Akaash Kumar
Shedding light on the molecular interactomes of fast, complex biological processes using multispectral imaging with uncompromised spatiotemporal resolution.
Keywords: Multispectral, Microscopy, Live-cell imaging, Cell biology, Spectral unmixing.
Fluorescence microscopy is a powerful tool for studying biological processes. In this technique, biological structures are labelled with fluorescent molecules (fluorophores) which can be imaged using a fluorescence microscope- providing background-free, high contrast images of the labelled structures. Importantly, this technique is compatible with visualising dynamics within live cells: several distinct structures can be labelled with different coloured fluorophores to determine where components are with respect to each other over time. These advantages make fluorescence microscopy particularly useful for establishing the spatiotemporal relationships of structures during a normal or perturbed biological process (for example cell division vs cancer).
However, there are two main, longstanding issues with this approach: 1) different coloured fluorophores are imaged sequentially leading to slow data acquisition and missed information and 2) due to a phenomenon known as spectral overlap only 2 or 3 (of the many dozens of proteins involved in a process) can be imaged within the specimen. These two limitations mean that it is not possible to capture the intricacies and molecular details of many fast, complex cellular processes.
To this end, my PhD thesis tackled these two longstanding issues through a combination of software, hardware and labelling strategies. First, I designed and developed multispectral imaging hardware. This can be attached to any camera based microscope to allow it to image up to eight fluorophores simultaneously. Secondly, I developed an algorithm which is used to process the acquired data to resolve the labelled biological structures. The combination of these technologies provides any fluorescence microscope with the ability to image up to eight labelled structures with no loss in temporal resolution, for the first time, resulting in a more informative view of intracellular process. Furthermore, in collaboration with David Baker’s laboratory at the University of Washington, I demonstrated the use of de novo designed protein binders for the imaging of membrane receptors. By combining these binders with the multispectral imaging technology, I demonstrated that we can observe the simultaneous trafficking of several different receptors paving the way for the molecular dissection of these trafficking pathways.
The impact of my work is recognised through several awards, a patent, and submitted publications, including a first-author manuscript currently under review at Nature Biotechnology. As an interdisciplinary scientist, I am now using the technology I have developed to shed light on complex cellular processes such as endosomal sorting. This process is crucial to understand as defects in this process result in various neurodegenerative diseases, such as Alzheimer’s disease. The next stage of my research will apply the technology I have developed to determine the molecular mechanisms underpinning T-cell immunotherapies- I will use this insight to develop more efficacious therapeutics. By overcoming longstanding challenges in cellular imaging, my research opens new avenues for exploring and understanding the intricate dynamics of cellular life, contributing meaningfully to the advancement of my field.
How the work contributes to bridge fields of biology, physics and/or mathematics:
My research is highly interdisciplinary. For instance, the development of the multispectral imaging hardware required principles of physics (optics) and engineering where I optically simulated the light paths to maximise the imaging performance (using Zemax OpticStudio), mechanically designed the hardware (using SolidWorks) and built/aligned the system. The processing algorithm required coding (Python) employing mathematical principles related to image formation and deconvolution operations. Finally, I successfully integrated the hardware and algorithm to study complex intracellular processes (endosomal sorting) requiring advanced molecular biology techniques, including the development of novel polycistronic plasmids and computationally-designed receptor binders, as well as image analysis pipelines.
MRC Laboratory of Molecular Biology, University of Cambridge
James Manton Lab
