Membrane Biophysics & Soft Matter Physics

Membrane Biophysics &

Soft Matter Physics

Huey W. Huang

Rice University, Physics & Astronomy http://hwhuang.rice.edu

Taida: March 3, 2015

The talk is about

• Proteins interacting with membranes.

• Physical process rather than chemical reactions.

• Functions of proteins in membranes are well defined, in fact by phase transitions.

• Unsolved problems.

Subject

• Membrane-active Antibiotics, often called

antimicrobial peptides (AMPs). This talk is the story of how we found out how AMPs work.

What is the significance of AMPs?

What is the physics problem?

• Unsolved membrane problems

Recently a 2^nd kind of AMPs were discovered.

Alzheimer’s disease, mad cow disease, type II diabetes and other neurodegenerative diseases could also be membrane problems.

Cell Membranes

C. elegans The red part is a lipid bilayer.

target of conventional antibiotics

target of membrane- active antibiotics

Bacterial membranes

E Coli Filament Stop Solution

77% Stop Solution 55% Stop Solution 32% Stop Solution 15% Stop Solution

Cell membranes are not simple.

-DP

Live cell

Dead cell

Biophys J. 107, 2082 (2014) Bar=2.5um

It is very difficult to know what happens when an antibiotic attacks a bacteria except that it kills.

After Diffusion: LL-37 6uM+40% Stop Solution

5+0.5min 5+4.5min 5+8.8min 5+12.2min 5+12.4min 5+17.5min 5+20.7min

Model membranes and AMP

1 nm X 2-3 nm

A typical antibiotics or AMP (melittin)

5nm

Soft!

Vesicle

This is the problem:

Model membrane attacked by membrane-active antibiotics

-DP

30 mm

Green

Red dye

PNAS 110, 14243 (2013)

Parallel Multilayers of Membranes

Liposomes (vesicles) Multilayers (smectics)

side view

top view

Neutron scattering

in-plane scattering

Using D₂O to show the water pores

Natural lipid with D₂O or H₂O perdeuterated lipid with H₂O or D₂O

D₂O

D₂O H₂O

H₂O

Biophys. J. 70,, 2659 (1996)

Analysis of neutron scattering from fluid membranes

)2 ( rq F

( rq ) S

) 2 (

)

(qr S qr F

I 

Pore size ~4.4 nm diameter 4-7 melittin in the pore

Phase transition by dehydration

dehydration

Biophys. J. 79, 2002 (2000)

Diffraction by molecular crystalline

Diffraction by soft matter crystals Same S(q), but

Ex. 1D constant density In the unit cell

q

Anomalous Diffraction for centrosymmetric structures

2

"

'

) exp(

)

"

' (

) exp(

) (

f F if F f

i if

f f

i f

F

o n

k

n j

j

n j

 



















 q

Multiwavelength Anomalous Diffraction (MAD) Method

' 1 . 0

~ '

' _

 f

f

JACS 128, 1340 (2006)

di18:0(9,10Br)PC

Solving F

and F

F

and F

Complete electron density Label only electron density

Only one AMP forms

pores of the (barrel-stave model)

PNAS 105, 17379 (2008)

All AMPs except one form pores of the (toroidal model)

PNAS 105, 17379 (2008) PNAS 110, 14243 (2013) A topological question!

Physics of pore formation in membranes

• But why do the antibiotics make pores when DA/A exceeds ~4%?

• Note that making pores when DA/A exceeds

~4% represents a concentration on-off switch.

• All biological on-off switches are by concentrations!

The biggest problem in membrane biophysics is:

How to detect the physical state of proteins in membranes?

Method 1: Oriented circular dichroism

We measured the orientation of helices as a function of antibiotic concentration.

Method 2: Lamellar diffraction

We measured the membrane

thickness as a function of antibiotic concentration.

JCP 89, 2531 (1988) BJ 57, 797 (1990)

BJ 68, 2361 (1995)

Biochemistry 34, 16764 (1995);

BJ 84, 3751 (2003)

We detected a critical concentration.

Physics of pore formation in a thin layer









2 R R

E

 







 / E

R

Litster, Phys. Rev. Lett. A35, 193 (1975) Taupin et al. Biochemistry 14, 4771 (1975)

Phase transition as a function of P/L.







 ²

2 R R

E_R^o  

) / )(

/

(A A P L

K_{a P L}

o 



L P

P A

A

K_a ( _P / _L)(  _I ) /

 

(for P/L>P/L*)

R N

P_I  _p_ 2







 2 R R² (4/3) ²R³ (N / P) E_R    _o  _{o p}

R_o

P/L<P/L*

P/L>P/L*

P/L*

Phys. Rev. Lett. 92, 198304 (2004)

With this understanding, we can now go back to the case of live cells.

-DP

Live cell

Dead cell Bar=2.5um

5+0.5mi n

5+4.5mi n

5+8.8mi n

5+12.2 min

5+12.4 min

5+17.5 min

5+20.7 min

of pore formation story

Although the pore-forming antibiotics have not yet been approved as drugs.

The 2^nd kind of membrane-acting antibiotics (daptomycin) do not make pores.

Ca⁺⁺ dye Daptomycin, Ca⁺⁺

-12 -10 -8 -6 -4 -2 0 2 4 6

0.93 0.94 0.95 0.96 0.97 0.98 0.99 1 1.01 1.02

0 100 200 300

Intensity

Time (s)

control intensity dA/A

DA/A

No leakage

Biochemistry 53, 5384 (2014)

Amyloidoses

(one of the most important medical problems) Alzheimer’s disease

Type II diabetes Mad Cow disease

Parkinson’s diesease other ~20 amyloidoses

Common characteristics of amyloidoses

• Each disease is strongly correlated with a protein.

• The disease is associated with the presence of

protein plagues.

• But the fibrils and plagues do not harm cells.

Characteristics II (somewhat controversial)

• Protein-membrane interactions turn the proteins into the plaques. During this

interaction something happens to the cell membranes. But how?

We need experimental tools to study

proteins in membranes.

Acknowledgement

Rice University:

Glenn Olah (Los Alamos) Yili Wu

Ke He (Shell)

Steve Ludtke (Baylor MC) William Heller (ORNL) Thad Harroun (Brock U) Lin Yang (NSLS)

Thomas Weiss (SSRL) Lai Ding (Tuft)

Wangchen Wang (Baylor MC) Shuo Qian (ORNL)

Yen Sun (Harvard/Rice) Chang-Chun Lee (CGG) Tzu-Lin Sun

Collaborators:

Lin Yang (NSLS) Shuo Qian (ORNL) Ming-Tao Lee

(NSRRC) Wei-Chin Hung

(Mil.Acad.tw)

SUPPORTED by:

Method of Oriented Circular Dichroism

JCP 89, 2531 (1988) BJ 57, 797 (1990)

Above a critical concentration (P/L)*, Peptide orientation changes with (P/L)

S

I

BJ 82, 908 (2002)

All pore-forming peptides studied showed critical orientation transitions.

Membrane Thinning Effect

BJ 68, 2361 (1995); Biochemistry 34, 16764 (1995); BJ 84, 3751 (2003)

Membrane thinning and peptide orientation change have the same critical concentration.

Peptide-induced pores are stable.

3 3 2 2

1R c R c R

c

E_R   

) 3 / ( )

3 (

) 3

(c₂ c₃ c₂ c₃ ² c₁ c₃

R_o   





L P c

c₂ / ₃  3( / ) 4( _p / )_  3.1

R_o~1-2nm

c₁=2 decreases with P/L. _R

o P/L<P/L*

P/L>P/L*

P/L*

Phys. Rev. Lett. 92, 198304 (2004)

The diseases are each associated with the presence of plagues (fibrils) of one particular

protein that misfolds.

The disease is strongly correlated with the protein.

b-cells co-secret insulin and amylin (an amyloid protein).

Human amylin and rat amylin differ by a few amino acids.

Human has diabetes; rat has not. But if the rat gene is modified to human gene, rat develops diabetes.

Mad cow disease (bovine spongiform encephalopathy) can be transmitted to human beings by eating the

animal protein.

Penetratin binds to the membrane and comes out.

Biophys. J. 98, 2236 (2010)

Penetratin in membranes

Peptide donformation change: CD vs. P/L

A topological question.

 

[ 2 c₁ c₂ c ² c₁c₂ dA

H  _{  o }

 

Gauss-Bonnet Theorem (for a closed surface)

Helfrich (1973)