Shuttling Device for Field Cycling Experiment on NMR Relaxation Study

dinnerworkableΠολεοδομικά Έργα

16 Νοε 2013 (πριν από 3 χρόνια και 8 μήνες)

416 εμφανίσεις

Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


1

Shuttling Device for Field Cycling Experiment on NMR Relaxation
Study

Ching
-
Yu Chou
1
,3
, Chi
-
Fon Chang
2
, Tai
-
Huang Huang
1
,3

1

Institute of Biomedical Science, Academia Sinica, Taipei, Taiwan

2
Genomics Research Center, Academia Sinica, Taipei, Taiwan

3
Depart
ment of Physics, National Taiwan Normal
University

(
e
-
mail address:
cychou
@
gate
.sinica.edu.tw
)


Abstract:

Dipolar
-
dipolar (DD) interaction and chemical shift anisotropy (CSA) effect are
two major sources of nuclear magnetic relaxation at high field. Convent
ionally
molecular motions are extracted from analysis of longitudinal relaxation rate (R
1
),
transverse relaxation rate (R
2
) and Nuclear Overhauser effect (NOE) measured at a
fixed field which cannot distinguish the contribution from DD interaction and CSA
effect. Since CSA contribution is linearly proportional to the magnetic field whilst the
dipolar interaction is field
-
independent one can separate these two contributions from
relaxation rates measured at multiple fields. We have built a field cycling appa
ratus
based on a Bruker AVANCE600 NMR spectrometer. By shuttling samples vertically
to the desired heights inside the magnet for relaxation and back to the central position
for detection, one can measure relaxation at lower fields with high resolution (fr
om
0.04 Tesla to 14.1 Tesla). The round trip shuttling time from the central position to the
top of magnet is about 0.16s, which is suitable for measuring relaxation rate in the
range up to 10s
-
1
. In this poster we will present the details of field cycling

apparatus
design and some preliminary results of the dynamics of single amino acids and a
di
-
peptide based on field dependent
13
C T
1

measured with the shuttling device
.

Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


2

Introduction:


NMR relaxation study is the most popular experiment for mol
ecular dyna
mics
investigation.
It can provide a detailed
description of protein dynamics.
The relaxation
parameter
s

in NMR are longitudinal relaxation rate (R
1
), transverse relaxation rate (R
2
)
and nuclear Overhauser effect (NOE)

(1
-
7)
. These three parameters are

most often
acquired

and
used to
map out the spectral density
function

describing such motions at
a number of frequencies
.
Therefore, these three parameters are composed of field
dependent spectral density functions and

are field
dependent

function
s. Within these
three parameters, the most manifest field dependent parameter is R
1
.
(3, 8
-
10)


The field dependent curve of R1 is
convoluted

by two major
mechanism
s
affecting molecular
dynamics, dipole
-
dipole interaction (DD) and chemical shift
anisotropy (CSA).

In order to solve the spectral
density

function, there are several
models.

(5, 11, 12)

The most popular analysis model for solving the sp
ectral density
function is the
Lipari
-
Szabo

model, so called

Model
-
free analysis

.
(13
-
15)

In this
Model
-
free analysis,
assumed

to be

observing molecules have the electronic gradient
symmetry or chemical shift symm
etry.
(5, 16)

In other words, the chemical shift tensor
is a symmetric tensor. Under this condition, the spectral density functions contributed
by dipole
-
dipole
interaction and chemical shift anisotropy are the same. However, in
protein dynamics study, the most studied cases are
15
N and
13
C which have
asymmetric

CSA, especially
13
C.
(17)

In order to investigate protein dynamics
for real
cases
, the

field dependent relaxation
measurement has been developed.
(18)

By measuring the R1 at
various

fields, these two
mechanisms

could be
deconveluted
mathematically;

meanwhile, the molecular dynamics could be d
escribed
separately from these two effects.
Our
preliminary

study cases
were

single amino
acids, Ala and Glu
, and a short peptide with two amino acids, Ala
-
Gly.
According to
the solid state NMR observations on 20 single amino acids
(17)
, c
arboxyl has
a

large
asymmetry CSA.
Because

of this property of carboxyl atom in amino acids, the
dynamics
contributed by

CSA
could be

extracted from field dependent relaxation
curve

and
detail

dynamics information has been revealed as well.




Methods and

materials:


Field dependent R1 theory
.

The molecular dynamics information is accomplished
by measuring NMR relaxation parameters R
1
, R
2
, NOE. The connection between
NMR relaxation parameters is a spectral density function describing molecular
motions in f
requency dimension.
(5)

The contributed mechanisms of NMR
Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


3

relaxations are majorly dipol
e
-
dipole interaction (DD) and chemical shift anisotropy
(CSA). These two mechanisms are commuted in
Hamiltonian

so the mapping
relaxation rates are linear combination. The relaxation rate functions of spectral
densities are in the table I of the literature

(5)
.


In order to solve the spectral density J(

), Lipari and Szabo

had developed a
model called “model
-
free” analysis.
The most recent dynamics studies based on
relaxation data obtained at a fix high field, usually greater than 9Tesla.
This

analysis

method is

calculated based on R
1
, R
2
, NOE at a fixed

field

and in
the

case where

the
electric field gradient or chemical shift tensors have axial symmetry,

=0, the
required spectral density is the same in both mechanisms.
(5)

However, the relaxation
theory

shows that

the relaxation rates are field dependent functions. Among these
three
relaxation
parameters, R
1

has the apparent field dependen
t behavior and clear
relationship between spectra
l

density J(

). The relationship between spectra
l

density
J(

) and longitudinal relaxation rate (R
1
) can be
expressed

by
two
interaction
mechanisms
:

dipol
e
-
dipole

interaction and chemical shift anisotropy (C
SA).

R
1
=
R
1
dipolar
interaction
R
1
Chemical
shift
anisotropy

………………

[1]

The

dipolar interaction

term is
,
(3, 4)

R
1
dipolar
interaction
=
0
4
2

2
H
2
C
2
r
CH
6
[
0.1
J
H

X
0.3
J
X
0.6J
H
X
]
………………

[2]

Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


4

Where spectral density J(

) is the linear combination of Lorenze func
tions,
J
=
2
1
2

2
. The larmor frequency is


and correlation time is

.

T
he chemical shift anisotropy effect

is determined by
,

R
1
CSA
=
1
15

2
1
2
/
3

X
B
2

[
S
C
2
2
1
2
2
1

S
C
2
2
f
1
2
f
2
]

..
………

[3]

where B is the strength of magnetic field, is gyomagnetic ratio for X nuclear,



and


CSA tensor. The dynamic parameters rep
resent the qualitative analysis of molecular
motions, which are the molecular correlation time

, the effective proton distance
r
e
,
the CSA order parameter

S
C
2
, and the correlation time for
local

motions

f
.
(10)


According to the field dependent behavior of

R
1

relaxation rate, the mechanisms
could be deconvoluted and the dynamics parameters could be calculated without
assuming the
symmetry CSA as

in

model
-
f
ree analysis.
The
resulting

R
1

from DD and
CSA is shown in figure 1.

The field dependence analysis
could

be
divide
d

in
to

three
parts based
on
frequency

domain
. (i) In high field range (greater than 6 Tesla), the
field dependenc
e
is mainly contributed by CS
A effect.
Manifest increasing R1 with
magnetic field (positive correlation
with

magnetic field)
presents

asymmetry CSA
on

observing atom.
In case of
symmetry CSA
no
obvious field dependent

R
1

curve

at
high field

could be seen
.
(ii)
In

the middle magnetic f
ield range (
about 2 to 4 Tesla
)
,

dipolar effect and CSA have equal contributions.
This area presents
the
correlation
time
dependency

of R
1
.
The simulation curves with different correlation times are
shown in figure 2.
(iii) As for the fringe field range, d
ipolar effect has the principal
contribution.
R
1

value at the fringe field exhibits the distance between the nearby spin
and correlation time from dipolar effect.


Therefore, molecular dynamics
informatio
n could be revealed by studying field
dependent R
1
.


Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


5

CSA
Dipolar

Figure
1
.

The field dependent R1 curve. The data points are
13
C
-
R
1

on carbox
yl in the peptide bond of
the peptide, Ala
-
Gly. Simulation curves of R
1

contributed by dipolar effect and CSA in red and blue,
respectively. The fittin
g curve of these data points is in green.


Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


6


Figure
2
. The simulation curves of field dependent R
1

with

various correlation time

. The calculation
was from 1 ns to 1ms.



H
ardware setup
.

To measur
e field dependent

R
1

value
,
a device for
moving
sample to different

magnetic field
s

in short time is required for this field cycling
experiment.
(8
, 18
-
20)


This apparatus is designed for pneumatically move an aqueous liquid NMR
Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


7

sample from the center of a Bruker 600 MHz
magnet up to

any position
at

its
corresponding

field. The hardware setup is all homemade product based on original
design from Pro
f. Redfield
(18)
. We have made some modifications
to

assemble

sample

easily

and
stabilize

the

air control. The shuttling rate has also been increased
compared to the original design
.

The shuttle tube is d
esigned for shutting the NMR sample between acquisition
position (the center of the magnet) and target position (with desired magnetic field)

(see figure 3)
. The inhomogeneitic magnetic field induces that different vertical
positions have different magneti
c strengths. Therefore,
a single
high field magnet

could

not only provide high resolution measurement but also the field variety.
Figure
4 presents the shuttling direction and magnetic field
inhomogeneity

at a Bruker
AV600 spectrometer
.


Figure
3
. Homemade shuttle tube for shuttling
NMR
sample in
side

the spectrometer.

Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


8


Figure
4
. The cartoon of shuttling tube direction and magnetic field
inhomogeneity

with vertical
positions.



We
utilize
d

the air pressure

difference to shuttle the NMR sample in the shuttle
tube. The NMR tube
wa
s hold by a special design
,

which c
an

move smoothly in the
shuttle tube and tolerate high speed shuttling.

The whole piece is shown in figure 5
and called

shuttler

.

Figure
6

is the

gas control station
(
collaborated

with Festo
Company
) which

control
led

the gas in and out to the shuttle tube. While
rising

up the
sample from the center, the air pressure above the shuttler decrease
d
; while pushing
down the sample to the center, the air
pressure
was

operated oppositely.

Th
e round
trip shuttling time to

different
vertical

positions is exhibited in figure
7.
The round trip
time limit
wa
s about 160ms,

and the time scale is

suitable for relaxation rate
measurements in the range up to about 10
s
-
1
. The time
limit

wa
s shorter than the
Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


9

200ms in the
Magnetic Resonance in Chemistry, 2003

describe
d

by A. Redfield
(18)
.

Our
newly
setup
air

control system
could
operate the field cycling experiment
at
fas
ter

shuttling time and high
stability
.


Figure
5
. Sample holder for 5mm NMR tube.
The glass tube
contained

aqueous

NMR sample.



Figure
6
. The gas station for controlling the
upward and downward gas.


Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


10


Figure
7
. Round trip shuttling time with different vertical positions.


A
mino acid and dipeptide
.

Amino acids, Ala and Glu, were purchased in CIL
(Cambridge Isotropic Label). The peptide, Ala
-
Gly, was from BACHEM.


Experimental
design
.
The carboxyl atoms at peptide bond in this
Ala
-
Gly
peptide and COOH
for individual amino acid
have been
investigated

by field cycling
experiment.
R
1

was obtained by
13
C detection 1D NMR.



Results and Discussion:


We have
conducted

the field cyclin
g technique to determine the spin
-
lattice
relaxation rate
R
1

of single amino acids, Glu and Ala, and di
-
peptide Aly
-
Gly.


The
investigation

of
R
1

at middle field range
on carboxyl of Ala in the dipeptide
and single Ala presents qualitative and
qualitative

discoveries.
First,
comparing the
field
dependency

in the field range of 6 to 10 Tesla, figure 8 (a) showed the positive
correlation between R
1

and magnetic field while figure 9 showed the saturation curve
at the same range. This
comparison

indicated the q
ualitative discovery that t
he
carboxy
l

atom at the peptide bond of di
-
peptide experiences asymmetric CSA instead
of symmetric CSA which the single Ala does. This result present
ed

that the CSA
symmetry depends on the chemical environment. Second,

because of

the CSA
symmetry of the
single

Ala, the single Ala ha
d

no local fluctuation motion influenced
by CS
A

only. That is, the single Ala has the
isotropic

molecular motion while Ala ha
s

anisotropic molecular motion

in dipeptide

with about 1ns local
correlation

time
influenced

CSA.

Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


11

Concerning peptide bond
mobility
, the carboxyl atom in the dipeptide, Ala and
Gly were calculated
by

the equation [3]. T
he carboxyl atom of peptide bond ha
d

longer CSA correlation time (~1ns) than C
-
terminal

(Gly carboxyl) ha
d

(95.7ps)
. This
result indicate
d

peptide bond reduce
d

the mobility significantly.

In addition to the relaxation investigation of Ala
, another amino acid with long

side chain, Glu, was also studied.

Field dependent relaxation rate studies of t
he
carboxyl atom and
carbon in acid COOH were presented in figure 10. Molecular
motion

of carboxyl atom on different type of amino acid was exhibited by comparing
the relaxation data of the single Ala and the single Glu in figure 9 and 10
, respectively
.
Since t
he R
1

as the
fun
ction
of magnetic field at the range

2Tesla to 14Tesla re
flected
CSA symmetry difference,

the

different field correlation of

carboxyl R
1

o
n

Glu
brought asymmetry CSA out.
Therefore,
c
arboxyl atom of Glu experience
d

the
asymmetry CSA while
carboxyl

atom of
Ala
did

not.
In addition to

the carboxyl
dynamics, the COOH dynamics was also investigated
in

figure 8 (b.) and figure 10.
These two similar field
-
dependent R
1

curves of COOH at the dipeptide and Glu
presented the similar molecular motion.
Consequently,

th
e long side chain influence
d

CSA at carboxyl atom and
the carboxyl dynamics behavior
; however,
the long side
chain

did

not influence
COOH dynamics
.

Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


12


Figure
8
. F
ield dependent R
1

curves of dipeptide, Ala and Gly. The fitting curve
s are in red. The
deconvoluted curves contributed by dipolar effect and CSA are in blue and green, respectively. The
nonlinear
regression
analysis of CSA effect
is

separated in high field and low field which are in dash
line and in solid line, respectively
.
(a.) is the carboxyl atom on Ala; (b.) is the carboxyl atom on Gly.


Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


13


Figure
9
.
Field dependent
13
C
-
R
1

on carboxyl of a single
Ala
.
The R1 curve has the saturation tendency
at the high field range, greater than 10Tesla.



Fig
ure
10
.
Field dependent
13
C
-
R
1

on carboxyl
and COOH
of a single Glu. Carboxyl atom relaxation
curve and data points are in red; carbon relaxation curve

and data

points

of COOH
are in yellow.



Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


14

Summery:

According to the
fundamenta
l

NMR relaxation theory, the molecular
motion
is
a

frequency dependent relaxation function. Hence, t
he field cycling technique is
essential for a
dvanced relaxation measurement,

especially
for

nuclei

31
P

or

13
C
with
large CSA
effect. It could deal molecular

m
otion not only isotropic or axial symmetric
but also anisotropic CSA. Our results show
ed

that the field dependent relaxation
could provide detail information of molecular motion as described by molecular
correlation time, the CSA order parameter
S
C
2
, and

the local correlation time for fast
internal motion. This measurement deconvoluted the dipolar relaxation and CSA
effects successfully.
We have conducted t
his approach on
13
C

-
R
1

on carboxyl in
amino acids
.
The

preliminary data

exhibited molecular motion
investigation with
out

symmetric CSA model
, which was in a real case
. The local fluctuation time
influenced by CSA was successfully calculated from
R
1
(

).
Our relaxation
investigation displayed additionally t
he carboxyl relaxation study on the peptide bond
of Ala
-
Gly
which
experience
d

a
symmetry
CSA
and
ha
d

manifest

rigid motion
.
Summarily, the carboxyl field dependent relaxation has

the potential to apply on
protein systems for dynamics investigation.


Acknowledgement:


Special t
hank
s to

Prof. A. G. Redfield

for kind ad
vice and discussion, especially

for providing his original design of the shuttling device.
Additionally t
hank Mr.
Fong
-
Ku Shi and Mr. Jimmy Wu (Rezwave Technology Inc.) for technical advice and
assistant. This work is supported by the High
-
Field Nuclear Magnetic Resonance
Center (HFNMRC), National Research Program for Genomic Medicine, NSC,
Taiwan
.


Shuttling Device for Field Cycling Experiment on NMR Relaxation Study


15

Referenc
e:


1.

Bloch F (1957)
Physical Review

105,

1206.

2.

Bloembergen N, Purcell EM, & Pound RV (1948)
Physical Review

73,

679.

3.

Solomon I (1955)
Physical Review

99,

559.

4.

Solomon I & Bloembergen N (1956)
The Journal of Chemical Physics

2
5,

261
-
266.

5.

Wittebort RJ & Szabo A (1978)
The Journal of Chemical Physics

69,

1722
-
1736.

6.

Overhauser AW (1953)
Physical Review

92,

411.

7.

Overhauser AW (1953)
Physical Review

89,

689.

8.

Brougham DF, Horsewill AJ, & Trommsdorff HP (1999)
Chemical Phy
sics

243,

189
-
199.

9.

Kimmich R, Nusser W, & Gneiting T (1990)
Colloid Surface

45,

283
-
302.

10.

Roberts MF, Cui Q, Turner CJ, Case DA, & Redfield AG (2004)
Biochemistry

43,

3637
-
3650.

11.

Caldeira AO & Leggett AJ (1983)
Physica A: Statistical and Theoretic
al
Physics

121,

587
-
616.

12.

Woessner DE (1962)
The Journal of Chemical Physics

37,

647
-
654.

13.

Lipari G & Szabo A (1981)
Biophys J

33,

A307
-
A307.

14.

Lipari G & Szabo A (1982)
J Am Chem Soc

104,

4546
-
4559.

15.

Lipari G & Szabo A (1982)
J Am Chem Soc

104,

4559
-
4570.

16.

Schurr JM, Babcock HP, & Fujimoto BS (1994)
Journal of Magnetic
Resonance, Series B

105,

211
-
224.

17.

Fushman D, Tjandra N, & Cowburn D (1998)
J Am Chem Soc

120,

10947
-
10952.

18.

Alfred GR (2003)
Magnetic Resonance in Chemistry

41,

753
-
768.

19.

Kimmich R, Seitter R
-
O, Beginn U, Mueller M, & Fatkullin N (1999)
Chem
Phys Lett

307,

147
-
152.

20.

Kimmich R & Anoardo E (2004)
Prog Nucl Mag Res Sp

44,

257
-
320.