MEMBPLUGIN: studying membrane complexity in VMD

From W3C Wiki
< HTML‎ | Training
Jump to: navigation, search

MEMBPLUGIN [1] is a versatile tool for the Visual Molecular Dynamics (VMD) package [2] to analyze the results of complex membrane simulations. The plugin can be used to characterize a wide range of biophysical properties of biomembranes such as membrane thickness, fluidity or condensation. In the present case study, we illustrate some of the features of the plugin by analyzing the effect of cholesterol enrichment on a series of lipid bilayers simulations.


Cholesterol, a vital component of cells is know to regulate the biophysical properties of biological membranes. The presence of this sterol in phospholipid membranes has proven to increase membrane condensation [3, 4], ultimately impacting on the overall structure of the bilayer. Thus, cholesterol content can alter key biophysical properties of biological membranes such as thick- ness, fluidity or area per lipid. Relevant membrane microdomains such as the so-called lipid rafts [5] display a high amount of cholesterol when compared to other regions of the membrane. The study of lipid rafts by experimental means depends upon the development of complex technology due to the fluctu- ating and heterogeneous nature of these nanoscale assemblies [6]. On the other hand, molecular models can be used in an attempt to offer a molecular per- spective (i.e. more detailed) of cholesterol-enriched membranes such as lipid rafts. Despite of the known limitations of current computational techniques, some studies using atomistic [7, 8] and coarse-grained [9] molecular dynamics have already shed some light on the molecular aspects of cholesterol-enriched domains. For this sample case study, we focused on the effect of cholesterol on certain biophysical properties of standard model lipid bilayers.


Play all:

POPC:CHL1 - 1:0

POPC:CHL1 - 1:1

POPC:CHL1 - 2:1

POPC:CHL1 - 5:1


We simulated a set of 1-palmitoyl-2-oleyl-sn-glycero-3-phosphocholine (POPC) bilayers containing different amounts of cholesterol. Specifically, we built 4 different POPC:CHO systems, namely 1:0, 5:1, 2:1 and 1:1 (see Table 1). All membranes were built using the CHARMM-GUI membrane builder [10] ( http: // ). Each output was re-hydrated us- ing approximately 30 water molecules (TIP3 model) per lipid and neutral- ized with 150 mM NaCl. In addition, those lipid tails incorrectly trapped into the cholesterol ring were manually corrected. Subsequently, membranes were run in the NPT ensemble at 1.01325 bar and 310 K for 30 ns, us- ing the ACEMD simulation package [11]. The CHARMM36 [12] force field was used in all simulations. We discarded the first 10 ns of each trajec- tory and used MEMBPLUGIN [1] to analyze the area per lipid, lipid S cd order parameters, leaflet interdigitation and membrane thickness on the last 20 ns of each system. Usage instructions of each subtool can be found at .


To illustrate the condensing effect of cholesterol on a lipid bilayer, we first measured the average area per lipid of each lipid bilayer using the area per lipid tool of MEMBPLUGIN. This tool calculates both the total area per lipid of the membrane and the area per lipid for each of the lipid species present at the membrane, i.e. POPC or CHOL in our particular case.

Apart from the typical equilibration ramp showed within the first 0-5 ns, our simulations do not show major fluctuations in the calculated area per lipid during the last 20 ns of each trajectory. In any case, it is worth to note that due to the short time scale used for this case study, we can not assume the 3area per lipid to be in equilibrium. In agreement with experimental [14] and computational [12] estimates, the average area per lipid of POPC, as calculated for system 1:0 (i.e. pure POPC membrane), is 69 . 26 ± 0 . 75 Å 2 . The addition of cholesterol drops the total area per lipid of the system, in consonance with the known condensing effect of this molecule [3]. Increasing the level of cholesterol from 0% (i.e. system 1:0) to 17% (i.e. system 5:1), 33% (i.e. system 2:1, the approximate cholesterol level found in lipid rafts) and 50% (i.e. system 1:1) decrease the average area per lipid of POPC down to 61 . 69 ± 0 . 89 Å 2 , 60 . 11 ± 0 . 81 Å 2 and 57 . 56 ± 0 . 93 Å 2 , respectively . While the condensing effect of cholesterol is most dramatic within 0-17% cholesterol, this effect is less pronounced at higher molar cholesterol concentrations (i.e. 17%-50%).

Scd order parameters

With a view to show the effect of cholesterol on membrane fluidity, we also inspected the chain structure of POPC in our simulations by using the S cd tool of MEMBPLUGIN. This tool can compute the S cd parameters of each 4phospholipids present in the membrane under analysis. As described elsewhere [15], highly ordered membranes display high S cd values and vice versa. Our results display the highest order near the polar head of the phospholipid (i.e. towards C2) whereas the end of phospholipid tails (i.e. towards C18) are highly disorder, that is, a the typical S cd plot of a lipid bilayer. In addition, the results we obtained for the pure POPC system (i.e. system 1:0) are in agreement with previous experimental [16, 17] and computational [12] studies of this phospholipid.

In our simulations, increasing amounts of cholesterol yield higher S cd values of POPC tails, that is, more ordered membranes. This trend is clearly visible for both the unsaturated and the saturated chain of POPC. Therefore, as expected, affecting membrane fluidity is one of the underlying mechanisms behind cholesterol condensing effect on POPC membranes.

Membrane thickness

However, are highly-condensed and highly-ordered membranes thicker that more extended and fluid membranes? We address this question using the mem- brane thickness tool of MEMBPLUGIN to compute the so-called phosphate-to- 5phosphate distance of our membrane set. This tool can compute the average thickness of the lipid bilayer between phosphate atoms (or any user-defined atom) and generate local deformations maps based on this analyis. The mem- brane thickness we obtained for the pure POPC membrane (i.e. system 1:0) goes along the experimental value of this type of lipid bilayers [18]. Likewise, as previously described in experimental [19] and computational [4,20] studies, we observe that membrane thickness increases with higher molar concentrations of membrane cholesterol. The results from our simula- tions confirm that membrane thickness increases by approximately 10%, 18% and 26% in systems containing 17% (i.e. system 5:1), 33% (i.e. system 2:1) and 50% (i.e. system 1:1) of cholesterol, respectively.

Specifically, we found an average membrane thickness of 37 . 73 ± 0 . 54 Å, 41 . 76 ± 0 . 48 Å, 43 . 31 ± 0 . 42 Å and 44 . 72 ± 0 . 64 Å, during the last 20 ns simulation of system 1:0, 5:1, 2:1 and 1:1, respectively.

Leaflet interdigitation

The concept on leaflet interdigitation remains controversial [21], partly due to the lack of experimental tools able to approach this measurement. However, this parameter can give valuable information on the coupling extent between membrane leaflets. Thus, we used the lipid interdigitation tool of MEMB- PLUGIN to study the impact of cholesterol level on the coupling between membrane leaflets during the simulation. This tool offers three estimates of this parameter, namely the fraction mass overlap between the two leaflets, I ρ , the width of such region, w ρ , and the fraction of contacts between atoms of different leaflets, I C , (see Home/LipidInterdigitation/ for more details on this tool). Hereby, we com- puted the value of I ρ , w ρ and I C for each of the membranes comprising our simulation set.

As for previous calculations, the results show a correlation between mem- brane cholesterol levels and leaflet interdigitation. In this case, higher choles- terol levels generally seems to decrease the extent of interdigitation between 7membrane leaflets . Thus, it shows how molar cholesterol concentrations of 17% (i.e. system 5:1), 33% (i.e. system 2:1) and 50% (i.e. system 1:1) gradually decrease both the mass overlap between the two leaflets, I ρ and the width of such region, w ρ . Interestingly, as the fraction of contacts between leaflets, I C , does not seem to decrease at the same pace at high cholesterol concentrations (i.e. system 2:1 and system 1:1) when compared to lower ones. In fact, the most pronounced effect of cholesterol addition on this parameter occurs, to a high extent, at the lowest cholesterol level (i.e. 17%, system 5:1) .


We used MEMBPLUGIN to analyze the effect of cholesterol on the structure of POPC bilayers. On the one hand, MEMBPLUGIN is able to yield repro- ducible results in terms of area per lipid, S cd order parameters and membrane thickness and to highlight the already-known condensing effect of cholesterol on phospholipid bilayers. At the same time, the increased amount of leaflet interdigitation showed a probable decrease in the extent of coupling between membrane leaflets, a fact that goes along the increase of membrane thick- ness. Although the aim of this case study was to highlight the versatility of MEMBPLUGIN using a realistic case, a similar rationale could be followed to analyze the impact of membrane composition on the biophysical properties of more complex and heterogeneous bilayers such as membrane microdomains [7], bacteria model membranes or specific subcellular compartments [22].

See also[edit]


  1. Ramon Guixà-González, Ismael Rodriguez-Espigares, Juan Manuel Ramı́rez-Anguita, Hector Martinez-Seara, Toni Giorgino, Pau Carrió, and Jana Selent. MEMBPLUGIN: Studying membrane complexity in VMD. Bioinformatics, 2013 (submitted).
  2. W. Humphrey, A. Dalke, and K. Schulten. VMD: visual molecular dynamics. Journal of molecular graphics, 14(1):33–38, 1996.
  3. Wei-Chin Hung, Ming-Tao Lee, Fang-Yu Chen, and Huey W Huang. The condensing effect of cholesterol in lipid bilayers. Biophysical journal, 92(11):3960–7, June 2007.
  4. Frédérick de Meyer and Berend Smit. Effect of cholesterol on the structure of a phos- pholipid bilayer. Proceedings of the National Academy of Sciences of the United States of America, 106(10):3654–8, March 2009.
  5. Kai Simons and Mathias J. Gerl. Revitalizing membrane rafts: new tools and insights. Nature Reviews Molecular Cell Biology, 11(10):688–699, October 2010.
  6. Daniel Lingwood and Kai Simons. Lipid rafts as a membrane-organizing principle. Sci- ence (New York, N.Y.), 327(5961):46–50, January 2010.
  7. Perttu Niemelä, Samuli Ollila, Marja T Hyvönen, Mikko Karttunen, Ilpo Vattulainen, and Marja T Hyvo. Assessing the nature of lipid raft membranes. PLoS Comput. Biol, 3(2):e34, 2007.
  8. S.A. Pandit, S Vasudevan, S W Chiu, R Jay Mashl, Eric Jakobsson, and H L Scott. Sphingomyelin-cholesterol domains in phospholipid membranes: atomistic simulation. Biophysical journal, 87(2):1092–100, August 2004.
  9. H Jelger Risselada and Siewert J Marrink. The molecular face of lipid rafts in model membranes. Proceedings of the National Academy of Sciences of the United States of America, 105(45):17367–72, November 2008.
  10. Sunhwan Jo, JB Joseph B Lim, JB Jeffery B Klauda, and Wonpil Im. CHARMM-GUI membrane builder for mixed bilayers and its application to yeast membranes. Biophysical journal, 96(1):50–8, July 2009.
  11. MJ J Harvey, G Giupponi, and GD De Fabritiis. ACEMD: accelerating biomolecular dynamics in the microsecond time scale. Journal of Chemical Theory and Computation, 5(6):1–9, 2009.
  12. Jeffery B Klauda, Richard M Venable, J Alfredo Freites, Joseph W O’Connor, Dou- glas J Tobias, Carlos Mondragon-Ramirez, Igor Vorobyov, Alexander D MacKerell, and Richard W Pastor. Update of the CHARMM all-atom additive force field for lipids: val- idation on six lipid types. The journal of physical chemistry. B, 114(23):7830–43, June 2010.
  13. Hadley Wickham. ggplot2: elegant graphics for data analysis. Springer New York, 2009.
  14. Norbert Kucerka, Stephanie Tristram-Nagle, and John F Nagle. Structure of fully hy- drated fluid phase lipid bilayers with monounsaturated chains. The Journal of membrane biology, 208(3):193–202, December 2005.
  15. Louic S Vermeer, Bert L de Groot, Valérie Réat, Alain Milon, and Jerzy Czaplicki. Acyl chain order parameter profiles in phospholipid bilayers: computation from molecular dynamics simulations and comparison with 2H NMR experiments. European biophysics journal : EBJ, 36(8):919–31, November 2007.
  16. Saame Raza Shaikh, Michael R Brzustowicz, Noah Gustafson, William Stillwell, and Stephen R Wassall. Monounsaturated PE does not phase-separate from the lipid raft molecules sphingomyelin and cholesterol: role for polyunsaturation? Biochemistry, 41(34):10593–602, August 2002.
  17. B Perly, I C Smith, and H C Jarrell. Acyl chain dynamics of phosphatidylethanolamines containing oleic acid and dihydrosterculic acid: 2H NMR relaxation studies. Biochem- istry, 24(17):4659–65, August 1985.
  18. Fluid phase lipid areas and bilayer thicknesses of commonly used phosphatidylcholines as a function of temperature. Biochimica et biophysica acta, 1808(11):2761–71, November 2011.
  19. Norbert Kucerka, Jason D Perlmutter, Jianjun Pan, Stephanie Tristram-Nagle, John Katsaras, and Jonathan N Sachs. The effect of cholesterol on short- and long-chain monounsaturated lipid bilayers as determined by molecular dynamics simulations and X-ray scattering. Biophysical journal, 95(6):2792–805, September 2008.
  20. Christofer Hofsäss, Erik Lindahl, and Olle Edholm. Molecular dynamics simulations of phospholipid bilayers with cholesterol. Biophysical journal, 84(4):2192–206, April 2003.
  21. Interleaflet coupling mechanisms in bilayers of lipids and cholesterol. Biophysical journal, 94(5):L32–4, March 2008.
  22. Björn Sommer, Tim Dingersen, Christian Gamroth, Sebastian E. Schneider, Sebastian Rubert, Jens Krüger, and Karl-Josef Dietz. CELLmicrocosmos 2.2 MembraneEditor: a modular interactive shape-based software approach to solve heterogeneous membrane packing problems. Journal of Chemical Information and Modeling, 51(5):1165–1182, May 2011.