Creating super-scattering Raman-active genetically encoded proteins

Lead Research Organisation: Cardiff University
Department Name: School of Biosciences


The ability to study complex molecular interactions in situ in real time using light microscopy and spectroscopy has revolutionised our understanding across the physical and life sciences. In the context of bioimaging, light microscopy is still the only practical means of obtaining high spatial and temporal resolution within living cells and tissues. Fluorescence microscopy is a widely utilised method, whereby fluorescent 'tags' are attached to biomolecules of interest and provide high contrast and specificity. However, fluorescent dyes cannot easily be attached to a specific target molecule in a non-invasive manner inside a living cell. This limitation was largely overcome by the discovery of fluorescent proteins which can be genetically fused to a specific target protein. This discovery has revolutionised bioimaging and was recognised by the Nobel prize in Chemistry in 2008.

Despite being the method of choice in virtually any cell imaging application, fluorescence microscopy has some major drawbacks. Firstly, all organic fluorophores are prone to photobleaching, an irreversible photo-chemical degradation process quenching the emitted fluorescence intensity. Photobleaching severely limits observations as a function of time and is often accompanied by toxic effects damaging living cells. Moreover, the emission spectrum of organic fluorophores is quite broad. This generates a "colour barrier" that limits the number of distinguishable fluorescent probes, and corresponding biomolecules, typically to about five. Yet, the ability to directly visualize many distinct molecular species inside cells is increasingly essential for understanding complex systems and processes. For example, signalling pathways which are dysregulated in many cancer types typically involve >50 protein components, and these are impossible to track simultaneously with current techniques.

Complementary to fluorescence, vibrational microscopy based on Raman scattering offers photostability and spectrally narrow bands. Raman scattering can be regarded as an inelastic collision of light with a vibrating molecule. The energy difference between the incident and scattered photon equates to the vibrational energy gained or lost by the molecule. A major drawback, however, is that photon fluxes in detection are extremely low. As a result, conventional Raman micro-spectroscopy requires long integration times and/or large incident powers, often incompatible with live cell imaging.

To overcome these limitations, in this project, we will create new molecules which will Raman scatter light extremely strongly and will exhibit sharp Raman resonances that will enable unprecedented multi-colour imaging. We will achieve this by engineering proteins to contain genetically encoded non-natural Raman-active chemical bonds coupled to chromophores. In this way, there will be a huge increase of Raman scattered light when the frequency of the incident light is close to that of the electronic absorption in the chromophore. Importantly, via genetic encoding, our new molecules can be easily fused to natural proteins, thus providing a new class of photostable tags for bioimaging in living cells. Moreover, we will exploit the coherent nonlinear enhancement that is achieved when two incident laser fields are used to drive a molecular vibration via their beat note. As a result, all vibrational modes of a given type within the focal volume are coherently driven to oscillate in sync, and the Raman scattered light constructively interferes. Such coherent Raman scattering uses near-IR light suitable for deep penetration in living specimens and benefits from an intrinsic 3D optical sectioning.

This development has the potential to transform the field of live cell microscopy by providing probes and imaging methods with superior photo-stability, multi-colour capabilities, penetration depth, and highly targeted molecular specificity via genetic encoding.


10 25 50