Broadband coherent raman cell imaging for in situ protein co-localisation

The use of light microscopy is particularly attractive for the study of the ways in which proteins interact in biological cells, and control cell function, since they permit the study of such biological processes within the actual environment of the cell. The ability to tag proteins with fluorescent molecules enables the determination, with microscopy, of the locations of different proteins in the cell. The spatial 'co-localisation' of these proteins can then provide valuable information about how they interact. Although this approach is enormously successful, the use of fluorescence does present a number of inherent problems. We aim to investigate another spectroscopic imaging technique, coherent anti-Stokes Raman scattering (CARS) imaging, as an alternative to fluorescence microscopy which offers the potential to overcome these problems. CARS is a non-linear optical technique which allows the measurement of strong vibrational signals, of similar strength to fluorescence. In particular, we will implement Broadband-CARS, in which a CARS spectrum can be measured at each scan point, as the sample of interest is scanned through a tightly focussed laser spot. We will consider the use of conventional small fluorescent molecules as CARS labels, each with a well defined CARS spectrum. We will also apply numerical algorithms (principle components analysis) to separate CARS spectra into the constituent spectra from different labels. This will allow the simultaneous determination at each scan point of the local concentration of a number of labels, for multiplexed protein co-localisation assays. We will perform preliminary measurements on fixed cells and tissue sections, to assess the capabilities of the technique for further application with our biological collaborators and for further development of the technique, towards in vivo imaging.

Optical imaging techniques are particularly attractive for the investigation of protein interactions and dynamics in a cellular context, because they are minimally invasive and can be applied to live cells. In particular, the use of fluorescence microscopy for the determination of protein co-localisation is a now well-established and powerful experimental tool. However, the use of fluorescence presents a number of inherent problems: fluorescence photo-bleaching; sample autofluorescence; cross-talk; bleed-through; and the need to excite different fluorophores with different wavelengths. We aim to investigate coherent anti-Stokes Raman scattering (CARS) imaging as an alternative technique to fluorescence microscopy, offering low photo-bleaching and avoidance of sample autofluorescence with all the advantages of optical imaging retained. CARS is a non-linear optical technique which allows the measurement of strong vibrational signals, of similar strength to fluorescence. In particular, we will implement Broadband-CARS, in which the probe Stokes beam is a supercontinuum excited from a photonic crystal fibre, such that CARS spectra can then be measured at each scan point to enable multiplexed measurement. We will consider the use of conventional small molecule fluorescent probes as CARS labels, each with a well defined CARS spectrum. This, in combination with principle components analysis (PCA), will allow the simultaneous quantitative determination of the local concentration of a number of labels, for protein co-localisation assays. We will perform preliminary measurements on fixed cells and tissue sections, to assess the capabilities of the technique for further application with our biological collaborators and for further development of the technique, towards in vivo imaging.