Chemistry Reference
In-Depth Information
Fig. 1 Example of a combinatorial isotope labeling scheme that identifies adjacent amino acids in
the sequence [
106
]. In this simplified example, backbone structures for the same segment with two
differently labeled samples are shown on the
left
in a and b, with the corresponding
1
H-
15
N2D
projection of the HNCO spectrum (
smaller blue peaks
) superimposed on the
1
H-
15
N HSQC
spectrum from the same sample (
red peaks
). Peaks are labeled with residue numbers for this
tetrapeptide segment. In sample a, two of the amino acids with the same side chain (residues
2 and 4,
orange side chains
) are labeled with
15
N (highlighted in
red
in the backbone structure), and
one of the amino acids (
yellow
) is labeled with
13
C. The appearance of a peak in the HNCO
projection for one of the
15
N-labeled amino acids allows the identification of the preceding amino
acid type for that residue (i.e., the sequence is
yellow-orange
). In sample b residues 2 and 4 are now
labeled with carbonyl
13
C(
blue
), with
15
N-labeling of 'green' amino acids. The single peak that
would be seen in the HSQC spectrum for this sample would also be observed in the HNCO
projection, allowing the sequence to be extended to
yellow-orange-green
in this example. The
number of samples and specific amino acids that would need to be labeled for a full assignment will
depend on the method used to find the combination that gives the maximum number of inter-residue
correlations with a minimum number of samples [
102
,
107
,
108
]
Meanwhile, a combinatorial optimization method has subsequently been developed
that allows a wider range of auxiliary conditions to be factored into the design of
protein sequence-specific labeling protocols [
109
]. Overall, these tools should help
to make cell-free expression and selective labeling increasingly accessible to a
wider range of laboratories in membrane protein structural biology.
3 Membrane Mimetic Systems for Solution NMR
3.1 Micelles
One of the challenges for solution NMR of membrane proteins is the identification
of conditions that can mimic the native lipid bilayer environment while maintaining
the sample in a stable, folded state with a total complex size of ~100 kDa or less
[
110
-
112
]. This has most commonly been achieved through the use of detergents
that form smaller micelles (~10-30 kDa) with a relatively high critical micelle
concentration (cmc). However, the ability of some detergents to maintain mem-
brane proteins in a water-soluble state sometimes works against structural studies,
essentially solubilizing the protein so well that the native contacts are disrupted by
