1University of Rochester Medical School, Rochester, NY, USA2University of Arkansas, Fayetteville, AR, USA

To d D . R o m o 1 , J o s h u a N . H o r n 1 , D e n i s e V . G r e a t h o u s e 2 , A l a n G r o s s f i e l d 1To d D . R o m o 1 , J o s h u a N . H o r n 1 , D e n i s e V . G r e a t h o u s e 2 , A l a n G r o s s f i e l d 1

1University of Rochester Medical School, Rochester, NY, USA2University of Arkansas, Fayetteville, AR, USA

CHARACTERIZATION OF MEMBRANE INTERACTIONS WITH LACTOFERRICIN PEPTIDES BY ALL-ATOM AND COARSE-GRAINED MOLECULAR DYNAMICS SIMULATIONS, SOLID-STATE NMR,

AND FLUORESCENCE SPECTROSCOPY

CHARACTERIZATION OF MEMBRANE INTERACTIONS WITH LACTOFERRICIN PEPTIDES BY ALL-ATOM AND COARSE-GRAINED MOLECULAR DYNAMICS SIMULATIONS, SOLID-STATE NMR,

AND FLUORESCENCE SPECTROSCOPY

LfB6 (RRWQWR-NH2) is a small cationic antimicrobial peptide with broad spectrum effectiveness that is derived from bovine lactoferrin. The mechanism for interaction between the antimicrobial peptide and the bacterial cell membrane is hypothesized to depend on lipid composition. Bacterial membranes generally contain a significant frac-tion of negatively charged lipids in contrast with zwitterionic mammalian membranes. Previously, we characterized the interactions of an acylated LfB6 (C6-LfB6) with a model bacterial membrane (3:1 POPE:POPG) and a model mammalian membrane (POPC). Here, we investigate the interactions of the non-acylated LfB6 peptide with the same model membranes, using over 17 µs of all-atom molecular dynamics as well as 53 µs of coarse-grained simulations, and we compare our results to solid-state 2H NMR and fluorescence spectroscopy. Molecular dynamics simulations reveal that the LfB6 peptide backbone does not penetrate as deeply in the model membranes as C6-LfB6 and that there is no preference in order of side-chain binding, unlike C6-LfB6. Further, molecular dynamics indicates the LfB6 tryptophans are more deeply buried in the membrane than C6-LfB6, yet fluorescence spectroscopy suggests they are more water-exposed. Coarse-grained molecular dynamics reveals that LfB6 comes off the membrane more easily than C6-LfB6, explaining the tryptophan membrane location and water exposure. The results also show subtle changes in the membranes’ struc-ture between the acylated and non-acylated peptides.

Membrane Type Length (ns)

Avg Length (ns)

Area / Lipid (Å2)

Avg Area / Lipid (Å2)

POPE:POPG C6-LfB6

3100

3055

63.6

63.6POPE:POPG C6-LfB6 3417 3055 63.6 63.6POPE:POPG C6-LfB6 2701 3055 63.6 63.6POPE:POPG C6-LfB6

3002

3055

63.6

63.6

POPE:POPG LfB6

3428

3209

63.5

63.5POPE:POPG LfB6 3406 3209 63.5 63.5POPE:POPG LfB6 3001 3209 63.4 63.5POPE:POPG LfB6

3002

3209

63.4

63.5

POPC C6-LfB6

3549

3368

67.8

67.8POPC C6-LfB6 3620 3368 67.8 67.8POPC C6-LfB6 3102 3368 67.8 67.8POPC C6-LfB6

3202

3368

67.8

67.8

POPC LfB6

4300

3679

67.7

67.6POPC LfB6 4400 3679 67.6 67.6POPC LfB6 3003 3679 67.6 67.6POPC LfB6

3012

3679

67.7

67.6

POPE POPG Water Lipid POPC Water

Acyl Chain Position

-0.02

-0.015

-0.01

-0.005

0

0.005

0.01

0.015

0.02

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

LfB6 POPC

SC

D (P

eptid

e - L

ipid

)Acyl Chain Position

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

LfB6 POPEPOPG

Acyl Chain Position

-0.02

-0.015

-0.01

-0.005

0

0.005

0.01

0.015

0.02

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

C6-LfB6

SC

D (P

eptid

e - L

ipid

)

Acyl Chain Position2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

C6-LfB6

Distance from Membrane Interface (Å)

Nor

mal

ized

Den

sity

LfB6 POPE:POPG C6-LfB6 POPE:POPGLfB6 POPC C6-LfB6 POPC

0

0.04

0.08

0.12

0.16

−15 −10 −5 0 5 10

Backbone

0

0.04

0.08

0.12

0.16

−15 −10 −5 0 5 10

Trp

0

0.04

0.08

0.12

0.16

−15 −10 −5 0 5 10

Arg

0

0.04

0.08

0.12

0.16

−15 −10 −5 0 5 10

Backbone

0

0.04

0.08

0.12

0.16

−15 −10 −5 0 5 10

Trp

0

0.04

0.08

0.12

0.16

−15 −10 −5 0 5 10

Arg

300 320 340 360 380 400 420 440

0.0

0.2

0.4

0.6

0.8

1.0

LfB6 POPC, (352 nm) LfB6 POPE:POPG, (339/342 nm) C6-LfB6 POPC, (345 nm) C6-LfB6 POPE:POPG, (339 nm)

Nor

mal

ized

Flu

ores

cenc

e In

tens

ity, a

.u

Emission, nm

Dis

tanc

e fr

om M

embr

ane

Cen

ter (

Å)

Time (µs)

All-AtomAll-Atom Coarse GrainedCoarse Grained

-0.020

-0.015

-0.010

-0.005

0

0.005

0.010

0.015

0.020

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

C6-LfB6

Acyl Chain Position

SC

D (P

eptid

e - L

ipid

)

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

C6-LfB6

Acyl Chain Position

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

POPEPOPG

LfB6

Acyl Chain Position

-0.020

-0.015

-0.010

-0.005

0

0.005

0.010

0.015

0.020

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

POPCLfB6

Acyl Chain Position

SC

D (P

eptid

e - L

ipid

)

2 6 10 11 12 13 14 15 16

LfB6 POPEPOPG

-0.020

-0.015

-0.010

-0.005

0

0.005

0.010

0.015

0.020

2 6 10 11 12 13 14 15 16

LfB6 POPC

2 6 10 11 12 13 14 15 16

C6-LfB6

-0.020

-0.015

-0.010

-0.005

0

0.005

0.010

0.015

0.020

2 6 10 11 12 13 14 15 16

C6-LfB6

SC

D (P

eptid

e - L

ipid

) S

CD (P

eptid

e - L

ipid

)

Frac

tiona

l Con

tact

s

C6-LfB6 LfB6 C6-LfB6 LfB6Methods• 5Å probe radius• Count atoms within the sphere• Fractional contribution by different components• Peptide heavy atoms probed for entire peptide,

all arginine atoms, all tryptophan atoms, and all C6 tail atoms (not shown)

• Time series averaged across all simulations

POPE:POPG• Acylated: - Arg touches first, followed by Trp and then C6 - Trp has slightly more lipid contacts than Arg - POPG contacts nearly equal POPE despite 3:1

ratio in membrane.• Non-Acylated: - No order seen in contact - POPG contacts nearly equal POPE despite 3:1

ratio in membrane - Trp makes more lipid contacts than acylated

POPC• Acylated: - C6 tails touch first (not shown), followed by

Arg and Trp - Trp makes slightly more lipid contacts than

acylated POPE:POPG• Non-Acylated: - No preference in contact order - Trp makes more lipid contacts than acylated

and non-acylated POPE:POPG

Methods• Simulation order parameters calculated using

LOOS• Acyl C-H bond orientation relative to mem-

brane normal:

• Experimentally measured by deuterium quadru-polar splitting in solid state NMR

Discussion• Subtle changes in membrane order for acylated

peptide.• Relative pattern of membrane order agrees be-

tween MD and NMR, despite differing absolute order parameters:

- POPC < POPG < POPE

3:1 POPE:POPG• 2 peptides• 100 lipids per leaflet - POPE in green, POPG in blue• Solvated to 50% w/w (7,900 waters)• 50 mM salt (plus neutralizing)• ~49,000 atoms

POPC• 2 peptides• 90 lipids per leaflet - POPC in red• Solvated to 50% w/w (7,850 waters)• 50 mM salt (plus neutralizing)• ~48,000 atoms

3:1 POPE:POPG• 2 peptides• 100 lipids per leaflet• 2,000 waters• 50 mM salt (plus neutralizing)• NPT at 50ºC

POPC• 2 peptides• 90 lipids per leaflet• 2,000 waters• 50 mM salt (plus neutralizing)• NPT at 50ºC

• CHARMM 27 forcefield• Electrostatics using PME• 10 Å vdW cutoff• NPγT at 50ºC

• γ = 32.5 dyn/cm• 2 fs time step, RATTLE• NAMD-2.6 for BlueGene/P

SCD = −1

23 cos2 θCD − 1

Lateral Distance (Å) Lateral Distance (Å)

Mol

ecul

ar O

rder

Par

amet

er

Methods• Electron density along the membrane

normal is calculated using LOOS• Membrane interface is defined as

Z-axis location of peak lipid head group density

• Electron density for peptide backbone, tryptophans, and arginines plotted relative to membrane interface

• Density is normalized for visualization purposes

Discussion• Backbone of acylated peptide resides

deeper in both membranes than the non-acylated

• Backbone is buried more deeply in POPC than in POPE:POPG

• Trp is more deeply buried in POPC• Arg remains near the membrane

interface• The acylated Arg is more buried than

the non-acylated

Methods• Trp emission fluorescence spectroscopy• 1:50 peptide:lipid ratio

Discussion• Non-acylated Trp in POPC is blue-shifted

suggesting greater water exposure• Lipid/water contacts and electron density

plots indicate Trp is more deeply buried in the POPC membrane

Methods• POPC membrane• Distance along the membrane

normal from the center of mass of the peptide to the membrane center plotted

• Representative simulations shownDiscussion

• Non-acylated peptides come off the membrane and rebind

• Acylated peptides do not leave the membrane during 12 µs of simula-tion in aggregate

• Unbinding explains greater water exposure of tryptophans despite deeper burial

The first 100ns is considered equilibration and excluded from calculations

The first 500ns is considered equilibration and excluded from calculations

• The acylated and non-acylated peptides both have subtle effects on the membrane

• The relative order is consistent between MD and NMR:

- POPC < POPG < POPE• Both peptides show significant membrane

effects at short range• Unlike the acylated peptide, the non-

acylated peptide shows no preference in binding sequence

• The acylated peptide binds deeper in the membrane than the non-acylated one

• Tryptophans reside deeper in POPC mem-branes than POPE:POPG

- Fluorescence suggests greater water ex-posure for Trp in POPC

• Coarse-grained MD shows that the non-acylated peptide comes off the POPC membrane, exposing the Trp to water

• Combining all-atom and coarse-grained MD reconciles the seemingly contradic-tory fluorescence and MD data

• Acylation increases the “stickiness” of the peptide

Methods• Only use lipids on same leaflet as peptide - Lipids must be within 10 Å of a peptide in the

plane of the membrane• Order parameters calculated using LOOS as above

Discussion• Peptide significantly decreases chain order at short

range• Slightly larger decrease in short range chain order

for acylated peptide• Larger decrease in POPC order, followed by POPG,

and then POPE

Methods• Calculate principal components for lipids• Use 2nd and 3rd principal components in lieu of

C-H bond• “Molecular Order Parameter” calculated as above• Consider only lipids on the same leaflet as peptide• “Molecular Order Parameter” binned based on

lateral distance to nearest peptide

Discussion• Significant short-range effect on membrane order

in all systems and peptides• Acylated peptide decrease short-range order more

than non-acylated• POPC < POPG < POPE• Good agreement between all-atom and coarse-

grained molecular dynamics• Origin of “hump” in coarse-grained data unclear

Time (µs)

MARTINI forcefield v2.1 with GROMACS 4.5.3 and 4.5.4

Acyl Chain Position Acyl Chain Position

Acyl Chain Position Acyl Chain Position

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.1 0.2 0.3 0.4 0.5

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.1 0.2 0.3 0.4 0.5

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.1 0.2 0.3 0.4 0.5

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6

Membrane Type Tension (dyn/cm)

Length (ns)

Avg Length (ns)

Area / Lipid (Å2)

Avg Area / Lipid (Å2)

POPE:POPG Neat 32.5242

23965.4

65.7POPE:POPG Neat 32.5 237 239 64.9 65.7POPE:POPG Neat 32.5238

23966.8

65.7

POPE:POPG C6-LfB6 32.5

536

430

65.5

65.5POPE:POPG C6-LfB6 32.5

532

430

66.3

65.5POPE:POPG C6-LfB6 32.5

530

430

65.6

65.5POPE:POPG C6-LfB6 32.5 530 430 65.4 65.5POPE:POPG C6-LfB6 32.5 350 430 65.5 65.5POPE:POPG C6-LfB6 32.5

345

430

65.1

65.5POPE:POPG C6-LfB6 32.5

333

430

65.4

65.5POPE:POPG C6-LfB6 32.5

281

430

65.3

65.5

POPE:POPG LfB6 32.5

862

1332

65.2

64.9POPE:POPG LfB6 32.5 1006 1332 64.5 64.9POPE:POPG LfB6 32.5 1661 1332 64.9 64.9POPE:POPG LfB6 32.5

1799

1332

64.8

64.9

POPC Neat 32.5

348

381

70.4

69.9POPC Neat 32.5 345 381 68.3 69.9POPC Neat 32.5 479 381 70.6 69.9POPC Neat 32.5

351

381

70.5

69.9

POPC C6-LfB6 32.5

585

643

71.1

71.1POPC C6-LfB6 32.5 672 643 71.1 71.1POPC C6-LfB6 32.5 664 643 71.1 71.1POPC C6-LfB6 32.5

652

643

71.1

71.1

POPC LfB6 32.5

1145

996

70.9

70.8POPC LfB6 32.5 839 996 70.7 70.8POPC LfB6 32.5 840 996 70.8 70.8POPC LfB6 32.5

1160

996

70.8

70.8 0.10

0.12

0.14

0.16

0.18

0.20

0.22

0.24

0.26

0.28

0.30

0 5 10 15 20 25 30 35 40

LfB6 POPELfB6 POPG

C6-LfB6 POPEC6-LfB6 POPG

LfB6 POPCC6LfB6 POPC

0.10

0.12

0.14

0.16

0.18

0.20

0.22

0.24

0.26

0.28

0.30

0 5 10 15 20 25 30 35 40

LfB6 POPELfB6 POPG

C6-LfB6 POPEC6-LfB6 POPG

LfB6 POPCC6LfB6 POPC

0 10 20 30 40

0 10 20 30 40

0 10 20 30 40

0 10 20 30 40

0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5 2.0 2.5

C6-LfB6LfB6

AbstractAbstract

ConclusionsConclusions

System ConstructionSystem Construction

SimulationsSimulations

Lipopeptide Binding MechanismLipopeptide Binding Mechanism

Effects on Membrane Structure: 2H Order ParametersEffects on Membrane Structure: 2H Order Parameters

Peptide Water ExposurePeptide Water Exposure

All-

Ato

mA

ll-A

tom

Coa

rse

Gra

ined

Coa

rse

Gra

ined

Entir

e Pe

ptid

eEn

tire

Pept

ide

Arg

inin

esA

rgin

ines

Tryp

toph

ans

Tryp

toph

ans

Distance-Based Order ParametersDistance-Based Order Parameters

Molecular Order ParametersMolecular Order ParametersMolecular Order Parameters

Electron DensityElectron Density

Fluorescence SpectroscopyFluorescence Spectroscopy

Coarse Grained MDCoarse Grained MD

All-AtomAll-Atom Coarse GrainedCoarse Grained

All-AtomAll-Atom Coarse GrainedCoarse Grained

Lightweight Object Oriented Structure Analysis

G ro s s f ie l d L a b

LOOS (Lightweight Object Oriented Structure analysis library) is a project of the Grossfield Lab and is an open-source library using C++ and BOOST to provide an easy to use and easy to extend framework for rapidly developing analytical tools for molecular simulations. LOOS is available through SourceForge at: h t t p : / / l o o s . s o u r c e f o r g e . n e t

See Poster B174on Feb. 26th

forRelated Work…

We thank the Center for Integrated Research Computing at the University of Rochester for providing the BlueGene/P. The NMR facility and mass spec facility are supported by NIH grant 1 P30 RR 031154. Funding for DVG was also provided by RR016460 and the Arkansas Biosciences Institute.

http://tinyurl.com/6t92pab

NM

RN

MR

All-

Ato

m M

olec

ular

Dyn

amic

sA

ll-A

tom

Mol

ecul

ar D

ynam

ics

All-

Ato

m M

olec

ular

Dyn

amic

sA

ll-A

tom

Mol

ecul

ar D

ynam

ics