Von Willebrands Factor = proteininvolved in blood clotting,

tobascothwackΠολεοδομικά Έργα

15 Νοε 2013 (πριν από 3 χρόνια και 6 μήνες)

62 εμφανίσεις

http://www.google.com/images?rlz=1T4GGLF_enUS278US27
8&q=von+willebrands+factor+clotting&um=1&ie=UTF
-
8&source=og&sa=N&hl=en&tab=wi&biw=987&bih=503

Von Willebrands Factor = protein

involved in blood clotting,



forms large multimers, binds platelets and vessel wall,



subject to large shear forces.

How does force affect its interactions?

Novel experimental set
-
up to study effect of force



on interaction between A1 and GPIb
a

E
xperimental readout of force at which domains separate

Note they are watching
individual

molecules


they can repeatedly test the
same

molecule



unbinding and rebinding forces are stochastic







Some technical details


How long is linker? 43 amino acids,
~
16nm fully extended


H
as to be long enough to see rupture


Why use DNA “handles”? To keep proteins off surfaces,


separate laser trapped bead from rest of assembly,


apply force at single point


How to attach proteins to DNA? via S
-
S bonds with extra


unique cys in protein and SH in synthetic DNA

How does the laser trap work?

I
dea: refraction changes direction (momentum) of light

Since momentum is conserved, bead’s momentum



changes by


D
p


F =
-
d
D
p/dt, mainly directed toward





most intense part of beam

D
p

http://www.youtube.com/watch?v=BL9gmMzpRr4&feature=PlayList&p=DB423384C9BB437D&index=1



for more detailed explanation of trap set
-
up used here

Watch videos at
http://tweezerslab.unipr.it

or

Optics


counterpropagating

laser traps

beam

position

force

Infer bead position from force, knowing stiffness, x=F/k

R
elation between force and
D
extension



D
x = length of “open” A1
-
linker
-
GPIb
a

+ DNA


-

length of A1
-
GPIb
a

complex + DNA


DNA length should be constant at given F


If A1 and GPIb
a

domains don’t stretch,
D
x due to


extension of flexible aa linker and rot. of A1, GPIb
a









Bin data by force,

show mean and

v
ariance of
D
x

a
t rupture or rebinding


Results c/w expected

s
tretch in linker w/force,

h
ence, c/w single tether

How does loading rate, F(t), affect d
n

of unbinding forces?

E

p
ulling direction x

D
E

Rate of escape at F=0
~
e
-
D
E/kT




For constant pulling force F,



energy barrier decreases by F
D
x,



so rate
~
e
-
(
D
E
-
F
D
x)/kT
=> av. lifetime


<
t
(F)> = <
t
(0)> e
-
F(
D
x/kT)
= e
-
F
s/kT

D
x

analysis to determine relationship between probability of
unbinding at force f for a given F(t) and the average lifetime
at
constant

force,
t
(F), which should depend exp’lly on F

If F changes with time, F(t),

d
n

of unbinding forces

changes, and need careful

D
udko
-
H
ummer
-
S
zabo provide this analysis



(PNAS 105:15755 (2008))


Relation between p(f) given F(t) and
t
(F) is:

If data grouped into

force bins of size
D
F;

h
i

= p(rupture in i
th


force bin); in k
th

bin,

F = F
0
+(k
-
1/2)
D
F

e
valuated at F=F
0
+(k
-
1/2)
D
F

=

calculated <
t
(F)>

Obs.

p
(f)

Kim et al observe p(f) at different loading rates, find bimodal







distributions at







intermediate







loading rates,







then calculate







<
t
(F)>
for each peak







according to D
-
H
-
S

k
1(or2),off

=
t
-
1

extrapolated to F=0

2

Note ln
t
(F)
~
F for each

peak as in simple theory


Their interpretation:

2 different bound states,

e
ach with its own exp’l

dependence on F (slope
s
/kT)

and predicted zero force

unbinding rate k
1,off

, k
2,off

1/k
2,off

1/k
1
,off

Repeat expts at constant pulling F close to value

at which bound/unbound states are equally likely,

w
atch transitions back and forth B<
-
>UB

From these data calculate prob.
o
f staying in bound

state for time t

Two exponentials at intermediate forces, c/w 2 bound

states with different unbinding rates (that are fns of F)

s
traight line => p
survival

~

e
-
kt

; k = rate of unbinding

exp’l

survival prob. expected for process with single


random event (like radioactive decay)

Their model: 2 bound states interconvert at (force
-

dependent) rates k
12

and k
21

estimated from data

Note both bonds are weakened by force (
s
>0, “slip” rather

than “catch”) but conversion to bond with lower k
off

and

smaller
s

-
> tighter binding at high shear (“flex” bond)

Ristocetin (antibiotic) and botrocetin

(
snake venom toxin) are drugs

known to affect vWF


They promoted conversion to state 2

and decreased response to force (
s
)

Main points


Novel set
-
up to repeatedly measure interaction between

2 protein domains in single protein as function of force


Nice “corroboration” of D
-
H
-
S theory that relates unbinding

force distributions at different loading rates to lifetimes at

constant force, with complication that…


Their system had bimodal unbinding force distributions at

some loading rates. This suggests protein complex exists in

2 different states, each state behaving as expected for

simple model with ln(
t(
F
)
)
=

F
s
/kT, but different

sensitivities (
s
’s) for the 2 states

The 2 bound
-
state model is biologically interesting since

it leads to a more graded response to force loading for

a
n interaction naturally subject to comparable forces


Will it turn out that many protein interactions have alter
-

native binding states?


Single
-
molecule pulling experiments reveal complexities of
protein interactions that would be hard to see in bulk assays


Compared to AFM, newer pulling methods allow study in

lower force regime (pN vs nN), and away from surfaces

which can
-
> artifacts

Relevance to nanotechnology in other areas



methods to detect



forces
~0.1
-
10pN



distance changes
~
10nm



ways to relate rupture
-
force distributions



at different pulling rates to lifetimes



at constant force



? interest from ME point of view


combining



2 “slip” bonds with different force sensitivities



can mimic “catch” bond (increase in F
-
>



sudden increase in bond strength)