Chemistry Reference
In-Depth Information
aqueous systems (e.g., ~0.2 mM for KcsA), this is offset by the increase in cryogenic
probe sensitivity (via an increase in probe quality factor Q) that comes from low
sample conductance [
230
]. In general, these results provide an encouraging indication
of future benefits that reverse micelles may bring, particularly for the elimination of
the requirement for deuterium isotope labeling for some large complexes (described in
Sect.
4.1.1
).
3.4 Approaches for Sample Screening
A consistent theme that has emerged in membrane protein structural biology is that
the selection of an appropriate membrane-mimetic solvent is a largely empirical
process. This was well illustrated for the pSRII GPCR, which required over 20
different sets of conditions to be systematically explored before C
7
-DHPC was
identified as the most appropriate solvent [
40
]. In this case screening was greatly
facilitated by the ability to use the unique absorbance properties of folded pSRII to
identify promising conditions rapidly. For those proteins with folding states that
cannot be so conveniently monitored it is still possible to screen a wide range of
small samples by performing stability measurements. This involves incubation of
each sample under conditions that would simulate those of a typical 3D NMR
experiment, followed by SDS-PAGE of the soluble fraction separated from
aggregated precipitates [
37
]. Identification of factors that preserve membrane
protein solubility can be used to narrow the range of conditions to be examined
by more informative but time- and sample-intensive approaches. These include
experiments that estimate size (e.g., size exclusion chromatography with or without
dynamic light scattering) [
40
,
138
], translational diffusion [
146
,
231
], effective
rotational correlation time measurements [
70
,
138
,
207
,
232
,
233
], small angle
X-ray scattering [
37
], analytical ultracentrifugation [
234
], and spectral quality
(e.g., 1D
1
Hor2D
1
H-
15
N HSQC (TROSY) spectra, [
37
,
149
,
159
]).
While cryoprobe technology has reduced the sample concentrations required to
evaluate the quality of reconstituted samples [
235
], the need for sample volumes
that exceed ~275
L limits the number of conditions that can be tested from a single
protein preparation. In addition, the detergents themselves can constitute a signifi-
cant fraction of the expense, particularly since a large excess is required to ensure a
1:1 micelle:protein ratio [
111
,
236
]. However, with the advent of smaller-diameter
NMR probes it is now possible to reduce the amount of sample used in each
experiment, since the increase in the probe coil length-to-diameter ratio confers
an increase in probe mass sensitivity [
237
-
239
]. Although acquisition times for
biomolecules are lengthened due to the reduction in total sample quantities, it is
nonetheless still possible to use sample concentrations similar to those used in
conventional probes to obtain 2D spectra within a few hours [
240
-
242
]. For
example, a 1-mm Bruker TXI microcoil probe with a ~10
m
L sample volume was
used to evaluate refolding efficiencies of the
E. coli
outer membrane protein OmpX
[
151
]. In this case it took ~9 h to record a
1
H-
15
N TROSY with ~1.2 mM detergent
m
