strategy begun by using coarse-grained molecular dynamics (CG-MD) simulations to efficiently and comprehensively determine how lipids bind to proteins\u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e, \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e78\u003c/span\u003e\u003c/sup\u003e. From the first round of screening based on CG-MD simulations, lipid-protein complexes for each system were saved and converted into all-atom models. In the second round of screening, all-atom molecular dynamics (AA-MD) simulations was employed to analyze the binding details at each site on the atomic level\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e, \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e79\u003c/span\u003e, \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e80\u003c/span\u003e\u003c/sup\u003e. Subsequently, we used the X and Y components of the distances between each lipid and the surrounding residues within 10 Å as collective variables (CVs) for enhanced sampling via well-tempered metadynamics (Wt-metaD), based on AA-MD results, to explore lipid binding and dissociation processes at each site\u003csup\u003e\u003cspan additionalcitationids=\"CR82\" citationid=\"CR81\" class=\"CitationRef\"\u003e81\u003c/span\u003e–\u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e83\u003c/span\u003e\u003c/sup\u003e. Key residues involved in lipid binding are identified from AA-MD simulations, followed by quantum mechanics (QM) simulations for a deeper analysis\u003csup\u003e\u003cspan additionalcitationids=\"CR85\" citationid=\"CR84\" class=\"CitationRef\"\u003e84\u003c/span\u003e–\u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e86\u003c/span\u003e\u003c/sup\u003e. This stepwise multiscale MD simulation approach is designed to reveal critical binding information across scales, from large to small, from coarse to fine, and from qualitative to quantitative, all in a highly efficient and precise manner. Moreover, this method maximizes the strengths of CG-MD, AA-MD/Wt-metaD, and QM while mitigating their individual limitations, providing a comprehensive and systematic approach to studying lipid-protein interactions (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb).\u003c/p\u003e\u003cp\u003eNext, based on the above stepwise multiscale MD simulation results, we conducted a series of advanced analyses. First, we performed a two-step screening process for the binding sites. The initial screening was based on CG-MD results, using the PyLipID package to identify all potential binding sites across the eight systems\u003csup\u003e\u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e87\u003c/span\u003e\u003c/sup\u003e. For each binding site, we performed analyses of occupancy, surface area and duration, followed by ranking to select the top-ranked binding modes. These binding modes were then converted into all-atom models for AA-MD simulations, where we calculated the RMSD of the lipids at each binding site. Using these RMSD values, we performed a second screening to identify the stably bound sites (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec).\u003c/p\u003e\u003cp\u003eWhile the previous two screenings focused on vertical analyses within each system, we also conducted a horizontal comparison to explore the differences in lipid binding across systems and assess the effects of different states and cofactors. To achieve this, we applied dimensionality reduction and clustering analysis on the interaction data from all systems’ binding sites. Specifically, we used Uniform Manifold Approximation and Projection (UMAP) for dimensionality reduction and Leiden clustering for grouping\u003csup\u003e\u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e88\u003c/span\u003e\u003c/sup\u003e. UMAP is a nonlinear dimensionality reduction technique that maps high-dimensional data into low-dimensional space, usually in two or three dimensions, while preserving local data structures for more intuitive visualization\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. Leiden is a community detection algorithm that optimizes modularity to cluster similar data points into the same group\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. Compared to other clustering algorithms, Leiden is faster, more accurate, and ensures the connectivity of each cluster, avoiding fragmentation\u003csup\u003e\u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e89\u003c/span\u003e\u003c/sup\u003e. First, we applied UMAP to map the interaction data of binding site residues from all systems into a low-dimensional space, allowing the similarities between binding sites to be visualized more intuitively. We then applied Leiden clustering to the reduced data, grouping similar binding sites into distinct clusters. This approach effectively identified subgroups of binding sites, capturing biological heterogeneity in the data and providing insights into the types and states of binding sites (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec).\u003c/p\u003e\u003cp\u003eTo further investigate the key interaction patterns of the screened binding sites, we performed a stepwise analysis of protein-lipid interactions of contact number analysis based on AA-MD to Hirshfeld partitioning and Atoms in Molecules (IGMH-AIM) analysis analysis based on QM simulations, transitioning from qualitative to quantitative insights \u003csup\u003e\u003cb\u003e\u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e90\u003c/span\u003e\u003c/b\u003e\u003c/sup\u003e. This comprehensive multiscale approach provided reliable interaction information, revealing key residues closely associated with lipid binding. Finally, based on the Wt-metaD trajectories of lipid binding and dissociation, we examined the thermodynamic and kinetic differences of lipid binding at different sites, uncovering the dynamic effects of protein states and cofactors on lipid binding (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec) \u003csup\u003e\u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e91\u003c/span\u003e, \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e92\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn summary, the SIMDA method systematically investigates lipid-protein interactions through a progressive multiscale MD simulation framework, transitioning from CG-MD to AA-MD, and further to QM or Wt-metaD simulations. This approach captures both macro-level structural features and micro-level interaction dynamics, providing a robust foundation for in-depth analysis. It follows four key steps: identifying and screening the most stable and functional binding sites using progressively refined simulations for precision and reliability; applying dimensionality reduction and clustering analysis to classify binding site characteristics and reveal functional relationships; transitioning from qualitative to quantitative analysis to explore binding strength, interaction modes, and dynamic changes under varying conditions; and analyzing thermodynamics and kinetics to uncover energy barriers, stability, regulatory mechanisms, and microscopic kinetic behaviors.\u003c/p\u003e\u003cp\u003eBy integrating multiscale MD simulations, binding site identification, and interaction analysis, SIMDA comprehensively elucidates PIP2 binding sites, state-dependent interactions, and cofactor modulation in complex membrane proteins. Beyond structural analysis, it quantifies thermodynamic and kinetic properties, including energy transformations, interaction patterns, and dissociation pathways, while identifying regulatory sites and demonstrating how cofactors like PIP2 modulate membrane protein function. SIMDA’s flexibility across simulation scales ensures seamless integration of macro-structural insights with micro-dynamic analyses, offering a holistic understanding of membrane protein-lipid interactions. This strategy not only reveals energy distributions and kinetic pathways but also provides actionable insights for designing targeted drugs and exploring lipid-binding mechanisms, membrane protein dynamics, and interaction regulation, establishing SIMDA as a powerful tool for biomedical research and therapeutic development.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData Availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe initial coordinates for the active and inactive structures of KCNQ1 were retrieved from the RCSB Protein Data Bank with accession codes 6v01 and 6v00, respectively. All CG-MD, AA-MD, and WT-metaD were performed using Gromacs 2021.4. QM calculations and IGMH-AIM analysis were conducted using ORCA software. PIP2 binding site identification, screening, classification, and UMAP-Leiden clustering were carried out using Python (version 3.10). Structural visualization was done with PyMOL (version 2.5) and VMD (version 1.9.4). The source data and parameter settings for the multi-scale molecular dynamics simulations have been deposited in Zenodo (https://zenodo.org/records/14885900). Detailed methods, along with additional data analysis and figures, are provided in the Supplementary Information.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode Availability\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll code of SIMDA framework are freely available on GitHub at https://github.com/xiaoxiaowang1201/SIMDA-framework and on Zenodo at https://zenodo.org/records/14885900 with an MIT license.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Science and Technology Development Fund, Macau, SAR (No. 0091/2022/A2) and Macao Polytechnic University (No. RP/FCA-02/2023)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eL. W. contributed to the main ideas, coding, data analysis, and writing of the manuscript. S. L. contributed to the initial modeling, data analysis, and manuscript revisions. Y. Z. and H. S. helped review the paper and provided suggestions for revisions. Q. L., W. Z., X. L., X. Y., and H. T. participated in discussions on the method implementation. H. Liu provided computational resources and the initial concept for the paper.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interest.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBarhanin J, Lesage F, Guillemare E, Fink M, Lazdunski M, Romey GJ (1996) N., KvLQT1 and IsK (minK) proteins associate to form the I Ks cardiac potassium current. 384:78\u0026ndash;80\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSanguinetti MC, Curran ME, Zou A, Shen J, Specter P, Atkinson D, Keating MJN (1996) Coassembly of KvLQT1 and minK (IsK) proteins to form cardiac I Ks potassium channel. 384:80\u0026ndash;83\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWarth R, Barhanin JJT (2003) J. o. m. b., Function of K\u0026thinsp;+\u0026thinsp;channels in the intestinal epithelium. 193, 67\u0026ndash;78\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng W, Verlander JW, Lynch IJ, Cash M, Shao J, Stow LR, Cain BD, Weiner ID, Wall SM, Wingo CS (2007) J. A. J. o. P.-R. P., Cellular distribution of the potassium channel KCNQ1 in normal mouse kidney. 292:F456\u0026ndash;F466\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAl-Hazza A, Linley J, Aziz Q, Hunter M, Sandle GJB (2016) communications, b. r., Upregulation of basolateral small conductance potassium channels (KCNQ1/KCNE3) in ulcerative colitis. 470:473\u0026ndash;478\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbbott GW, Tai K-K, Neverisky DL, Hansler A, Hu Z, Roepke TK, Lerner DJ, Chen Q, Liu L, Zupan BJS (2014) s., KCNQ1, KCNE2, and Na+-coupled solute transporters form reciprocally regulating complexes that affect neuronal excitability. 7:ra22\u0026ndash;ra22\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcCrossan ZA, Abbott GW (2004) The MinK-related peptides. Neuropharmacology 47:787\u0026ndash;821\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKirchhofer A, Helma J, Schmidthals K, Frauer C, Cui S, Karcher A, Pellis M, Muyldermans S, Casas-Delucchi CS, Cardoso MC, Leonhardt H, Hopfner K-P, Rothbauer U (2010) Modulation of protein properties in living cells using nanobodies. Nat Struct Mol Biol 17:133\u0026ndash;138\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun J, MacKinnon R (2020) Structural Basis of Human KCNQ1 Modulation and Gating. Cell 180:340\u0026ndash;347e9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarro-Soria R, Perez ME, Larsson HP (2015) KCNE3 acts by promoting voltage sensor activation in KCNQ1. Proc Natl Acad Sci U S A 112:E7286\u0026ndash;E7292\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchroeder BC, Waldegger S, Fehr S, Bleich M, Warth R, Greger R, Jentsch TJ (2000) A constitutively open potassium channel formed by KCNQ1 and KCNE3. Nature 403:196\u0026ndash;199\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarro-Soria R, Ramentol R, Liin SI, Perez ME, Kass RS, Larsson H (2017) P., KCNE1 and KCNE3 modulate KCNQ1 channels by affecting different gating transitions. Proc Natl Acad Sci U S A 114:E7367\u0026ndash;E7376\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun J, MacKinnon R (2017) Cryo-EM Structure of a KCNQ1/CaM Complex Reveals Insights into Congenital Long QT Syndrome. Cell 169:1042\u0026ndash;1050e9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbbott GW, Tai KK, Neverisky DL, Hansler A, Hu Z, Roepke TK, Lerner DJ, Chen Q, Liu L, Zupan B, Toth M, Haynes R, Huang X, Demirbas D, Buccafusca R, Gross SS, Kanda VA, Berry GT (2014) KCNQ1, KCNE2, and Na+-coupled solute transporters form reciprocally regulating complexes that affect neuronal excitability. Sci Signal 7:ra22\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGhosh S, Nunziato DA, Pitt GS (2006) KCNQ1 assembly and function is blocked by long-QT syndrome mutations that disrupt interaction with calmodulin. Circ Res 98:1048\u0026ndash;1054\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKasimova MA, Zaydman MA, Cui J, Tarek M (2015) PIP₂-dependent coupling is prominent in Kv7.1 due to weakened interactions between S4-S5 and S6. Sci Rep 5:7474\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y, Zaydman MA, Wu D, Shi J, Guan M, Virgin-Downey B, Cui J (2011) KCNE1 enhances phosphatidylinositol 4,5-bisphosphate (PIP2) sensitivity of IKs to modulate channel activity. Proc Natl Acad Sci U S A 108:9095\u0026ndash;9100\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLoussouarn G, Park KH, Bellocq C, Bar\u0026oacute; I, Charpentier F, Escande D (2003) Phosphatidylinositol-4,5-bisphosphate, PIP2, controls KCNQ1/KCNE1 voltage-gated potassium channels: a functional homology between voltage-gated and inward rectifier K\u0026thinsp;+\u0026thinsp;channels. Embo j 22:5412\u0026ndash;5421\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang H, He C, Yan X, Mirshahi T, Logothetis DE (1999) Activation of inwardly rectifying K\u0026thinsp;+\u0026thinsp;channels by distinct PtdIns(4,5)P2 interactions. Nat Cell Biol 1:183\u0026ndash;188\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChuang HH, Prescott ED, Kong H, Shields S, Jordt SE, Basbaum AI, Chao MV, Julius D (2001) Bradykinin and nerve growth factor release the capsaicin receptor from PtdIns(4,5)P2-mediated inhibition. Nature 411:957\u0026ndash;962\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRodr\u0026iacute;guez-Menchaca AA, Adney SK, Zhou L, Logothetis DE (2012) Dual Regulation of Voltage-Sensitive Ion Channels by PIP(2). Front Pharmacol 3:170\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDelmas P, Coste B, Gamper N, Shapiro MS (2005) Phosphoinositide lipid second messengers: new paradigms for calcium channel modulation. Neuron 47:179\u0026ndash;182\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuh BC, Leal K, Hille B (2010) Modulation of high-voltage activated Ca(2+) channels by membrane phosphatidylinositol 4,5-bisphosphate. Neuron 67:224\u0026ndash;238\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng Y, Liu H, Chen Y, Dong S, Wang F, Wang S, Li GL, Shu Y, Xu F (2022) Structural insights into the lipid and ligand regulation of a human neuronal KCNQ channel. Neuron 110:237\u0026ndash;247e4\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWillegems K, Eldstrom J, Kyriakis E, Ataei F, Sahakyan H, Dou Y, Russo S, Van Petegem F, Fedida D (2022) Structural and electrophysiological basis for the modulation of KCNQ1 channel currents by ML277. Nat Commun 13:3760\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMandala VS, MacKinnon R (2023) The membrane electric field regulates the PIP(2)-binding site to gate the KCNQ1 channel. Proc Natl Acad Sci U S A 120:e2301985120\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun J, MacKinnon R (2020) Structural Basis of Human KCNQ1 Modulation and Gating. Cell 180:340\u0026ndash;347e9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen H, Kim LA, Rajan S, Xu S, Goldstein SAJ (2003) N., Charybdotoxin binding in the IKs pore demonstrates two MinK subunits in each channel complex. 40:15\u0026ndash;23\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorin TJ, Kobertz WR (2008) J. P. o. t. N. A. o. S., Counting membrane-embedded KCNE β-subunits in functioning K\u0026thinsp;+\u0026thinsp;channel complexes. 105, 1478\u0026ndash;1482\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu X, Perez ME, Noskov SY, Larsson HP (2021) A general mechanism of KCNE1 modulation of KCNQ1 channels involving non-canonical VSD-PD coupling. Commun Biol 4:887\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuenze G, Vanoye CG, Desai RR, Adusumilli S, Brewer KR, Woods H, McDonald EF, Sanders CR, George AL, Jr.;, Meiler J (2020) Allosteric mechanism for KCNE1 modulation of KCNQ1 potassium channel activation. \u003cem\u003eElife\u003c/em\u003e 9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRodriguez-Menchaca AA, Adney SK, Tang QY, Meng XY, Rosenhouse-Dantsker A, Cui M, Logothetis DE (2012) PIP2 controls voltage-sensor movement and pore opening of Kv channels through the S4-S5 linker. Proc Natl Acad Sci U S A 109:E2399\u0026ndash;E2408\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiao J, Zhen XG, Yang J (2003) Localization of PIP2 activation gate in inward rectifier K\u0026thinsp;+\u0026thinsp;channels. Nat Neurosci 6:811\u0026ndash;818\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang H, Craciun LC, Mirshahi T, Roh\u0026aacute;cs T, Lopes CM, Jin T, Logothetis DE (2003) PIP(2) activates KCNQ channels, and its hydrolysis underlies receptor-mediated inhibition of M currents. Neuron 37:963\u0026ndash;975\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRobbins J (2001) KCNQ potassium channels: physiology, pathophysiology, and pharmacology. Pharmacol Ther 90:1\u0026ndash;19\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChoveau FS, De la Rosa V, Bierbower SM, Hernandez CC, Shapiro MS (2018) Phosphatidylinositol 4,5-bisphosphate (PIP(2)) regulates KCNQ3 K(+) channels by interacting with four cytoplasmic channel domains. J Biol Chem 293:19411\u0026ndash;19428\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTobelaim WS, Dvir M, Lebel G, Cui M, Buki T, Peretz A, Marom M, Haitin Y, Logothetis DE, Hirsch JA, Attali B (2017) Competition of calcified calmodulin N lobe and PIP2 to an LQT mutation site in Kv7.1 channel. Proc Natl Acad Sci U S A 114:E869\u0026ndash;e878\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu Y, Xu X, Gao J, Naffaa MM, Liang H, Shi J, Wang HZ, Yang ND, Hou P, Zhao W, White KM, Kong W, Dou A, Cui A, Zhang G, Cohen IS, Zou X, Cui J (2020) A PIP(2) substitute mediates voltage sensor-pore coupling in KCNQ activation. Commun Biol 3:385\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Q, Zhou P, Chen Z, Li M, Jiang H, Gao Z, Yang H (2013) Dynamic PIP2 interactions with voltage sensor elements contribute to KCNQ2 channel gating. 110:20093\u0026ndash;20098\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang Q, Corey RA, Hedger G, Aryal P, Grieben M, Nasrallah C, Baronina A, Pike ACW, Shi J, Carpenter EP, Sansom MS (2020) P., Lipid Interactions of a Ciliary Membrane TRP Channel: Simulation and Structural Studies of Polycystin-2. \u003cem\u003eStructure\u003c/em\u003e 28, 169\u0026ndash;184 e5\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePant S, Zhang J, Kim EC, Lam K, Chung HJ, Tajkhorshid E (2021) PIP2-dependent coupling of voltage sensor and pore domains in Kv7.2 channel. Commun Biol 4:1189\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNaughton FB, Kalli AC, Sansom MS (2016) Association of Peripheral Membrane Proteins with Membranes: Free Energy of Binding of GRP1 PH Domain with Phosphatidylinositol Phosphate-Containing Model Bilayers. J Phys Chem Lett 7:1219\u0026ndash;1224\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMuller MP, Jiang T, Sun C, Lihan M, Pant S, Mahinthichaichan P, Trifan A, Tajkhorshid E (2019) Characterization of Lipid-Protein Interactions and Lipid-Mediated Modulation of Membrane Protein Function through Molecular Simulation. Chem Rev 119:6086\u0026ndash;6161\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcInnes L, Healy J, Melville J (2018) J. a. p. a., Umap: Uniform manifold approximation and projection for dimension reduction\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGhojogh B, Ghodsi A, Karray F, Crowley M (2021) J. a. p. a., Uniform manifold approximation and projection (UMAP) and its variants: tutorial and survey\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKasimova MA, Zaydman MA, Cui J, Tarek M (2015) PIP(2)-dependent coupling is prominent in Kv7.1 due to weakened interactions between S4-S5 and S6. Sci Rep 5:7474\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIng\u0026oacute;lfsson HI, Lopez CA, Uusitalo JJ, de Jong DH, Gopal SM, Periole X, Marrink SJ (2014) The power of coarse graining in biomolecular simulations. 4:225\u0026ndash;248\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang L, Zhang Q, Tong HHY, Yao X, Liu H, Li G (2024) Computational methods for unlocking the secrets of potassium channels: Structure, mechanism, and drug design. 14, e1704\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarrink SJ, de Vries AH, Mark AE (2004) Coarse Grained Model for Semiquantitative Lipid Simulations. J Phys Chem B 108:750\u0026ndash;760\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarrink SJ, Risselada HJ, Yefimov S, Tieleman DP, de Vries AH (2007) The MARTINI Force Field: Coarse Grained Model for Biomolecular Simulations. J Phys Chem B 111:7812\u0026ndash;7824\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEnkavi G, Javanainen M, Kulig W, Rog T, Vattulainen I (2019) Multiscale Simulations of Biological Membranes: The Challenge To Understand Biological Phenomena in a Living Substance. Chem Rev 119:5607\u0026ndash;5774\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCorradi V, Sejdiu BI, Mesa-Galloso H, Abdizadeh H, Noskov SY, Marrink SJ, Tieleman DP (2019) Emerging Diversity in Lipid-Protein Interactions. Chem Rev 119:5775\u0026ndash;5848\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarrink SJ, Corradi V, Souza PCT, Ingolfsson HI, Tieleman DP, Sansom MS (2019) P., Computational Modeling of Realistic Cell Membranes. Chem Rev 119:6184\u0026ndash;6226\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuh BC, Hille B (2008) PIP2 is a necessary cofactor for ion channel function: how and why? Annu Rev Biophys 37:175\u0026ndash;195\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCzech MP (2000) PIP2 and PIP3: complex roles at the cell surface. Cell 100:603\u0026ndash;606\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRodr\u0026iacute;guez Menchaca AA, Adney SK, Zhou L, Logothetis DE (2012) Dual Regulation of Voltage-Sensitive Ion Channels by PIP2. 3\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuh B-C, Leal K, Hille B (2010) Modulation of High-Voltage Activated Ca\u003csup\u003e2+\u003c/sup\u003e Channels by Membrane Phosphatidylinositol 4,5-Bisphosphate. Neuron 67:224\u0026ndash;238\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim RY, Pless SA, Kurata HT (2017) PIP2 mediates functional coupling and pharmacology of neuronal KCNQ channels. 114:E9702\u0026ndash;E9711\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMandala VS, MacKinnon R (2023) The membrane electric field regulates the PIP\u003csub\u003e2\u003c/sub\u003e-binding site to gate the KCNQ1 channel. 120, e2301985120\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim H-J, Jeong M-H, Kim K-R, Jung C-Y, Lee S-Y, Kim H, Koh J, Vuong TA, Jung S, Yang H, Park S-K, Choi D, Kim SH, Kang K, Sohn J-W, Park JM, Jeon D, Koo S-H, Ho W-K, Kang J-S, Kim S-T, Cho H (2016) Protein arginine methylation facilitates KCNQ channel-PIP2 interaction leading to seizure suppression. eLife 5:e17159\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCooper EC, Abreo T, Tran B (2022) KCNQ channel PIP2 modulation: Two loose links, three rings, and a twist. Neuron 110:178\u0026ndash;180\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Q, Zhou P, Chen Z, Li M, Jiang H, Gao Z, Yang H (2013) Dynamic PIP2 interactions with voltage sensor elements contribute to KCNQ2 channel gating. Proc Natl Acad Sci U S A 110:20093\u0026ndash;20098\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePipatpolkai T, Mitrovic D, Cui J, Delemotte L (2023) PIP2 binding at the voltage sensor domain facilitates KCNQ1 VSD activation and gating. Biophys J 122:29a\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLoussouarn G, Park KH, Bellocq C, Bar\u0026oacute; I, Charpentier F, Escande D (2003) P PIP2 controls KCNQ1/KCNE1 voltagae gated potassium channels: a functional homology between voltage gated and inward rectifier K\u003csup\u003e+\u003c/sup\u003e channels. EMBO J 22:5412\u0026ndash;5421\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKasimova MA, Zaydman MA, Cui J, Tarek M (2015) PIP2-dependent coupling is prominent in Kv7.1 due to weakened interactions between S4-S5 and S6. Sci Rep 5:7474\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCorradi V, Mendez-Villuendas E, Ingolfsson HI, Gu RX, Siuda I, Melo MN, Moussatova A, DeGagne LJ, Sejdiu BI, Singh G, Wassenaar TA, Delgado Magnero K, Marrink SJ, Tieleman DP (2018) Lipid-Protein Interactions Are Unique Fingerprints for Membrane Proteins. ACS Cent Sci 4:709\u0026ndash;717\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCorey RA, Vickery ON, Sansom MSP, Stansfeld PJ (2019) Insights into Membrane Protein-Lipid Interactions from Free Energy Calculations. J Chem Theory Comput 15:5727\u0026ndash;5736\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCorey RA, Stansfeld PJ, Sansom MSP (2020) The energetics of protein-lipid interactions as viewed by molecular simulations. Biochem Soc Trans 48:25\u0026ndash;37\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKefauver JM, Hakala M, Zou L, Alba J, Espadas J, Tettamanti MG, Gajić J, Gabus C, Campomanes P, Estrozi LF, Sen NE, Vanni S, Roux A, Desfosses A, Loewith R (2024) Cryo-EM architecture of a near-native stretch-sensitive membrane microdomain. Nature 632:664\u0026ndash;671\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAnsell TB, Corey RA, Viti LV, Kinnebrew M, Rohatgi R, Siebold C, Sansom MSP (2023) The energetics and ion coupling of cholesterol transport through Patched1. Sci Adv 9:eadh1609\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArnold WR, Mancino A, Moss FR, Frost A, Julius D, Cheng Y (2024) Structural basis of TRPV1 modulation by endogenous bioactive lipids. Nat Struct Mol Biol 31:1377\u0026ndash;1385\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBartocci A, Grazzi A, Awad N, Corringer P-J, Souza PCT, Cecchini M (2024) A millisecond coarse-grained simulation approach to decipher allosteric cannabinoid binding at the glycine receptor α1. Nat Commun 15:9040\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKmiecik S, Gront D, Kolinski M, Wieteska L, Dawid AE, Kolinski A (2016) Coarse-Grained Protein Models and Their Applications. Chem Rev 116:7898\u0026ndash;7936\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChan KC, Lu L, Sun F, Fan J (2017) Molecular Details of the PH Domain of ACAP1(BAR-PH) Protein Binding to PIP-Containing Membrane. J Phys Chem B 121:3586\u0026ndash;3596\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSong W, Corey RA, Ansell TB, Cassidy CK, Horrell MR, Duncan AL, Stansfeld PJ, Sansom MS (2022) P., PyLipID: A Python Package for Analysis of Protein\u0026ndash;Lipid Interactions from Molecular Dynamics Simulations. J Chem Theory Comput 18:1188\u0026ndash;1201\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAnsell TB, Song W, Coupland CE, Carrique L, Corey RA, Duncan AL, Cassidy CK, Geurts MMG, Rasmussen T, Ward AB, Siebold C, Stansfeld PJ, Sansom MS (2023) P., LipIDens: simulation assisted interpretation of lipid densities in cryo-EM structures of membrane proteins. Nat Commun 14:7774\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSejdiu BI, Tieleman DP (2021) ProLint: a web-based framework for the automated data analysis and visualization of lipid\u0026ndash;protein interactions. Nucleic Acids Res 49:W544\u0026ndash;W550\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbraham MJ, Murtola T, Schulz R, P\u0026aacute;ll S, Smith JC, Hess B, Lindahl E (2015) GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1\u0026ndash;2, 19\u0026ndash;25\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarducci A, Bonomi M, Parrinello M, Metadynamics (2011) Wiley Interdisciplinary Reviews-Computational Mol Sci 1:826\u0026ndash;843\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBennett WF, Tieleman DP (2013) Computer simulations of lipid membrane domains. Biochim Biophys Acta 1828:1765\u0026ndash;1776\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarducci A, Bussi G, Parrinello M (2008) Well-Tempered Metadynamics: A Smoothly Converging and Tunable Free-Energy Method. Phys Rev Lett 100:020603\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang L, Li S, Xiang S, Liu H, Sun H (2024) Elucidating the Selective Mechanism of Drugs Targeting Cyclin-Dependent Kinases with Integrated MetaD-US Simulation. J Chem Inf Model 64:6899\u0026ndash;6911\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLaio A, Parrinello M (2002) Escaping free-energy minima. Proc Natl Acad Sci U S A 99:12562\u0026ndash;12566\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrimme S, Bannwarth C, Shushkov P, Robust A (2017) Accurate Tight-Binding Quantum Chemical Method for Structures, Vibrational Frequencies, and Noncovalent Interactions of Large Molecular Systems Parametrized for All spd-Block Elements (Z\u0026thinsp;=\u0026thinsp;1\u0026ndash;86). J Chem Theory Comput 13:1989\u0026ndash;2009\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLu T, Chen Q (2022) Independent gradient model based on Hirshfeld partition: A new method for visual study of interactions in chemical systems. J Comput Chem 43:539\u0026ndash;555\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGroenhof GJ (2013) B. s. m.; protocols, Introduction to QM/MM simulations. 43\u0026ndash;66\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSong W, Corey RA, Ansell TB, Cassidy CK, Horrell MR, Duncan AL, Stansfeld PJ, Sansom MS (2022) P., PyLipID: A Python Package for Analysis of Protein-Lipid Interactions from Molecular Dynamics Simulations. J Chem Theory Comput 18:1188\u0026ndash;1201\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTraag VA, Waltman L, Van Eck NJ (2019) J. S. r., From Louvain to Leiden: guaranteeing well-connected communities. 9, 1\u0026ndash;12\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAnuar SHH, Abas ZA, Yunos NM, Zaki NHM, Hashim NA, Mokhtar MF, Asmai SA, Abidin ZZ, Nizam AF (2021) Comparison between Louvain and Leiden algorithm for network structure: a review. In Journal of Physics: Conference Series, ; IOP Publishing: 2021; Vol. 2129; p 012028\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBader RF (1985) J. A. o. c. r., Atoms in molecules. 18, 9\u0026ndash;15\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLu H, Marti J (2020) Cellular absorption of small molecules: free energy landscapes of melatonin binding at phospholipid membranes. Sci Rep 10:9235\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLelimousin M, Limongelli V, Sansom MSP (2016) Conformational Changes in the Epidermal Growth Factor Receptor: Role of the Transmembrane Domain Investigated by Coarse-Grained MetaDynamics Free Energy Calculations. J Am Chem Soc 138:10611\u0026ndash;10622\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Stepwise Integrated Multi-scale Dynamics and Advanced Analysis, Protein-lipid Interaction, PIP2-KCNQ1 binding","lastPublishedDoi":"10.21203/rs.3.rs-6059518/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6059518/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eVoltage-gated potassium channel KCNQ1 (Kv7.1) is essential for various physiological processes, including cardiac repolarization, epithelial ion transport, and inner ear function. Its functional versatility arises from interactions with auxiliary KCNE proteins, calmodulin (CaM), and the lipid phosphatidylinositol 4,5-bisphosphate (PIP2), which modulate its gating properties, trafficking, and activity in a tissue-specific manner. Despite advancements in structural and functional studies, the precise molecular mechanisms underlying PIP2's role in KCNQ1 activation, as well as the contribution of KCNE3 and CaM to PIP2-KCNQ1 binding, remain unclear. Here, we present the Stepwise Integrated Multi-scale Dynamics and Advanced Analysis (SIMDA) framework, which incrementally integrates coarse-grained and all-atom molecular dynamics, quantum mechanics, and well-tempered metadynamics, along with advanced clustering and energy analysis techniques. Over 280 \u0026micro;s multi-scale simulations revealed eight PIP2-binding sites, including new regions on the S0 segment and the S6-HA junction. We also observed KCNE3 enhances the \u0026ldquo;twist\u0026rdquo; effect at KCNQ1\u0026rsquo;s C-terminal, promoting PIP2 binding. Furthermore, eight PIP2 dissociation pathways revealed transitions across binding sites, which highlight its dynamic transfer behavior. These findings provide a comprehensive understanding of PIP2-mediated regulation of KCNQ1 and establish SIMDA as a robust tool for studying lipid-protein dynamics.\u003c/p\u003e","manuscriptTitle":"Revealing the Molecular Mechanisms of PIP2 Binding and Regulating KCNQ1: Twists, Links, and Binding-Site Transfers via the Developed SIMDA Strategy","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-03-07 09:02:03","doi":"10.21203/rs.3.rs-6059518/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"7a2f3c38-38e8-48e0-96d0-2d9de2a1d67d","owner":[],"postedDate":"March 7th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":45311977,"name":"Biological sciences/Biophysics/Membrane biophysics"},{"id":45311978,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":45311979,"name":"Biological sciences/Molecular biology"}],"tags":[],"updatedAt":"2026-04-15T18:20:15+00:00","versionOfRecord":[],"versionCreatedAt":"2025-03-07 09:02:03","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6059518","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6059518","identity":"rs-6059518","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}