4 Notes
1. Fragment screening by NMR typically requires the preparation
of large numbers of samples, often 50–100 or more at a time.
While these can be prepared manually, automated (such as a
liquid handling robot) or semiautomated approaches (such as a
multichannel pipette) significantly reduce the manual workload
required. A number of experimental tips are useful to consider
when using multichannel pipettes to prepare samples for NMR.
(a) It is advised to prepare the samples in deep-well 96-well
plates to allow thorough mixing and to transfer the
prepared samples to NMR tubes positioned in empty
96 position pipette tip racks; these racks hold the sample
tubes securely and in the correct format for loading with a
multichannel pipette (Fig. 5a). Sample racks can also be
printed readily using a 3D printer.
(b) Gel-loading tips are useful when loading NMR tubes with
a multichannel pipette to avoid the formation of air bubbles close to the top of the sample tube.
(c) Gentle spinning of the sample tubes in a hand centrifuge
improves the quality of the shimming by ensuring
uniform sample depth and removing bubbles; labeling of
the rotor buckets allows ready sample tracking (Fig. 5b). A
hand centrifuge is preferred to a benchtop centrifuge in
order to reduce the risk of breaking the NMR tubes.
2. Commonly used acquisition parameters and details of
LO-NMR experiments:
(a) Excitation sculpting [39] has proven to be a robust
method of solvent suppression suitable for use under
automation.
(b) A 2 s relaxation delay in the 1D
1 H NMR experiment is
sufficient to allow approximate quantitation of the ligand
concentration with respect to an internal standard.
(c) In our experience, a T2 relaxation filter using a CPMG
train is reliable and generic and is preferred over a T1ρ
filter using a spinlock.
(d) Mixing delays of 2.2 s (STD), 2 s (water-LOGSY), and
400 ms (T2 relaxation filter) are suitable for most
LO-NMR FBS situations. Longer CPMG filters are
advised only when fragments are binding with low affinity
(K D ! 1 mM).
3. Approximate compound concentration can be readily determined from a 1D
1
H NMR experiment acquired with 50 μM
DSS (4,4-dimethyl-4-silapentane-1-sulfonic acid) present in
266
Ben J. Davis
1. Fragment screening by NMR typically requires the preparation
of large numbers of samples, often 50–100 or more at a time.
While these can be prepared manually, automated (such as a
liquid handling robot) or semiautomated approaches (such as a
multichannel pipette) significantly reduce the manual workload
required. A number of experimental tips are useful to consider
when using multichannel pipettes to prepare samples for NMR.
(a) It is advised to prepare the samples in deep-well 96-well
plates to allow thorough mixing and to transfer the
prepared samples to NMR tubes positioned in empty
96 position pipette tip racks; these racks hold the sample
tubes securely and in the correct format for loading with a
multichannel pipette (Fig. 5a). Sample racks can also be
printed readily using a 3D printer.
(b) Gel-loading tips are useful when loading NMR tubes with
a multichannel pipette to avoid the formation of air bubbles close to the top of the sample tube.
(c) Gentle spinning of the sample tubes in a hand centrifuge
improves the quality of the shimming by ensuring
uniform sample depth and removing bubbles; labeling of
the rotor buckets allows ready sample tracking (Fig. 5b). A
hand centrifuge is preferred to a benchtop centrifuge in
order to reduce the risk of breaking the NMR tubes.
2. Commonly used acquisition parameters and details of
LO-NMR experiments:
(a) Excitation sculpting [39] has proven to be a robust
method of solvent suppression suitable for use under
automation.
(b) A 2 s relaxation delay in the 1D
1 H NMR experiment is
sufficient to allow approximate quantitation of the ligand
concentration with respect to an internal standard.
(c) In our experience, a T2 relaxation filter using a CPMG
train is reliable and generic and is preferred over a T1ρ
filter using a spinlock.
(d) Mixing delays of 2.2 s (STD), 2 s (water-LOGSY), and
400 ms (T2 relaxation filter) are suitable for most
LO-NMR FBS situations. Longer CPMG filters are
advised only when fragments are binding with low affinity
(K D ! 1 mM).
3. Approximate compound concentration can be readily determined from a 1D
1
H NMR experiment acquired with 50 μM
DSS (4,4-dimethyl-4-silapentane-1-sulfonic acid) present in
266
Ben J. Davis