Development of multiphoton label-free super-resolution microscopy techniques for biomedical imaging

Abstract

Biology exists across a range of spatial scales from whole organisms (up to ~25 m) down to individual molecules (<1 nm). Optical microscopy has allowed the study of many biological processes at scales that are not resolvable by the human eye (~< 100 μm). However, due to diffraction, the resolution of an optical microscope at visible wavelengths is ~200 nm, prohibiting the study of biological processes below this limit. A group of techniques developed to overcome this limit called super-resolution techniques, have improved the resolution to ~1-10 nm. However, these super-resolution techniques require fluorescent labelling of biological samples limiting their applicability especially to live cell or long term imaging. Multiphoton label-free imaging techniques do not require labelling of the sample but are still limited by diffraction. In this work two methods of label-free super-resolution with multiphoton imaging are established and explored. A physical method employing the photonic nanojet (PNJ) phenomenon is used to improve the resolution of second harmonic generation (SHG) microscopy. Optimal imaging parameters are established alongside developments in sample preparation and imaging workflow to maximise imaging performance. The limit of resolution is established and found to be 125 nm a 2.7 x improvement over the diffraction limit for SHG excited at 800 nm. The method is used to detect ultrastructural changes in fibrillar collagen in lung disease unobservable under diffraction-limited imaging. The method is used further to characterise other extracellular matrix proteins such as elastin, revealing new biophysical insight. A computational method based on signal fluctuations is also developed for application to cellular autofluorescence signals. Resolution improvement was demonstrated for widefield fluorescence and point-scanning imaging systems. A 3D model sample was refined and lightsheet imaging employed to facilitate multiphoton imaging. The method is used to image mitochondria using the autofluorescence from the metabolic co-enzymes nicotinamide adenine dinucleotide (NADH) and flavin adenine dinucleotide (FAD). This thesis successfully implements label-free super-resolution using multiple methods and imaging modalities. These tools have led to new biological understanding; moreover, they are simple to implement allowing for widespread use and application to disease research.

More information

Identifiers

ORCID for Sumeet Mahajan: orcid.org/0000-0001-8923-6666

Catalogue record

Export record