Binding: the waterLOGSY experiment

NMR is one of the best methods for studying molecular interactions. To identify small molecules that bind to macromolecules many different experiments can be used, one of which is the waterLOGSY experiment. This is a 1D, ¹H-detected experiment that can be run in five minutes and is best suited for weakly interacting complexes (mM K_Ds), a binding regime that is difficult to probe by other techniques. To demonstrate how it works I will show spectra of a mixture of L-tryptophan and D-sucrose in the presence and absence of bovine serum albumin.

Bovine serum albumin (BSA) binds small aromatic molecules, like tryptophan, but does not bind sugars, such as sucrose, so this system makes a good test case for the waterLOGSY experiment. The figure below shows normal 1D ¹H spectra with presaturation of the water recorded using a mixture of tryptophan and sucrose. The aromatic tryptophan resonances lie between 7.0 and 8.0 ppm, the sucrose anomeric resonance appears near 5.4 ppm and the other sucrose signals between 3.0 and 4.5 ppm. The upper, red spectrum was recorded with BSA, while the lower, blue one was recorded without BSA. In the upper, red spectrum the BSA resonances can be seen as low broad peaks on the baseline from 0.0 to 9.0 ppm. Note that in this spectrum the aromatic tryptophan resonances are less intense and broader than the sucrose peaks, whereas in the spectrum without BSA (lower, blue) the tryptophan and sucrose peak intensities are similar. This is an indication that tryptophan binds BSA while sucrose does not.

The next figure shows waterLOGSY spectra recorded on the two samples. Again, the upper red spectrum was recorded with BSA added, while the lower blue spectrum was recorded without BSA. In the lower, blue spectrum the tryptophan and sucrose peaks are all positive. In the upper, red spectrum, where BSA has been added, the aromatic tryptophan peaks are now negative while the sucrose peaks remain positive. You may also notice some negative peaks around 4.0 and 3.5 ppm, these are the aliphatic tryptophan peaks. The inversion of the tryptophan peaks indicates that the tryptophan is binding to the BSA. Since the sucrose peaks remain positive we can conclude that sucrose does not bind to BSA.

The waterLOGSY experiment is essentially a 1D selective NOE experiment that identifies NOEs to water. It relies upon the change in sign of the NOE with correlation time. If the water shows an NOE to a small molecule with a short correlation time the peaks will be positive. If the water shows an NOE to a large molecule with a long correlation time then the peaks will be negative. When the tryptophan binds to the BSA it acts like a large molecule and so gives negative peaks. The sucrose doesn't bind to the BSA so it continues to act like a small molecule and gives positive peaks.

A very long NOE mixing time is used to exploit all possible pathways for magnetisation transfer from the bulk water. Magnetisation can be transferred to the ligand from the bulk water, to the ligand from water bound to the surface of the macromolecule that is in exchange with the bulk water, and to the ligand by spin diffusion from the macromolecule which has been in contact with the bulk water or has surface bound waters in exchange with the solvent.

The waterLOGSY experiment is very useful for screening mixtures of compounds to identify molecules that bind to a specific macromolecular target. Even where the resonances of molecules in the mixture overlap it is still possible to identify binding.

Acknowledgements
Implementing the waterLOGSY experiment was suggested by Julius Bogomolovas in the Ju Chen Lab. Julius also prepared and provided the samples.