-
Notifications
You must be signed in to change notification settings - Fork 24
Tutorial: PEGylated lipid bilayers
With polyply it becomes easy to combine your polymer with any existing bio-molecules. One such use case is the creation of PEGylated lipid bilayers. PEGylated lipids are lipids to which PEG is covalently attached. Depending on the grafting density (i.e. concentration of those lipids in the bilayer) the chain dimensions fall into one of two regimes. In the low concentration regime, known as “mushroom regime”, the chains are well-separated, interact minimally, and therefore move freely within a space approximating a half-sphere.[1] In contrast, in the high grafting density regime (“brush regime”), the polymer chains are close in space and repel each other. This repulsion leads to more extended chain dimensions than in the mushroom regime. More information about PEGylated lipid bilayers can be found here [2,3]. In this tutorial we learn how to create PEGylated bilayers in the two regimes using the Martini force-field.
First we need to generate an itp file for the PEGylated lipid we want to study. To do so we have to download the lipid itp file from the Martini database. Note not all lipid types are currently supported. Once we have the lipid itp file we simply invoke the itp file generation tool (polyply gen_itp) as follows:
polyply gen_params -f <lipid>.itp -lib martini2 -o pegylated_lipid.itp -name PEL -seq <lipid-name>:1 PEO:<number of monomers>
For this command we have to provide the lipid itp file using the -f flag (i.e. -f <lipid>.itp) Next we have to tell polyply we want to use the library of Martini polymers using -lib martini2_polymers. Note in place of "martini2_polymers" other martini force-fields can also be used. Besides the name of the output file provided with -o and the name of the molecule set by -name, we also need to define the macro molecule sequence. In this case we want one lipid molecule (i.e. <lipid-name>:1) to be connected to several monomers of PEO (i.e. PEO:<number of monomers>). The lipid-name is the molecule name specified in the lipid itp file.
Let us generate a specific example namely PEGylated POPC with 50 residues of PEG terminated with an OH group. We will use this lipid as an example in the rest of the tutorial. To generate the itp file first download POPC then run the following command:
polyply gen_params -f POPC.itp -lib martini2 -o PEL.itp -name PEL -seq POPC:1 PEO:50 OHter:1
Now that we have the itp-file, we will learn how to generate the starting coordinates. This is done in two steps: (1) First use insane or TS2CG to generate a lipid bilayer of your choice. Don't add water or ions yet! (2) Subsequently use the polyply coordinate generation tool to add the polymer. For the sake of this Tutorial we generate a lipid bilayer of 100 POPC lipids in each leaflet. If you are using insane this can be done by running the following command:
python2 insane.py -l POPC:100 -o bilayer_only.gro -dz 20.0 -x 8.0 -y 8.0
Now that we have the bilayer let's begin by generating coordinates for the mushroom state.
The first thing we need is a topology file. Let's call it system.top. This file needs to describe the complete system that we want to generate. Note that you'll have to download the martini parameter file as well. The topology file should then look like this:
#include "martini_v2.0_PEO_PS_CNP.itp"
#include "POPC.itp"
#include "PEL.itp"
[ system ]
; name
INSANE! Membrane UpperLeaflet>POPC=1.0 LowerLeaflet>POPC=1.0 in water
[ molecules ]
; name number
PEL 5
POPC 190
PEL 5
The mushroom state is characterized by an unrestricted chain tethered to the lipid bilayer. So there is no specific directional restriction to the growing of the polymer chain. Thus we can use the following command to "grow" the polymer on top of the lipid bilayer.
polyply gen_coords -p system.top -o struct.gro -res PEO OHter -c bilayer_only.gro -name test -box 8.0 8.0 20.0 -split POPC:HEAD-NH3:TAIL-PO4,GL1,GL2,C1A,D2A,C3A,C4A,C1B,C2B,C3B,C4B
To generate the structure we always need to specify the starting lipid bilayer file (-c start.pdb), the residues that need to be built (-res PEO OHter) and the box (-box 8.0 8.0 20). In addition we choose to split the residue POPC into two separate residues describing the lipid head-group and the lipid tail region. Run polyply gen_coords -h to get more information on the splitting syntax. Splitting the residues is needed, because polyply generates a super-coarse grained 1 bead per residue model. However, PEO is attached to the head group and we want to start "growing" the chain from there. You will have to adopt the names depending on the lipid types which you use. However, the head group always contains an NH3 and PO4 bead. Finally you can add water to the system with your most favorite tool for solvation or skip ahead to use polyply.
In the brush regime the grafting density is higher the chains tend to be more elongated and more tightly packed. When we generate the initial structure we need to take those factors into account. This can be done very easily by creating a build file, which can be used to specify more complex building options. First we need to increase the grafting density. To do so change the number of PEL molecules in the topology file to 25 for each leaflet. Next create a build file, which looks as follows:
[ molecule ]
; molname
PEL 0 25
[ rw_restriction ]
; resname start stop normal angle
PEO 3 52 0 0 1 90.0
[ molecule ]
; molname
PEL 175 200
[ rw_restriction ]
; resname start stop normal angle
PEO 2 51 0 0 -1 90.0
In this build file we can specify building options for specific molecules in the system. Which molecules we want to target is specified by the [ molecule ] directive that is followed by the a line of the following format: <molecule_name> <starting molecule index> <ending molecule index>. Note molecule indices are counted as total not per species. For this example, we want to specify options for the first 25 PEGylated lipids and the last 25. In this specific case the first 25 are in the upper leaflet and the last 25 are in the lower leaflet. The last 25 molecules have the indices 175 through 200 as there are in total 200 molecules. Next we write the [rw_restriction] directive. This directive can be used to specify how we want to restrict the random-walk used to grow the polymer chains. It has the following syntax: <resname> <starting resid> <last resid> <x> <y> <z> <angle in degree>. Each line specifies which residue of the molecule we want to target by residue name and the range of residue ids, followed by a plane normal vector and an angle. Using this syntax the random-walk will be restricted by the following rule: A step is only allowed, iff the angle between the step-vector and a vector specified in the [rw_restriction] directive is less than the angle specified in that same directive. In our case we want the chains in the upper leaflet to be stretched in the positive z-direction and in the lower leaflet in the negative z direction. Thus we specify the positive and negative z axis as our reference vectors. The angle is set to 90 degrees, which effectively means a step can never go back in z-direction but is free to move in x and y. Decreasing the angle will give even more straight chains. Using this build-file the PEGylated bilayer brush is simply generated by running:
polyply gen_coords -p system.top -o struct.gro -res PEO OHter -c bilayer_only.gro -name test -box 8.0 8.0 20.0 -split POPC:HEAD-NH3:TAIL-PO4,GL1,GL2,C1A,D2A,C3A,C4A,C1B,C2B,C3B,C4B -b build_file.bld
Running this command will yield the desired polymer brush. Again after solvating the system and running an energy minimization the procedure is complete.
In principle polyply can be used to add water as well as ions to structures as well. The only downside is that you will have to more or less accurately guess the number of water molecules and ions. However, it also has the advantage that you can exclude regions in space where water should not be added. For example, gmx_solvate might add water in the lipid bilayer tail region, which can perturb the bilayer. Let us try water placement for the brush we just created. First we need to add the water to the topology file. Add the following line to the topology file from the last step: W 8000. Next we add the following entry to the build file:
[ molecule ]
; molname
W 201 8202
[ rectangle ]
; resname start stop inside-out x y z a b c
W 1 2 out 4.0 4.0 10. 4.1 4.1 1.5
Now we invoke the rectangle directive, which looks very similar to the the previous directive. However, in this case the syntax is: <resname> <starting resid> <last resid> <in or out> <x> <y> <z> <a> <b> <c>, where we specify a point and three distances along (a,b,c) from a central point to the faces of a rectangle. Together with the keyword out it means all water molecules need to be outside a rectangle defined by this syntax. Having edited all files run the following command:
polyply gen_coords -p system.top -o struct.gro -res PEO OHter -c start.pdb -name test -box 8.0 8.0 20.0 -split POPC:HEAD-NH3:TAIL-PO4,GL1,GL2,C1A,D2A,C3A,C4A,C1B,C2B,C3B,C4B -b build_file.bld
After energy minimization, your system is ready to go.
- implemented lipids DPPE, DOPC, POPC, DPPC, DOPE
- if you'd like another PEGylated lipid please raise an issue and we'll help you asap
- currently only flat bilayers are supported
- always run an energy minimization before running an NpT relaxation
- to see other libraries implemented run
polyply list-libs - this tutorial was affected by issue #93; please use the master branch when running the tutorial
[1] Grunewald, F., Rossi, G., de Vries, A.H., Marrink, S.J. and Monticelli, L., 2018. Transferable MARTINI model of poly (ethylene oxide). The Journal of Physical Chemistry B, 122(29), pp.7436-7449.
[2] Lee, H. and Larson, R.G., 2016. Adsorption of plasma proteins onto PEGylated lipid bilayers: the effect of PEG size and grafting density. Biomacromolecules, 17(5), pp.1757-1765.
[3] Lee, H. and Pastor, R.W., 2011. Coarse-grained model for PEGylated lipids: effect of PEGylation on the size and shape of self-assembled structures. The Journal of Physical Chemistry B, 115(24), pp.7830-7837.