The retina is an essential part of the eye, as it is responsible for transforming light into electrical signals, which are processed in the brain later on. It effectively works as a photodetector of the eye. Imaging the structure and function of the living retina is crucial for the effective diagnosis and treatment of eye diseases and drug developments. Nowadays, structural information about the retina can be obtained from, e.g., OCT studies. However, functional alterations are the first signs of early pathological processes and often precede structural changes; this information is currently very challenging to obtain.
The retina has a layered structure, which is packed with various fluorophores. For example, retinal pigments epithelium (RPE) contains lipofuscin, a byproduct of the visual cycle. Lipofuscin is accumulated overage, but also with a progression of the disease. Other examples are retinol and retinyl esters (vitamin A derivatives active in a visual cycle), melanin, FAD, NADH, collagen, and elastin. Those substances can provide valuable information about the retina’s health and can be a valuable tool for discovering functional alterations during age-related macular degeneration, diabetic retinopathy, or glaucoma.
Standard ophthalmic autofluorescence imaging visualizes the distribution of retinal fluorophores, but only intensity information is available. As a result, signals from various fluorophores cannot be distinguished. Lipofuscin is a dominant fluorophore, and its strong signal needs to be discriminated from other, usually much weaker, sources. Luckily, various fluorophores differ in fluorescence properties, i.e., fluorescence lifetime and fluorescence spectrum. This adds as an additional discrimination parameter.
The eye is the window to the body, but it also has a specific transmission range. Consequently, many fluorophores with excitation spectra in the UV/blue spectral range (<420 nm) cannot be excited, and the information contained in them is unavailable. Our solution to this problem is to use a two-photon excitation. This scheme uses short (femtosecond) pulses in the near-infrared (twice the wavelength), bypassing those limitations. For example, using femtosecond pulses at 730 nm is equivalent to single-photon excitation at 365 nm, which would not be possible in the living eye. Additional advantages of this method are improved resolution, lower phototoxicity, and lower scattering.
In our research, we aim at visualizing the structure and molecular composition of ocular tissues using both spectral and temporal discrimination. For example, the image shows fluorescence lifetime (FLIM) distribution of retinal pigment epithelium cells of Abca4PV/PV mouse (a model of Stargardt disease in humans). The image reflects differences in the subcellular distribution of endogenous fluorophores in the RPE. Shorter lifetimes (blue-green color) are associated with A2E, a component of lipofuscin. Red granules (longer lifetime) can be associated with retinyl esters.
Similar discrimination can be done by looking at the fluorescence spectra. Pixels displaying spectral features of retinosomes were color-coded in red, while those with A2E-like properties were in blue. The analysis shows that retinosomes are predominantly located near RPE cell borders, and some are near cell nuclei. Such studies require specialized ultrafast light sources, scanning ophthalmoscopes, sensitive light detectors, and dedicated software for analysis.
Grazyna Palczewska, Jakub Boguslawski, Patrycjusz Stremplewski, Lukasz Kornaszewski, Jianye Zhang, Zhiqian Dong, Xiao-Xuan Liang, Enrico Gratton, Alfred Vogel, Maciej Wojtkowski, Krzysztof Palczewski, “Noninvasive two-photon optical biopsy of retinal fluorophores”, Proceedings of the National Academy of Sciences 117(36), 22532-22543 (2020).
Grazyna Palczewska, Patrycjusz Stremplewski, Susie Suh, Nathan Alexander, David Salom, Zhiqian Dong, Daniel Ruminski, Elliot H. Choi, Avery E. Sears, Timothy S. Kern, Maciej Wojtkowski, Krzysztof Palczewski, “Two-photon imaging of the mammalian retina with ultrafast pulsing laser”, Journal of Clinical Investigation Insight 3(17), e121555 (2018).
Text: Jakub Bogusławski, PhD