The influence of trehalose on melting and dynamics in dehydrated phospholipidbilayers

Janna K. Maranas

User friendly packages are available for simulations in biology

NanoscaleMolecular Dynamics

NAMD GROMACSGroningen Machine for Chemical Simulationswww.ks.uiuc.edu/Research

/namd/ www.igc.ethz.ch/GROMOS

AMBERAssisted Model Building with Energy Refinement

http://amber.scripps.edu/

PACKAGES

FORCE FIELDS

AMBER

all-atom

ff94 ff96 ff99 ff99SB ff02 ff03

CHARMM

all-atom

CHARMM22 CHARMM27

united atomCHARMM19

GROMOS

united atom

GROMOS87 GROMOS96

AMBER

Nucleotides

Membranes

Proteins

Accessible properties are similar to polymers; presence of water and initial configuration are different

Initial protein structures

www.rcsb.org

Computational methods for structure prediction

Experimentally determined NMR structures

http://boinc.bakerlab.org/rosetta/

Adapted from Ding et al., Trends Biotechnol, 2005. 23,9:450-455

Complete dynamics are difficult to access

102

103

107

Simulation size of biological systems

19751970 1980 1985 1990 1995 2000 2005 2010

104

105

106

Sanbonmatsu K et al. J.Struc.Biol. 2007. 15,470-480

Num

ber o

f ato

ms

Modeling solvent effects

water models: TIP3P, TIP4P, SPC/E,F3C

Tim

e sc

ale

(s)

10-15

10-14

10-13

10-12

10-11

10-10

10-9

10-8

10-7

10-6

10-5

10-4

10-3

10-2

10-1

110102

103

Backbone librations

Large conformational

changesProtein Folding

Protein aggregation

All-atom molecular dynamics

Side chain motions

10-11 10-10 10-9 10-8 10-7 10-6

Length scale (m)

Small moleculesProtein complexes

Proteins

Conformational dynamics of trialanine in waterMu et al., J.Phys.Chem B 2003, 107:5064-5073

Possible conformations for trialanine from simulation

Ion permeation through a narrow membrane channelAllen et al., Biophys. J. 2006, 90:3447-3468

Conductance prediction across membrane (pS)

0.300.816.921

GROMOS87CHARMM27AMBER94Experimental

Educate yourself when choosing a force field

CHARMM22AMBER ff96

Ramachandran Plot (φ, ψ angles)OPLS96

GROMOS96

Models for watersimulation of ubiquitin

Showalter et al., J. Chem. Theory Comput. 2007, 3(3):961-975

AMBER force fields variantssimulation of alanine tetrapeptide

Hornak et al., Proteins: Struct. Funct. Bioinf. 2006,65: 712-725

Ramachandran plots

AMBER99NPT

AMBER99NVT

AMBER99SBNPT

AMBER99SBNVT

0

0.02

0.04

0.06

0.08

0.1SPC TIP3P

Variations within force field families and with water are less important

Resurrection PlantResurrection Plant((SelaginellaSelaginella))

Trehalose and sucrose: preservation of biologicals•Proteins: enzymes, antibodies•Cells: stem cells, platelets, bacteria, sperm, seed embryos

Nature uses sugars to survive dessication

PhospholipidPhospholipid

Head GroupHead Group

Alkyl Alkyl TailsTails

De-hydration Re-hydration

hydrated bilayer

bilayer dried with trehalose

hydrated bilayer

bilayer dried without trehalose

What drives melting of freeze-dried liposomes: tails & heads?

What is the nature of the transition and “liquid crystalline” state with trehalose?

How does trehalose affect the dynamics of lipid heads & tails?

Trehalose stabilizes membranes during re-hydration

•hTrehalose•dhDPPC•hdDPPC•hhDPPC+dTrehalose•ddDPPC+hTrehalose•dhDPPC+dTrehalose•hdDPPC+dTrehalose

hhDPPC

dTrehalose

dhDPPC

This study combines simulation with QENS

dhDPPC

QENS follows proton motion

“hide” specific parts with deuteration

Systems investigated•128 DPPC molecules•100 water molecules•256 trehalosemolecules

•384 DPPC molecules•300 water molecules

QENS

Atomistic MD

Layer thickness

The GROMOS force field is accurate for this system

Area per lipid, 320 K

Expt: 64 Å2

[ave. value over xray, NMR, neutron]

Sim: 60 Å2

Hydrated bilayers

Dry bilayers

We combined HFBS and DCS dynamic runs and used HFBS for elastic scans

Δt ≅ 0.5-40 ps Δt ≅ 240-5000 ps

NG4 Disk-chopper time-of-flight spectrometer

NG2 Backscattering spectrometer

Dynamic run

Elastic scan

dhDPPC hdDPPC

Q=0.62-1.75 Å-1

melting transition: lipid tailsSignificant mobility before Tm

Elastic Scans: melting ofTails, Heads andWhole Lipid

Melting in dry bilayers without trehalose is driven by lipid tails

0

0.5

1

1.5

2

2.5

3

3.5

4

200 250 300 350 400 450 500Temperature (K)

Inte

nsity

(arb

. uni

ts)

hPMMA

hPEO

amorphousSemi-crystalline

Melting: lipid tails

Three regimes in tail dynamics:1. Vibrational motion2. “Localized” motion3. Translational motion

Dry Bilayer395K

Dry Bilayer310K

Molecular Simulations

Previous QENS hydrated bilayers: equivalence of tail protons

(König et al.,J. Phys II, 1992)

Head and Tail mean displacement

Mobility varies between heads and tails and with tail position

( )2 2exp / 3I Q u∝ −No translational motion:

u = r – re

re

Trehalose restricts headgroupmobilityTrehalose and phospholipidheadgroups: similar mobility

Q=0.62-1.75 Å-1

TrehaloseHeadgroups

Trehalose+LipidHeadgroups+Sugar

Trehalose restricts headgroup mobility

QENS: elastic scans

QENS T=395K HFBS

0

0.2

0.4

0.6

0.8

1

100 1000 10000time (ps)

I (q,

t)

d-DPPC + h-trehalose at q=1.60

hd-DPPC + d-trehalose at q=1.60

QENS: dynamics at 395 K

PMMA

“poly methyl methacrylate”

This behavior is encountered in polymer mixtures

Without hydrogen bonds: ΔTg = 180K

-4

-2

0

2

4

6

1.5 2 2.5 3 3.5

1000/T (K-1)

log

tau

(ns)

CH2 C

CH3

CH3

C

O

O

CH2 OCH2

PEO “poly ethylene oxide”

Tg ~ 220 K, more mobile

Tg ~ 400 K, less mobile

Significant sugar-headgroup hydrogen bonding is likely

* Zhang, Runt, et al., Macromolecules (2004, 2003)

0

0.2

0.4

-2 -1 0 1 2 3 4 5 6log10[Frequency (Hz)]

95908580757065605552

40% PVPh/PVEE

One dynamic response: 30 - 70% PVPh

Hydrogen bonding = similar dynamics

Trehalose and lipid headgroups have significant hydrogen bonding

With hydrogen bonds: ΔTg = 185 K

Q=0.62-1.75 Å-1

Trehalose lowers melting transition and restricts mobilitySome differences below TmLarge difference above Tm

Tails with and without, sugar

Trehalose also restricts mobility of lipid tails

Dry Bilayer395K

Dry Bilayerwith trehalose395K

??

No translational motion:

Trehalose decreases the mean displacements and restrains the lipid mobility

( )2 2exp / 3I Q u∝ −Tails mean displacements withand without trehalose

Molecular Simulations

Dry Bilayer/Trehalose395K

Dry Bilayer/Trehalose310K

Molecular nature of mobility is different when trehalose is present

The melting transition differs with trehalose

310 K

395 K

Tm“melting” = bottom of tails

Without trehaloseWith trehalose

Tm“melting” = top of tails395 K

310 K

Dynamic runs were made on tail labeled samples

0.0001

0.001

0.01

0.1

1

-6 -4 -2 0 2 4Energy transfer (meV)

Inte

nsity

(arb

. uni

ts.)

dhDPPC 390KResolution DCS

HFBS

Δt ≅ 0.5-40 ps Δt ≅ 240-5000 ps

NG4 Disk-chopper time-of-flight spectrometer

NG2 Backscattering spectrometer

τ: characteristic timeβ: distribution of times

E: motion in restricted geometry

DCS HFBS

Tail motion in QENS window has two processes

( )( ) ( )[ ]i

iiii tEEKWW

KWWKWWAtqIβτ/exp1

, 21

−−+=

=

τ: typical time scale for trans/gauche isomerization

If varying mobilities, β < 1

KWW Limiting cases

β= 1: same τ for all protons

E = 0: no spatial restriction

Fit lines

Fast process: trans/gauche isomerization 7-45 ps

Slow process: lipid rotation 2-12 ns

fast processQ=0.98

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

1 10 100 1000

time (ps)

I(Q

,t)

T=310 dhDPPC T=330 dhDPPCT=350 dhDPPC T=370 dhDPPCT=395 dhDPPC T=370 dhDPPC+dtreh

Isolating the fast process reveals conformational transitions

( ) ( )[ ]11111 /exp1 βτtEEKWW −−+=

trehalose decreases conformational transitionseffect is similar to decreasing T to between 310 & 330larger spatial restriction of protons

below Tm

fast process Q=1.58

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

1 10 100 100time (ps)

I(q,

t)

T=310 dhDPPCT=330 dhDPPCT=350 dhDPPCT=370 dhDPPCT=395 dhDPPCT=370 dhDPPC+dtreh

0

1

10

100

-0.20 -0.10 0.00 0.10 0.20 0.30 0.40

log Q (angstrom-1)

τ1

T=310 dhDPPC T=330 dhDPPCT=350 dhDPPC T=370 dhDPPCT=395 dhDPPC T=370 dhDPPC+dtreh

y = 7.1081x-1.9801

y = 3.0589x-2.1704

0

2

4

6

8

10

12

14

16

18

0.00 0.50 1.00 1.50 2.00 2.50

Q (angstrom-1)

τ1

T=310 dhDPPCT=330 dhDPPCT=350 dhDPPCT=370 dhDPPCT=395 dhDPPCT=370 dhDPPC+dtreh

Trehalose slows conformational transitions and increases spatial localization

Characteristic times: Slower with decreasing T

Little change with trehalose

Hydrated bilayers, T=333 K: τconf = 7-45 psVenable, et al, Science, 262, 223 (1993)

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.50 1.00 1.50 2.00 2.50Q (angstrom-1)

E1

T=310 dhDPPCT=330 dhDPPCT=350 dhDPPCT=370 dhDPPCT=395 dhDPPCT=370 dhDPPC+dtreh

Spatial localizationLarger with decreasing

temperature AND with trehalose

Modeling of spatial localization quantifies the variation in mobility along lipid tails

Q dependence of EISF

( ) ( ) 2

13⎥⎦

⎤⎢⎣

⎡=

QrQrjQE

( ) ( )∑=

⎥⎦

⎤⎢⎣

⎡=

N

n n

n

QrQrj

NQE

2

2

131

Single sphere size

varying sphere sizes

E1

E1

Pure lipid tails

Diffusion in impenetrable spheres

Simulation directly probes difference in moblity with carbon position

Simulation: neat lipid tails

E1

QENS with simulation input

Five things to rememberlipid headgroups and trehalose:

significant hydrogen bonding

Spatial extent of mobility varies with tail postion

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.50 1.00 1.50 2.00 2.50Q (angstrom-1)

E1

T=310 dhDPPCT=330 dhDPPCT=350 dhDPPCT=370 dhDPPCT=395 dhDPPCT=370 dhDPPC+dtreh

Trehalose decreases spatial extent of mobility

Melting: lipid tails

nature of main transition may be different with trehalose