I’m always happy when a new analytical technique is worked out, especially one that’s applicable to biological binding assays, doesn’t require labeling of the species involved, and is orthogonal to the existing methods. We need all the reality checks we can get, and this one (reported by a large multinational team led out of Oxford) uses light scattering to get more mass resolution than I would have thought possible.
Light scattering by small particles depends on their refractive index and the particle volume. Proteins have basically the same refractive index (the indices of the individual amino acids don’t vary that much, and it’s all averaged out). Likewise, the volume occupied by each amino acid is very similar as well, and putting these together means that if you do very accurate light scattering measurements (using interferometry) the signal you get should be proportional to the number of amino acid residues or the total mass. This effect has been predicted, but recent improvements in instrumentation have made it possible to actually observe this and use it in real applications. The group gives the technique the deliberately memorable name of iSCAMS (interferometric scattering mass spectrometry), and we’ll have to get used to the idea of being able to do “mass spec” with a microscope and visible light.
Looking over a range of proteins (both monomers and larger oligomers), the team found a very good correlation (within 1.8%, plus or minus half a per cent) between the interferomic signature and protein mass, one that (importantly) was not influenced by the three-dimensional shape of the protein itself. The correlation also held for lipid nanodisc species, which is a good sign. And since I was writing just the other day about glycosylated proteins, I note that they also observed various states of a highly glycosylated (up to half the molecular weight) HIV envelope complex, and could spot the differences there as well. So it look like the effect holds for a number of different biomolecule types, although I expect that there will be refinements to the scattering models for these as the technique is developed.
This looks like it could be a very good technique for studying protein-protein interactions in vitro, and it has the potential advantage of spatial resolution as well. I’m not sure if you could ever get something like this to work in a living cell (the signal/noise would be a huge problem), but there’s a lot of biomolecular behavior that you could get new details on short of that. There are a lot of variations to try, such as different surfaces, and immobilized binding partners. This could be a good complement to techniques like SPR, only this time at a single-protein level. It’ll be interesting to see if it can get to the point of delivering on- and off-rates, which is one of SPR’s great advantages. As it stands now, the resolution of iSCAMS doesn’t seem good enough to detect small-molecule binding to individual proteins, but it could certainly pick up small molecule effects on protein complexes, and presumably effects on protein-DNA, protein-RNA, and protein-polysaccharide interactions as well. That should keep everyone occupied!