John M. Leveritt III, Almudena Pino-Angeles, Themis Lazaridis*.

Transcription

1 The structure of a melittin-stabilized pore John M. Leveritt III, Almudena Pino-Angeles, Themis Lazaridis*. Department of Chemistry, The City College of New York, 160 Convent Ave, New York, NY, 10031, USA Supplementary Material System Setup The initial melittin structure was taken from the Protein Data bank 1, entry 2MLT. A close-packed tetramer was then generated with the helices pointing their hydrophobic groups away from each other (Fig. S1). Preparation of the tetramer was done with the CHARMM program 2. The remainder of the system was built with CHARMM-GUI 3, in particular the membrane builder 4 6. The lipids were composed of DMPC for the zwitterionic system and 3: 1 DMPC:DMPG for the 25% anionic system. The P/L ratio was 1:18 in both cases. The replacement method was used to insert the tetramer into the lipid bilayer. The TIP3P model was used to model the water, and chloride ions were added to neutralize the system. The total number of water molecules was 4096 and the number of chloride ions was 20. The total number of atoms for these two systems was approximately 20,000. Figure S1 Initial structure of the melittin tetramer. Lipids are not shown. Simulation Details Equilibration was done in a standard manner where first the water molecules, ions, phosphates of the lipid headgroups, and the protein backbone were harmonically constrained (k = 1 kcal/mol/å 2 ). Gradually the constraints were released; first the

2 phosphates, followed by the water/ions, and finally the peptides. Brief NPT production simulations were run locally using the NAMD software package 7. Periodic boundary conditions and the Particle Mesh Ewald (PME) method were used for these local runs. SHAKE constraints for the bonds involving hydrogen were also employed, and the step size for integration was 2 fs. The main production runs were run on the Anton supercomputing machine 8. The initial DMS (Desmond structure) file was generated from the final structure of the locally run production runs. To generate the DMS it is required to extract the extended system coordinate file, velocities, coordinates, and psf. The CHARMM36 force field 9,10 was used for the entire system during these long production runs and the TIP3P model was used for water. Most of the parameters used in the simulations were the suggested parameters from the various commands within Anton (guess_chem, refinesigma, and subboxer). Details on how the parameters were generated can be found on the Anton wiki. The box size was 56 x 56 x 60 Å, a 32 x 32 x 32 PME grid was used, and non-bonded interactions were cut off at 11 Å. The pressure was set to a constant value of 1 bar, and the temperature was set to 300 K. The step size for the simulation was 2 fs and coordinates were written every 1.08 ns. SHAKE was also implemented in the Anton runs. For higher efficiency, in the production runs we chose to use the Multigrator integrator 11, as opposed to a pure NPT ensemble. For most steps of the simulation, Multigrator runs NVE dynamics and applies a thermostat or barostat only periodically. In our case we used a Nose-Hoover thermostat, updated every 24 steps, and an MTK barostat, updated every 240 steps. To compute quantities such as pore radius, tilt angle, shape, and lipid headgroup distribution, trajectories were analyzed locally using the CHARMM program. All trajectories were converted from the Anton format to dcd files using VMD 12. Pore radius was determined by counting the number of water molecules in each z coordinate slice using the COOR SEARCH command and computing the radius of an effective cylinder that contains them. The tilt angles were determined with the COOR HELIX command with the helix axis defined by residues 5 to 22. Helicities were calculated with the program CPPTRAJ 11 from the Ambertools package using DSSP rules. Results: Melittin tetramer in DMPC The tilt angles as a function of time are shown in Figure S2. As one can see there is significant tilting of several of the melittin monomers and there is enough tilting present for the monomers to enter a surface state (tilt angle of 90 degrees). The average tilt angles over the entire simulation are: 20.7±10.8, 27.6±13.7, 42.4±23.3, and 31.6±14.1. Average tilt angles over the last μs are: 25.4 ± 11.3, 18.0 ± 9.0, 48.0 ± 11.6, and 29.6 ± The kink angle ranges from to o, compared to o in the crystal structure. Thus, in the membrane environment the peptide is less kinked than in the crystal structure, consistent with solid state NMR studies 13.

3 Figure S2 Melittin monomer tilting as a function of time. The four monomers present in the simulation are indicated with a unique color. Previous work suggested the importance of Lys7 as a determinant of toroidal pore formation 14. Interaction energies with lipids showed that Lys7 contributed about kcal/mol in comparison to nearly kcal/mol contributed by the C-terminal charges 15. Interaction energies calculated from this longer simulation show nearly the same contribution of Lys7 ( ± kcal/mol) but a significantly less stabilizing effect from the C-terminal charges ( ± kcal/mol). These results support the important role of Lys7, although long mutant simulations would have to be performed to confirm this hypothesis. Results: Melittin trimer in DMPC Due to the interesting behavior of one of the melittin monomers (undergoing consistent T to S state transitions), we wondered whether a trimer can stabilize a toroidal pore. To test this hypothesis we took one of the early frames with three T- state peptides and one S-state peptide and removed the S-state monomer. We then re-equilibrated and ran the trimer for 2 μs on Anton. We observed an initial pore that was stable for one microsecond, and then a degradation of the pore. This is shown in Figure S3. Results: Melittin in 3:1 DMPC: DMPG In contrast to the melittin tetramer in DMPC, there are three aspects in which the behavior of the system changes with the incorporation of DMPG lipids. The first is the effect on pore radius, the second the dynamics of the monomers (tilt angle), and lastly the lack of movement of head groups into the membrane. The pore radius as a function of time is shown in Figure S4. It is clear that the pore closes periodically throughout the simulation and eventually closes entirely for the last microsecond. Additionally the tilting dynamics of the monomers changed with the incorporation of anionic lipids. As shown in Figure S5, the monomers remain

4 transmembrane throughout the simulation. No transition from transmembrane to a surface state is observed. Figure S3 Representative pictures of the melittin trimer are shown. The top left corner indicates the initial state and the top right corner is the state after one microsecond. The final state at two microseconds is shown in the lower panel and one can see that the pore tends to close. Figure S4 Pore radius as a function of time in 3:1 DMPC/DMPG. A small initial pore begins to decay and periodically close. This is due to one of the monomers separating from the pore as described in the manuscript

5 Figure S5 Tilt angle of the four peptides as a function of time for the melittin tetramer in 3:1 DMPC/DMPG. There is significantly less tilting in this simulation than in the DMPC simulation The distribution of lipid headgroups within the membrane also changes when comparing the tetramer in DMPC vs. DMPC: DMPG (3:1). Initially as the monomers separate a few headgroups enter the membrane to stabilize a pore. Eventually as the pore decays the lipid headgroups return to their respective leaflets. Toward the end of the simulation there is at best one headgroup within the membrane but the majority reside at the membrane surfaces. References (1) Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.; Bhat, T. N.; Weissig, H.; Shindyalov, I. N.; Bourne, P. E. Nucleic Acids Res. 2000, 28, 235. (2) Brooks, B. R.; Brooks, C. L.; Mackerell, a D.; Nilsson, L.; Petrella, R. J.; Roux, B.; Won, Y.; Archontis, G.; Bartels, C.; Boresch, S.; Caflisch, A.; Caves, L.; Cui, Q.; Dinner, a R.; Feig, M.; Fischer, S.; Gao, J.; Hodoscek, M.; Im, W.; Kuczera, K.; Lazaridis, T.; Ma, J.; Ovchinnikov, V.; Paci, E.; Pastor, R. W.; Post, C. B.; Pu, J. Z.; Schaefer, M.; Tidor, B.; Venable, R. M.; Woodcock, H. L.; Wu, X.; Yang, W.; York, D. M.; Karplus, M. J. Comput. Chem. 2009, 30, (3) Jo, S.; Kim, T.; Iyer, V. G.; Im, W. J. Comput. Chem. 2008, 29, (4) Wu, E. L.; Cheng, X.; Jo, S.; Rui, H.; Song, K. C.; Dávila-Contreras, E. M.; Qi, Y.; Lee, J.; Monje-Galvan, V.; Venable, R. M.; Klauda, J. B.; Im, W. J. Comput. Chem. 2014, 35, (5) Jo, S.; Lim, J. B.; Klauda, J. B.; Im, W. Biophys. J. 2009, 97, 50. (6) Jo, S.; Kim, T.; Im, W. PLoS One 2007, 2, e880. (7) Phillips, J. C.; Braun, R.; Wang, W.; Gumbart, J.; Tajkhorshid, E.; Villa, E.; Chipot, C.; Skeel, R. D.; Kalé, L.; Schulten, K. J. Comput. Chem. 2005, 26, (8) Shaw, D. E.; Dror, R. O.; Salmon, J. K.; Grossman, J. P.; Mackenzie, K. M.; Bank, J. A.; Young, C.; Deneroff, M. M.; Batson, B.; Bowers, K. J.; Chow, E.; Eastwood, M. P.; Ierardi, D. J.; Klepeis, J. L.; Kuskin, J. S.; Larson, R. H.; Lindorff-larsen, K.; Maragakis, P.; Moraes, M. A.; Piana, S.; Shan, Y.; Towles, B.; York, N. Proc. ACM/IEEE Conf. Supercomput. (SC09), Portl. Oregon 2009.

6 (9) Best, R. B.; Zhu, X.; Shim, J.; Lopes, P. E. M.; Mittal, J.; Feig, M.; Mackerell, A. D. J. Chem. Theory Comput 2012, 8, (10) Klauda, J. B.; Venable, R. M.; Freites, J. A.; O Connor, J. W.; Tobias, D. J.; Mondragon-Ramirez, C.; Vorobyov, I.; MacKerell, A. D.; Pastor, R. W. J. Phys. Chem. B 2010, 114, (11) Lippert, R. a; Predescu, C.; Ierardi, D. J.; Mackenzie, K. M.; Eastwood, M. P.; Dror, R. O.; Shaw, D. E. J. Chem. Phys. 2013, 139, (12) Humphrey, W.; Dalke, A.; Schulten, K. J. Molec. Graph. 1996, 7855, 33. (13) Naito, A.; Nagao, T.; Norisada, K.; Mizuno, T.; Tuzi, S.; Saitô, H. Biophys. J. 2000, 78, (14) Mihajlovic, M.; Lazaridis, T. Biochim. Biophys. Acta 2012, 1818, (15) Mihajlovic, M.; Lazaridis, T. Biochim. Biophys. Acta - Biomembr. 2010, 1798, 1485.