Conformational change of a biomolecule—a change in its structure—is the ultimate basis for signaling in biology. A convenient and direct way of detecting structural changes in protein targets, and one that is scalable and requires small amounts of sample, would be a big help in definitively identifying compounds that act as agonists or antagonists, and especially useful for discovering allosteric ones. Conventional techniques such as NMR and x-ray crystallography provide atomic-level resolution but are not particularly suitable for drug discovery because of their intensive labor, sample, and time requirements.

This article reviews second-harmonic generation (SHG), a new screening platform for detecting conformational change, that is fast, scalable, and requires tiny amounts of sample. It can be used in either a label or label-free format. It is also suitable for fragment screening up to millimolar concentrations. It provides a direct answer to the question of whether a hit is an antagonist or an agonist, which is often difficult and time-intensive to ascertain with cell-based assays, particularly with fragment screens. In this methodology, protein samples must be purified, and in the case of membrane proteins reconstituted in one of several ways. Because only a picomole of target is needed per well, however, a little protein goes a long way.

As a detection modality, SHG works in a different way than fluorescence. It is a surface-sensitive technique because it only detects molecules right at an interface. Conformational change is detected when biomolecules are immobilized to the surface. Rather than absorbing and re-emitting light as in fluorescence, light is reflected in a nonlinear way off of a surface. The interface (surface plus biomolecules) converts a small amount of red pulsed laser light into second harmonic light at half the wavelength of the incident light.

For example, 800 nm light is converted to 400 nm, a wide spectral difference. Therefore, the shift is in the opposite direction than what is observed in fluorescence. In practice, this makes separating the signal from the incident light easy to do.

For many years, SHG has been used in the physical sciences to detect molecules and their orientation on surfaces (Figure 1). Only some molecules are SH-active, just as only some are fluorescent. As protein molecules are not generally SH-active, Biodesy has developed both label and label-free formats for detecting targets and their conformational changes on a surface.

For example, in label format purified, labeled protein is immobilized to a surface. A single type of label is used and it does not need to be applied site specifically. Incident light from a laser is directed to the surface. The surface and labels convert a small fraction of the incident light to second harmonic light, the baseline signal.

When the protein changes its structure upon binding ligand or drugs, this changes the amount of second harmonic light produced. In effect, the technique provides an instantaneous and direct readout of conformational change and is thus a molecular-level, functional screen. It is complementary to cell-based assays and can be used in hit generation or the hit-to-lead phase. Only a picomole of protein is needed per assay well. The signal-to-noise ratios are at least 20:1 for a four-second average with z'-factors of about 0.8. The response is in real time. With replicates, one can also measure IC50.