A molecular simulation method for studying the effects of phospholipid structure on biological membrane fission

Abstract

The present invention provides a molecular simulation method for studying the effect of phospholipid structure on biological membrane fission. The method uses the MD (Molecular Dynamics) simulation method to model the membrane tube system, accurately constructs the membrane tube fission model of different phospholipid components, explores the microscopic dynamic process of membrane fission, and systematically analyzes the microscopic action mechanism of different phospholipid structures on the fission process. The invention helps to explain the guiding relationship between phospholipid structure and fission path selection at the molecular scale, and can deepen the current theoretical research on phospholipid structure. The implementation of this computational simulation method will also help explain how phospholipid structure regulates the fission process, lay a theoretical foundation for a deep understanding of the diversity of phospholipid structure and the specific functions of biological membranes, and provide research directions for the study and mechanism exploration of diseases such as Parkinson's disease.

Description

Technical Field

[0001] The invention relates to the technical field of molecular simulation, in particular to a molecular simulation method for studying the influence of phospholipid structure on biological membrane fission. Background Art

[0002] Vesicle transport is an important way of exchanging substances inside and outside cells. When the vesicle transport function is abnormal, the organelle activities and cell functions related to this function will be affected, thus causing some biological diseases. The process of vesicle transport is accompanied by the fission of biological membranes, and phospholipids are the basic skeleton of membranes. Therefore, the behavior of phospholipids in the fission process has attracted more and more attention from researchers.

[0003] The general form of phospholipids is a hydrophilic head connected to two hydrophobic hydrocarbon tails. Phospholipid membranes composed of different polar head groups and tail chains show differences in splitting. The different morphological ratios of the head group and tail chain lead to differences in the thickness, curvature, and deformation ability of the membrane. The different unsaturation and length of the tail chain lead to different properties such as membrane stiffness and stacking defects. The physical properties of the membrane and the interaction between phospholipids will cause different phenomena in the splitting process. Although experimental techniques have been continuously developed and improved in recent years, providing an unprecedented perspective for the microstructural observation of the splitting process and the extraction of phospholipid arrangement features, it is difficult to accurately capture the dynamic details of phospholipid molecules and the potential energy changes in the splitting process using conventional experimental methods, and a new technical method is urgently needed. Summary of the invention

[0004] In view of this, the present invention aims to propose a molecular simulation method for studying the influence of phospholipid structure on biological membrane fission, and to study the computational technology of the time evolution of microscopic systems under classical mechanics through numerical simulation methods. The present invention adopts the MD simulation method under the Martini coarse-grained force field, constructs a membrane tube contraction and fission model composed of phospholipids with different structures, and combines the umbrella-enhanced sampling method to systematically explore the influence of phospholipid structure on the fission process, and clarifies the changing laws of various phospholipid structures in the fission process and the microscopic mechanism acting on the fission.

[0005] To achieve the above object, the technical solution of the present invention is achieved as follows:

[0006] A molecular simulation method for studying the effect of phospholipid structure on biomembrane fission, the method comprising the following steps:

[0007] (1) Constructing membrane tube splitting models with different structural phospholipid components:

[0008] Phospholipids with different structures were selected according to the head group, tail chain length, and hydrocarbon chain unsaturation. Small module membrane systems with different phospholipid combinations were first constructed using the CHARMM-GUI website to reflect the compositional diversity of biological membranes. Then, the small module membrane systems were expanded to the required size using GROMACS, and several groups of different membrane tube models were constructed using the BUMPY script tool. Finally, coarse-grained water droplets were used to simulate probes and probe rings were constructed around the membrane tube model to obtain a membrane tube splitting model.

[0009] (2) Simulating splitting and shrinking behavior:

[0010] GROMACS was used to simulate the splitting and contraction behavior: first, a constant force was applied to each probe toward the center point of the membrane tube at the horizontal plane where the probe was located to ensure that the probe could contract evenly until the lipids in the inner layer contacted each other. This process was called the mechanical contraction stage. Then, the probe in the last frame of the mechanical contraction stage was continuously pulled until the membrane tube split. This process was called the splitting completion process.

[0011] (3) Mechanism analysis:

[0012] Perform umbrella sampling on the two processes of step (2), and analyze the results of umbrella sampling using the gmx wham module in GROMACS;

[0013] VMD visualization software is used to capture the dynamic process and details of membrane splitting, and VMD's built-in script is used to analyze the overall evolution rate and change behavior during membrane splitting.

[0014] Furthermore, the membrane layer of each group of membrane tubes includes common components and variable components. The common components are the phospholipids shared by all groups of membrane tubes, and the variable components are the phospholipids that distinguish each group of membrane tubes from other membrane tubes. The variable components are divided into head group groups, chain length groups and unsaturation groups according to the selected phospholipid structures. The ratio of the added numbers of the common components and the variable components is in the range of 1:1-1.5.

[0015] Furthermore, the membrane tube model was established using the BUMPY script tool, the Martini3 coarse-grained force field was selected, the temperature was set to 310 K to simulate the temperature environment in the body, and the membrane tube was periodically docked up and down.

[0016] Furthermore, the initial membrane tube model needs to undergo energy minimization, pre-equilibrium simulation and equilibrium simulation to obtain a stable membrane tube model; after obtaining the stable membrane tube model, insert a probe, the probe is facing the center of the membrane tube, a 5nm water layer and ions are added above and below the membrane tube, and then energy minimization, pre-equilibrium simulation and equilibrium simulation are performed to obtain a stable membrane tube splitting model before shrinkage and splitting.

[0017] Furthermore, energy minimization was performed using two 5000-step steepest descents;

[0018] The pre-equilibrium simulation was performed 5 times under isothermal and isobaric ensemble, each time for 1 ns, with the time step gradually increasing and the position restriction on the head group reduced in each pre-equilibrium;

[0019] The equilibrium simulation before adding the probe was a 200 ns equilibrium simulation in the NPT ensemble, using the Parrinello-rahman pressure control method and a V-rescale thermal bath;

[0020] The equilibrium simulation after adding the probe was a 1 ns equilibrium simulation under the NPT ensemble, using the Parrinello-rahman pressure control method and a V-rescale thermal bath.

[0021] Furthermore, a probe ring is composed of 10-16 evenly distributed probes, each of which is composed of 64 water beads arranged in a 4×4×4 cube. The beads are connected by 144 direct lattice bonds, 216 planar diagonal bonds, and 108 diagonal bonds. The harmonic force constant of each bond is 1250 kJ·nm -2 ·mol -1 .

[0022] Furthermore, the mechanical contraction process is a uniform stretching dynamics simulation of each probe, and the probe movement speed is 0.001 nm×ps -1 ;

[0023] The splitting process is to relax the end state of the mechanical contraction process, that is, firstly -1 The probe is pulled at a speed to advance 0.1 nm, and then a force is used to fix the probe to relax and deform the membrane tube for 50 ns. This process is repeated until the membrane tube splits.

[0024] Furthermore, the conditions for the relaxation of the splitting completion process after completing one traction are as follows: a force potential with a velocity of 0 is applied to each probe under the PR pressure bath.

[0025] Furthermore, the specific method of umbrella sampling in step (3) is:

[0026] During the mechanical contraction stage, the probe displacement distance was used as the reaction coordinate, with a window of every 0.2 nm to ensure overlap between windows to obtain a continuous energy function;

[0027] In the splitting completion stage, the relaxed configuration was taken as the initial window configuration, the reaction coordinates of adjacent window configurations were 0.1 nm apart, the sampling simulation parameters followed the relaxation simulation parameters and restrictions, and the sampling time was 20 ns.

[0028] The present invention also provides an application of the molecular simulation method as described above for studying the influence of phospholipid structure on biomembrane fission in studying the influence law and microscopic mechanism of phospholipid structure on biomembrane fission from a microscopic perspective.

[0029] Compared with the prior art, the molecular simulation method for studying the effect of phospholipid structure on biomembrane fission described in the present invention has the following advantages:

[0030] (1) The molecular simulation method for studying the effect of phospholipid structure on biological membrane fission described in the present invention can realize the use of different phospholipid structures to model membrane tubes and simulate the fission process, overcome the difficulty of analyzing the fission process and molecular dynamic details, and realize the correspondence between structural differences and free energy barrier differences, meeting the quantitative investigation needs of phospholipid structure action mechanism analysis. The method is simple to operate, has strong applicability, and the phospholipid composition and membrane component composition are adjustable. It is suitable for the reproduction of various biological systems, making up for the shortcomings of cumbersome operation, long cycle and high cost in experimental research; at the same time, the use of coarse-grained force fields can also improve the calculation efficiency and extend the space and time scale of simulation. In general, the present invention can deeply explore the molecular action mechanism of phospholipid structure in the fission process, and provide a convenient and effective way to explore the pathogenic mechanism related to lipids.

[0031] (2) The present invention analyzes the microscopic mechanism of the fission process from two aspects: the overall membrane layer and individual phospholipids. Because fission is highly correlated with the physical properties of the membrane, the overall membrane area, the thickness of the inner and outer membranes, and the rigidity and flexibility of the membrane all need to be quantitatively measured. The rigidity and flexibility of the membrane and a series of physical properties were obtained through script analysis tools such as FATSLiM and membrane fluctuation calculation, as well as the analysis modules provided by GROMACS (gmx density, etc.). The second stage of analysis focuses on the individual level. Since the phospholipid structures in the second stage will frequently contact and have more inter-chain and head group interactions, the analysis focuses on the quantitative analysis of the individual area (APL) of phospholipids, thickness changes, tail chain insertion performance, head group contact times, and phospholipid arrangement order. By analyzing the phospholipid structures of the two stages and combining the analysis results with the free energy barrier, the mechanism of the influence of phospholipid structure on the fission process is summarized, that is, structural differences lead to morphological differences of phospholipids, which in turn changes the spatial and aggregation state of phospholipids, affects the interaction between phospholipids, and thus affects the fission process. This analysis deepens the understanding of the fission process and analyzes the mechanism of the influence of phospholipids on fission from a rare perspective. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:

[0033] Figure 1Schematic diagram of the membrane tube splitting model structure, taking the components DOPC:DPPA (1:1) as an example; (a) is the front view of the membrane tube splitting model; (b) is the top view of the membrane tube splitting model, the dotted line mark beads are fixed W beads, which serve as a reference for the probe pulling direction during the contraction process; each probe contracts in the direction of the reference bead, and the forward direction is indicated by the dotted arrow; (c) is a coarse-grained schematic diagram of the phospholipids that make up the membrane tube, including common components and variable components; coarse-grained bead scheme: DOPC head group choline, DPPE head group ethanolamine, DPPG head group glycerol, DPPS head group serine;

[0034] Figure 2 The overall change diagram of the membrane tube morphology during the splitting and shrinking process. Taking the membrane tube system composed of DOPC and DPPA as an example, the probe, water and ions are hidden; (a) is the dynamic change diagram of the membrane tube splitting process, and (b) is the cross-sectional diagram of the membrane tube during the splitting process;

[0035] Figure 3 The free energy difference diagrams of membrane tube splitting composed of phospholipids with different structures; (a), (c), and (e) are the free energy difference diagrams of membrane tube splitting in the mechanical contraction stage of chain length group, unsaturation group, and head group, respectively; (b), (d), and (f) are the free energy difference diagrams of membrane tube splitting in the splitting completion stage of chain length group, unsaturation group, and head group, respectively;

[0036] Figure 4 The diagrams are the membrane thickness and area per lipid of phospholipids with different structures; (a) is the membrane thickness diagram in the mechanical contraction stage and the completion stage of the splitting system; (b) is the statistical diagram of the area per lipid in the mechanical contraction stage and the completion stage of the splitting system;

[0037] Figure 5 The insertion analysis diagram of the membrane tube splitting completion stage composed of phospholipids with different structures, taking the component DOPC:DTPA (1:1) as an example; (a), (b), (c) are the insertion frequency statistics of the variable components in each group of membrane tubes, and (d) is a schematic diagram of shallow insertion and deep insertion;

[0038] Figure 6 The diagrams are the bending stiffness and phospholipid enrichment index of membrane layers composed of phospholipids with different structures; (a), (b), and (c) are the stiffness statistics of different membrane layers; (d), (e), and (f) are the enrichment index diagrams of variable components in different membrane layers;

[0039] Figure 7 The figure is a statistical diagram of the contact number of the head groups of the membrane layer composed of phospholipids with different head groups during contraction;

[0040] Figure 8 This is a statistical diagram of the lipid arrangement order of membrane layers composed of different phospholipids. DETAILED DESCRIPTION

[0041] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] All molecular dynamics simulations involved in the present invention were completed using the open source simulation software GROMACS 2019.6, the model was built using the BUMPY script tool and Packmol software, and post-processing analysis was performed using VMD 1.9.3 software, FATSLIM and other analysis scripts.

[0044] Example 1 Construction of membrane tube splitting models with different structural phospholipid components

[0045] In order to better understand the mechanism of the influence of phospholipid structure on the division process, membrane tubes with different structural phospholipid components were constructed to reflect the compositional diversity of biological membranes.

[0046] The membrane layer of the membrane tube includes common components and variable components. The common components are phospholipids shared by each group of membrane tubes. In this example, DOPC phospholipids (1,2-dioleoyl-sn-glycero-3-phosphocholine) are selected. The variable components are phospholipids that distinguish each group of membrane tubes from other membrane tubes. The variable components select phospholipids with different structures according to the head group, tail chain length, and hydrocarbon chain unsaturation. Specifically, the chain length group selects DTPA, DLPA, and DPPA. The tail chain length ratio of DTPA, DLPA, and DPPA is 7:9:11. The unsaturation group selects DOPA, DAPA, and DRPA. The unsaturation ratio of DOPA, DAPA, and DRPA is 1:4:6. The head group group selects DPPE, DPPG, and DPPS. The DPPE head group is ethanolamine, the DPPG head group is glycerol, and the DPPS head group is serine. The mutual attraction ability of the head groups is: DPPE-DPPE (NH3)>DPPS-DPPS (CNO)>DPPG-DPPG (GL0). The structures of the common components and variable components are as follows Figure 1 As shown in (c), the ratio of the number of added common components and variable components is 1:1.

[0047] Table 1 Specific names of common components and variable components

[0048]

[0049] First, use the membrane layer construction module on the CHARMM-GUI website to select the appropriate system ratio, fix the phospholipid ratio of the upper and lower layers, set the appropriate system temperature to 310K, and construct a small module membrane system. Then, use the GROMACS built-in command to expand the small module to the required size of the system. In this example, the small membrane block is expanded 8 times in the X and Y directions.

[0050] The system at this time is used as an input file, and the BUMPY script tool is used to construct a membrane tube of the required length and radius for the study. The length of the membrane tube constructed in this embodiment is 40nm and the radius is 12nm. The Martini3 coarse-grained force field is selected, and the temperature is set to 310K to simulate the temperature environment in the body. The membrane tube is docked periodically up and down. The script tool will use the statistical experience of the number of lipids inside and outside the deformed membrane layer as the proportion basis, and will reserve hydrophobic holes for subsequent membrane tube equilibrium simulation.

[0051] In this embodiment, a total of nine groups of models are constructed, that is, three groups of models are set for each of the three types (such as head group, tail chain length, and hydrocarbon chain unsaturation), specifically including: DOPC+DTPA, DOPC+DLPA, DOPC+DPPA, DOPC+DOPA, DOPC+DAPA, DOPC+DRPA, DOPC+DPPE, DOPC+DPPG, DOPC+DPPS. The constructed membrane tube model is as follows Figure 1 (a) and Figure 1 (b) shown.

[0052] When the membrane tube was subjected to equilibrium simulation, no water layer was added above and below the membrane tube, which was periodic. The initial membrane tube model was first constructed by using the steepest descent method twice with 5000 steps to minimize the energy, and then 5 pre-equilibrium simulations were performed in the isothermal and isobaric ensemble, each time for 1ns pre-equilibrium, the time step gradually increased, the simulation time step was 2fs to 20fs, and the position restriction on the tail chain was reduced in each pre-equilibrium. The electrostatic interaction was calculated using the reaction field method proposed by CHARMM-GUI MartiniMaker with a dielectric constant of 15 and a cutoff distance of 1.1nm, and the calculation method of the van der Waals interaction had a cutoff distance of 1.1 nm. Finally, a 200ns equilibrium simulation was performed in the NPT ensemble, using the Parrinello-rahman pressure control method and the V-rescale heat bath.

[0053] After obtaining a stable membrane tube model, the position of the membrane tube is restricted up and down, and a probe is inserted, the probe is facing the center of the membrane tube, and a 5nm water layer and ions are added to the top and bottom of the membrane tube. After the same energy minimization and pre-equilibrium simulation, a 1ns equilibrium simulation is completed to obtain a stable membrane tube splitting model before contraction and splitting. During the first membrane equilibrium, the membrane tube height is consistent with the system height to achieve periodic membrane tube modeling.

[0054] Among them, the construction of the probe ring is based on the basic morphology and characteristics of motor proteins. Since the function of motor proteins is relatively complex and contains multiple domains, in order to realize the contraction and shearing effect of motor proteins on the membrane layer, the present invention uses coarse-grained water droplets to construct probe rings to simulate the function of motor proteins. A probe ring is composed of 12 probes arranged symmetrically, with the central angle of the center of adjacent probes being 30°, and the radius of the probe ring being slightly larger than the radius of the membrane tube. Each probe is composed of 64 water droplets arranged in a 4×4×4 cube. The beads are connected by 144 direct lattice bonds (0.3nm), 216 planar diagonal bonds and 108 diagonal bonds, and the simple harmonic force constant of each bond is 1250 kJ·nm -2 ·mol -1 .

[0055] Example 2: Simulation and analysis of splitting and shrinking behavior using GROMACS

[0056] 1. Simulate splitting and shrinking behavior

[0057] Mechanical contraction stage: First, a constant force is applied to each probe toward the center of the membrane tube in the horizontal plane to ensure that the probe can contract evenly. The contraction time step is 10fs. A W bead at the center of the probe ring is fixed as a traction reference using a position constraint. Only the X\Y direction of the probe is allowed to be subjected to force (the Z direction is the length direction of the membrane tube). The center of mass of the 12 probes is stretched, and the traction direction is toward the reference bead. The traction force constant is set to 3000 kJ·mol -1 ·nm -1 , the pulling speed is 0.001 nm·ps -1 The whole process continues until the lipids in the inner layer come into contact with each other. This stage of contraction is called the mechanical contraction stage. The membrane tube in the mechanical contraction stage deforms relatively quickly, and the lipids do not have time to rearrange in the membrane layer.

[0058] Splitting completion process: The second stage begins to slowly shrink. The probe in the last frame of the previous stage is pulled at a speed of 0.00002 nm·ps. -1 The total pulling distance is 0.1nm. After completing one pulling, the model needs to be relaxed. Under the PR pressure bath, a force potential with a speed of 0 is applied to each probe (as a preparatory simulation for the enhanced sampling window), and the relaxation time is 50ns. At this time, the lipids will begin to shrink and generate curvature, and the lipids in contact with each other in the inner layer will also interact due to accumulation. A total of 24 relaxations were performed on this group of systems, and finally the splitting was completed.

[0059] 2. Enhanced sampling simulation

[0060] Umbrella is used as the bias potential type to perform umbrella sampling in the two contraction stages. The sampling window is selected in a different way. In the mechanical contraction stage, the probe displacement distance is used as the reaction coordinate, and a window is set every 0.2nm. This ensures that the windows overlap to obtain a continuous energy function. For the splitting completion stage, the relaxed configuration is taken as the initial window configuration, and the reaction coordinates of adjacent window configurations are 0.1nm apart. The sampling simulation parameters use the relaxed simulation parameters and restrictions, and the sampling time is 20ns. After all sampling is completed, the sampling window configuration histogram is used to determine whether the sampling is sufficient.

[0061] 3. Data Analysis

[0062] The VMD visualization software can be used to capture the dynamic process and details of membrane splitting. The Tcl script provided by VMD can be used to analyze the overall evolution rate and change behavior during the membrane splitting process.

[0063] In order to compare the effects of various structures on the splitting barrier, the gmx wham module in GROMACS was used to analyze the results of umbrella sampling. The free energy barriers in the contraction stage were compared, and the key factors were extracted in combination with dynamic details to provide analytical directions for subsequent mechanism interpretation.

[0064] like Figure 2 As shown in the figure, the dynamic changes of the membrane layer during splitting are shown by taking the membrane tube system composed of DOPC and DPPA as an example. Figure 2 (b) The entire contraction process is divided into a mechanical contraction stage and a splitting completion stage, with the contact between the inner and outer phospholipids as the boundary. In the mechanical contraction stage, the membrane tube deforms uniformly, and the curvature of the deformed part gradually increases. In the splitting completion stage, the inner phospholipids of the membrane layer come closer to each other under the action of contraction. At the same time, due to tension and hydrophobicity, the phospholipids are pulled to both sides, and then a semi-split state is formed; the phospholipids will accumulate at the neck, and the arrangement of the phospholipids at the neck is more disordered than that at the slightly deformed part.

[0065] After completing the contraction simulation of all systems, the displacement distance of the probe from the initial position to the position when the splitting is completed is used as the reaction coordinate, and the PMF curve is calculated using umbrella sampling to obtain the free energy barrier for the splitting of the membrane tube system, such as Figure 3 shown.

[0066] The free energy change of the chain length group during the mechanical contraction stage is as follows: Figure 3 (a) As the chain length increases, the free energy barrier increases. The free energy of the chain length group at the completion stage of splitting is shown in Figure 3 (b) As the chain length increases, the free energy increases, and the difference between the DTPA group and the DLPA group is small.

[0067] The free energy change of the unsaturated group during the mechanical contraction stage is as follows: Figure 3(c) As the degree of unsaturation increases, the DOPA group has the highest free energy barrier, while the DAPA group has the lowest. The free energy changes of the unsaturated groups during the splitting process are shown in Figure 2. Figure 3 (d) As the degree of unsaturation increases, the free energy barrier is highest in the DRPA group and lowest in the DAPA group.

[0068] The free energy change of the head group during the mechanical contraction stage is as follows: Figure 3 (e) The free energies of the three groups are similar, with the highest free energy barrier being the DPPE group and the lowest being the DPPG group. The free energy changes of the head group at the completion stage of splitting are shown in Figure 2. Figure 3 (f), the highest is the DPPE group, and the lowest is the DPPS group. The free energy curves of the DPPS and DPPG groups show a platform in the middle, which may be related to the head group interaction.

[0069] Figure 4 The membrane tube thickness and lipid area of ​​different phospholipid structures are shown in Figure 1. Figure 4 As can be seen in (a), the longer the chain length, the thicker the membrane thickness; the thickness does not decrease with the increase of unsaturation; the difference between the head groups does not significantly affect the membrane thickness. The membrane thickness is related to the deformation ability of the membrane tube and the diffusion ability of phospholipids. Generally speaking, the ability of phospholipids to move in thicker membrane layers will decrease. Figure 4 (b) It can be seen that when the tail chain is saturated, the area per lipid of the chain length group and the head group is not much different, but as the degree of unsaturation increases, the area per lipid of the highly unsaturated phospholipid increases significantly. Generally speaking, the larger the area of ​​a single phospholipid, the higher the movement space of the phospholipid and the spreading area of ​​the tail chain, and the stronger the ability to split, open a gap and move.

[0070] Figure 5 This is an insertion analysis diagram of the completion stage of membrane tube splitting composed of phospholipids with different structures. It can also be called the thickness contribution rate, which refers to the contribution of the tail chain arrangement of a single phospholipid to the normal thickness of the membrane. By establishing a reference vector for the line connecting the phospholipid head group to the phospholipid core, and then establishing a vector from the head group to the end atom of the tail chain, the vertical projection distance of the vector relative to the reference vector direction is obtained, and this distance is divided by the thickness of a single phospholipid to obtain the thickness contribution rate of a single phospholipid. The thickness contribution rate is related to the mobility of phospholipids. The phospholipids with higher thickness contribution rates tend to be inserted deeper into the membrane layer, and the tail chains are arranged more compactly, which makes it difficult for phospholipids to move; the phospholipids with lower thickness contribution rates tend to move on the surface of the membrane layer, which is conducive to opening the gap in the membrane and promoting splitting.

[0071] from Figure 5 (a) It can be seen that the shorter the chain length of phospholipids, the more they tend to be inserted shallowly. As the chain length increases, the proportion of deep insertion increases. Figure 5(b) It can be seen that the thickness contribution rate of phospholipids containing unsaturation is relatively high, and the tendency of phospholipids with high unsaturation to be deeply inserted is the highest. Figure 5 (c) It can be seen that the difference between head groups has little effect on the thickness contribution rate, but the DPPE phospholipid with the smallest head group has the lowest contribution rate above 30%, which should be related to the ratio of head group size and tail chain length. Figure 5 (d) Schematic diagram of shallow insertion and deep insertion.

[0072] Figure 6 The bending stiffness diagram and phospholipid enrichment index diagram of the membrane layer composed of phospholipids with different structures are shown in Figure 2. Figure 6 (a) It can be seen that as the chain length increases, the bending stiffness increases. Figure 6 (b) It can be seen that as the degree of unsaturation increases, the bending stiffness of the membrane decreases, but when the phospholipid has a high degree of unsaturation, the bending stiffness of the membrane increases slightly. Figure 6 (c) It can be seen that when the phospholipid tail chain is a saturated hydrocarbon chain, the bending stiffness of the phospholipid membrane head group is very high, among which DPPS has the lowest bending stiffness.

[0073] The phospholipid enrichment index is the number of phospholipids within the cutoff distance of the same phospholipid. The higher the number, the higher the number of interactions between the head groups or tail chains of the phospholipids. Figure 6 (d) It can be seen that as the chain length increases, phospholipids tend to aggregate more, which leads to stronger interchain interactions. Figure 6 (e) It can be seen that the enrichment index of phospholipids first decreases and then increases with the increase of unsaturation. Combined with the analysis of each lipid area, the higher the unsaturation of phospholipids, the higher the area of ​​phospholipids, so the enrichment index of phospholipids will decrease. However, when the unsaturation continues to increase, the chain is too soft, which will lead to entanglement between chains and aggregation of phospholipid tail chains, and the enrichment index will increase instead.

[0074] Figure 7 The statistical diagram of the contact number of the head groups of the membrane layer composed of phospholipids with different head groups during contraction can reflect the interaction frequency between different head groups and analyze whether it is ipsilateral or contralateral aggregation in combination with the enrichment situation. Figure 7 It can be seen that DPPE has a higher frequency of occurrence at low contact numbers, while DPPG and DPPS have basically the same basic number and basic frequency and are higher than DPPE.

[0075] Figure 8 This is a statistical diagram of lipid order parameters of membrane layers composed of different phospholipids. The diagram can reflect the arrangement of phospholipids in the lipid layer. As can be seen from the figure, as the chain length increases, phospholipids tend to be arranged regularly in the membrane; as the degree of unsaturation increases, phospholipids tend to be arranged disorderly, and the interchain entanglement is stronger at this time; as the head group changes, the arrangement of phospholipids in the membrane is not much different.

[0076] By combining the basic physical properties of the membrane (such as Figure 4 , Figure 6 (a) Figure 6 (b) Figure 6 (c) Insertion of phospholipids (e.g. Figure 5 ), arrangement (such as Figure 8 ) and aggregation properties (e.g. Figure 6 (d) Figure 6 (e) Figure 6 (f)) By comparison, the molecular mechanism by which phospholipid structure affects the free energy barrier of splitting was obtained.

[0077] During the mechanical contraction stage, the physical properties of the membrane determine the free energy barrier for the splitting. As the chain length of the variable component increases, the free energy barrier for the mechanical contraction stage also increases, which is consistent with the high thickness of the membrane composed of long-chain phospholipids ( Figure 4 (a)), high hardness ( Figure 6 (a)). The longer the chain length, the harder it is for the membrane tube to deform, and the higher the free energy barrier ( Figure 3 (a)). Similarly, the change of free energy barrier of head group and unsaturation group in mechanical contraction stage can be analyzed. When the unsaturation is higher, the bending stiffness of the membrane is lower, the membrane layer is thinner, and the free energy barrier of the component is also lower ( Figure 3 (c)). For the head group, the head group has almost no effect on the physical properties of the membrane. Therefore, their same tail chain properties make the free energy barriers of the head group splitting almost the same during the mechanical contraction stage. The DPPE group with the highest thickness and higher bending stiffness has the highest free energy barrier, while the DPPG group with the lowest thickness has the lowest free energy barrier.

[0078] At the completion stage of cleavage, the interactions between lipids are more obvious, and the rearrangement and aggregation of variable components also make the probe contraction subject to different resistances. For chain length groups and unsaturation groups, different phospholipids have different insertion properties (e.g. Figure 5 (a) Figure 5 (b)). The longer the chain length of phospholipids, the more they tend to be arranged regularly and deeply inserted. The long-chain lipids have a small area per lipid and a high thickness contribution rate. The chain tails tend to be arranged in a straight chain and inserted into the membrane layer, so their mobility and rearrangement abilities are reduced. The higher the unsaturation, the more they tend to be arranged in a spread arrangement in terms of their area per lipid and insertion. As the unsaturation increases, the deeper the insertion into the membrane layer, the larger the area per lipid, and the higher the degree of mutual aggregation, which means that the long-chain phospholipids or highly unsaturated phospholipid chains will be entangled with each other (such as Figure 6 (d) Figure 6 (e) Figure 8). Interchain entanglement will cause the probe to provide higher energy to overcome the interchain interaction, so the membrane containing long-chain phospholipids or highly unsaturated phospholipids has a higher free energy barrier for splitting. It is worth noting that compared with low-unsaturated phospholipids, phospholipids will obtain a larger area when they have an appropriate degree of unsaturation ( Figure 4 (b) This provides space for the probe to shrink and the phospholipid to move, and the entanglement between the chains is also lower than that under high unsaturation, which reduces the free energy barrier to a certain extent. For the head group, the difference in head groups makes the number of interactions and the degree of aggregation between head groups vary greatly ( Figure 7 , Figure 6 (f)), DPPE with the smallest head group has the lowest phospholipid area and the lowest number of contacts between head groups, and DPPE tends to aggregate more. These factors lead to the existence of DPPE being conducive to same-side aggregation, so when the probe shrinks, the inner phospholipid contact cannot provide a stronger interaction, and the membrane gap is difficult to open, so DPPE has the highest free energy barrier. Other differences between DPPG and DPPS phospholipids are unlikely to cause differences in the free energy barrier of splitting.

[0079] Through the analysis of the above results, the mechanism of the effect of phospholipid structure on the free energy barrier of fission was obtained, and the microscopic mechanism of chain length, unsaturation and head group structure promoting fission was summarized as follows:

[0080] For the chain length group, as the chain length of the variable component increases, its free energy barrier will increase in both the mechanical contraction stage and the splitting completion stage. The mechanical contraction stage is mainly because the longer the chain length, the thicker the membrane thickness, the more compact the lipid accumulation, and the lower the bending and deformation ability of the membrane. In the splitting completion stage, as the chain length increases, the fluidity of the lipid decreases, and the lipid tail chain tends to be inserted and arranged in the direction of the membrane thickness.

[0081] For the unsaturation group, the order of free energy barriers of the three groups in the mechanical contraction stage and the splitting completion stage is the same. With the increase of unsaturation, the free energy barrier does not increase linearly but has a peak at a moderate degree of unsaturation. This is because increasing the unsaturation will increase the flexibility of the chain to a certain extent, expand the area of ​​phospholipids, thereby improving the fluidity of lipids, making it easier for neck lipids to move and open gaps, and thus reducing the free energy barrier. However, increasing the unsaturation will also cause interchain entanglement, increase interchain interactions, and aggregate tail chains, thereby increasing the free energy barrier. Therefore, unsaturation exists as an antagonistic factor.

[0082] For the head group, the order of free energy barriers of the three groups in the mechanical contraction stage and the splitting completion stage is different. In the mechanical contraction stage, when the bending stiffness of the three groups is high, the thickness can affect the free energy barrier more, and the membrane layer with the highest thickness has a higher free energy barrier for splitting. In the splitting completion stage, DPPS, which has the relatively thinnest membrane thickness and a larger single phospholipid area (APL), exerts an effect on the opposite membrane tube under high-frequency contact between head groups, making it easier for the membrane layer to open a splitting gap under the action of the probe and accelerate splitting.

[0083] To summarize the above rules, the membrane layer is easier to split when it is composed of short saturated tail chain phospholipids; the membrane layer is easier to split when it is composed of phospholipids containing appropriate unsaturation; the membrane layer is easier to split when it is composed of phospholipids that have weak mutual attraction and do not tend to aggregate.

[0084] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A molecular simulation method for studying the effect of phospholipid structure on biomembrane fission, characterized in that: The method comprises the following steps: (1) Constructing membrane tube splitting models with different structural phospholipid components: Phospholipids with different structures were selected according to the head group, tail chain length, and hydrocarbon chain unsaturation. Small module membrane systems with different phospholipid combinations were first constructed using the CHARMM-GUI website. Then, the small module membrane systems were expanded to the required size using GROMACS. Then, several groups of different membrane tube models were constructed using the BUMPY script tool. Finally, coarse-grained water droplets were used to simulate probes and probe rings were constructed around the membrane tube model to obtain a membrane tube splitting model. (2) Simulating splitting and shrinking behavior: GROMACS was used to simulate the splitting and contraction behavior: first, a constant force was applied to each probe toward the center point of the membrane tube at the horizontal plane where the probe was located to ensure that the probe could contract evenly until the lipids in the inner layer contacted each other. This process was called the mechanical contraction stage. Then, the probe in the last frame of the mechanical contraction stage was continuously pulled until the membrane tube split. This process was called the splitting completion process. (3) Mechanism analysis: Perform umbrella sampling on the two processes of step (2), and analyze the results of umbrella sampling using the gmx wham module in GROMACS; Use VMD visualization software to capture the dynamic process and details of membrane splitting, and use VMD's own scripts to analyze the overall evolution rate and change behavior during membrane splitting. The membrane layer of each group of membrane tubes includes common components and variable components. The common components are phospholipids shared by each group of membrane tubes, and the variable components are phospholipids that distinguish each group of membrane tubes from other membrane tubes. The variable components are divided into head group group, chain length group and unsaturation group according to the selected phospholipid structure. The head group selects DPPE, DPPG, and DPPS, the chain length group selects DTPA, DLPA, and DPPA, and the unsaturation group selects DOPA, DAPA, and DRPA. The addition ratio of the common components to the variable components is in the range of 1:1-1.

5. The membrane tube model was established using the BUMPY script tool, the Martini3 coarse-grained force field was selected, the temperature was set to 310 K to simulate the temperature environment in the body, and the membrane tube was periodically docked up and down.

2. The molecular simulation method for studying the effect of phospholipid structure on biomembrane fission according to claim 1, characterized in that: The initial membrane tube model needs to undergo energy minimization, pre-equilibrium simulation and equilibrium simulation to obtain a stable membrane tube model; after obtaining the stable membrane tube model, insert a probe, the probe is facing the center of the membrane tube, a 5nm water layer and ions are added above and below the membrane tube, and then energy minimization, pre-equilibrium simulation and equilibrium simulation are performed to obtain a stable membrane tube splitting model before shrinkage and splitting.

3. The molecular simulation method for studying the effect of phospholipid structure on biomembrane fission according to claim 2, characterized in that: Energy minimization was performed using two 5000-step steepest descents; The pre-equilibrium simulation was performed 5 times under isothermal and isobaric ensemble, each time for 1 ns, with the time step gradually increasing and the position restriction on the head group reduced in each pre-equilibrium; The equilibrium simulation before adding the probe was a 200 ns equilibrium simulation in the NPT ensemble, using the Parrinello-rahman pressure control method and a V-rescale thermal bath; The equilibrium simulation after adding the probe was a 1 ns equilibrium simulation under the NPT ensemble, using the Parrinello-rahman pressure control method and a V-rescale thermal bath.

4. The molecular simulation method for studying the effect of phospholipid structure on biomembrane fission according to claim 1, characterized in that: A probe ring consists of 10-16 evenly distributed probes. Each probe consists of 64 water beads arranged in a 4×4×4 cube. The beads are connected by 144 direct lattice bonds, 216 planar diagonal bonds and 108 diagonal bonds. The simple harmonic force constant of each bond is 1250 kJ·nm -2 ·mol -1 .

5. The molecular simulation method for studying the effect of phospholipid structure on biomembrane fission according to claim 1, characterized in that: The mechanical contraction process is a uniform stretching dynamics simulation of each probe, and the probe movement speed is 0.001nm×ps -1 ; The splitting process is to relax the end state of the mechanical contraction process, that is, firstly -1 The probe is pulled at a speed to advance 0.1 nm, and then a force is used to fix the probe to relax and deform the membrane tube for 50 ns. This process is repeated until the membrane tube splits.

6. The molecular simulation method for studying the effect of phospholipid structure on biomembrane fission according to claim 5, characterized in that: The conditions for the relaxation after completing a traction in the splitting process are as follows: a force potential with a velocity of 0 is applied to each probe under the PR pressure bath.

7. The molecular simulation method for studying the effect of phospholipid structure on biomembrane fission according to claim 1, characterized in that: The specific method of umbrella sampling in step (3) is: During the mechanical contraction stage, the probe displacement distance was used as the reaction coordinate, with a window of every 0.2 nm to ensure overlap between windows to obtain a continuous energy function; In the splitting completion stage, the relaxed configuration was taken as the initial window configuration, the reaction coordinates of adjacent window configurations were 0.1 nm apart, the sampling simulation parameters followed the relaxation simulation parameters and restrictions, and the sampling time was 20 ns.

8. An application of the molecular simulation method for studying the effect of phospholipid structure on biomembrane fission as claimed in any one of claims 1 to 7 in studying the effect of phospholipid structure on biomembrane fission and its microscopic mechanism from a microscopic perspective.

Images