Biomedical Engineering Reference
In-Depth Information
2
H,
15
N,
13
C-labelled protein. Analogously to deuteration, which removes the
unwanted DD interactions from remote sites to improve the resonances in
proton detection experiments, the resonances from low-c nuclei are expected to
be narrower basically because their lower c reduces DD relaxation. This effect
is clearly seen in the difference between the transverse relaxations of H
a
and C
a
[Figure 2.1(A)]. However, the relaxation rate of low-c nuclei, of course, largely
depends on directly bonded nuclei as well as its own CSA. As exemplified in
Figure 2.1(A), in a protonated
15
N,
13
C-labelled protein, the transverse
relaxation rate of a protonated C
a
is much faster than that of a high-c
nucleus, H
N
, simply because the large DD contribution from the directly
bonded H
a
overwhelms the benefit from its own low-c ratio. In an a-helical
region of a protein, for example, the relaxation induced by the DD interaction
between C
a
and H
a
is twice as large as the sum of the DD contributions
between H
N
and remote protons. The importance of the directly bonded
partner is already seen in high-c nuclei as the relaxation of H
a
attached to
13
C
a
is faster compared to H
N
attached to
15
N
H
.
For carbonyl carbons on the other hand, the large CSA is the dominant
source for transverse relaxation. Since the relaxation efficiency through CSA
goes with the second power of the magnetic field strength, the transverse
relaxation rates of carbonyl nuclei are largely accelerated by the increase of the
magnetic field. Although the C9 relaxation rate is half that of H
N
at 11.4 T, it
becomes larger than that of amide protons at 18.8 T. Thus, carbonyl
13
C-
detection has benefits for non-deuterated proteins of small to medium sizein
relatively low-field magnets. In fact, with its simple
13
C-
13
C one-bond scalar
coupling system in uniformly labelled proteins, C9 has been the preferred
choice for
13
C-detecting experiments established so far for moderate-size
paramagnetic proteins or intrinsically disordered proteins.
31,43
As for other low-c nuclei, direct detection of
15
N, which has the lowest c
within the protein backbone, has not been extensively exploited so far. In the
past, there were few examples where one-dimensional
15
N-direct-detection
experiments have been used to obtain structure information near active sites of
paramagnetic proteins.
44-48
In addition, it was shown that the slow
15
N
relaxation results in ultra-high-resolution spectra, which is also quite attractive
for diamagnetic proteins.
40
Considering its preferable transverse relaxation
properties in high-magnetic fields and/or at high molecular weights,
15
N-
detection is still largely unexplored.
The transverse
15
N
H
relaxation of a uniformly
15
N,
13
C-labelled protein in
H
2
O is slower than C9 relaxation in low magnetic fields. In addition, the
contribution of CSA to the relaxation is much smaller for
15
N
H
than C9, thus
the transverse relaxation rate of
15
N
H
is significantly less dependent on
magnetic field strength. The transverse relaxation of
15
N
H
can further be
reduced, simply by changing the solvent to D
2
O. In that case, even without
perdeuteration, the expected linewidth of
15
N
D
in a protein with a rotational
correlation time (t
c
) as long as 100 ns is still only a little more than 10 Hz (in
11.4 T magnets). In the case of the protein GB3, linewidths of ,1Hzand
