Noninvasive multiphoton fluorescence microscopy resolves retinol and retinal condensation products in mouse eyes

Abstract

Multiphoton excitation fluorescence microscopy (MPM) can image certain molecular processes in vivo. In the eye, fluorescent retinyl esters in subcellular structures called retinosomes mediate regeneration of the visual chromophore, 11-cis-retinal, by the visual cycle. But harmful fluorescent condensation products of retinoids also occur in the retina. We report that in wild-type mice, excitation with a wavelength of ∼730 nm identified retinosomes in the retinal pigment epithelium, and excitation with a wavelength of ∼910 nm revealed at least one additional retinal fluorophore. The latter fluorescence was absent in eyes of genetically modified mice lacking a functional visual cycle, but accentuated in eyes of older wild-type mice and mice with defective clearance of all-trans-retinal, an intermediate in the visual cycle. MPM, a noninvasive imaging modality that facilitates concurrent monitoring of retinosomes along with potentially harmful products in aging eyes, has the potential to detect early molecular changes due to age-related macular degeneration and other defects in retinoid metabolism.

Figure 1. Multi–photon excitation of a 6–month–old wild type (WT) mouse eye at 730 and 910 nm produced emission spectra indicating more than one fluorophore

(a) A series of TPM images of an intact mouse eye were obtained along the axis perpendicular to the RPE layer with an excitation wavelength of 730 nm. The main box reveals the enface image of RPE cells; the two cross–section images, one shown at the bottom and the other at the right edge, were assembled from a series of z-slice images. The yellow outlined rectangle represents the region from which fluorescence was collected for spectral analysis with the excitation light focused on the RPE. Scale bar, 75 μm. (b) Fluorescence emission spectra from the RPE of an intact mouse eye through the sclera (ts) and from flat–mounted (fm) mouse RPE are super–imposable. The second harmonic signal (SH) exhibits a sharp maximum at half of the 910 nm excitation wavelength. (c) A series of TPM images of an intact mouse eye obtained with an excitation wavelength of 910 nm. In the region of the blue outlined rectangle, a strong second harmonic signal from the sclera was dominant, as the curvature of the eye brought the sclera more into focus. Scale bar, 75 μm. The yellow outlined rectangle represents the region from which fluorescence was collected for spectral analysis shown in b. (d) TPM image of flat–mounted ex vivo RPE. Part of the RPE is folded over, exposing a sagittal view of retinosomes, indicated with the yellow arrow. Scale bar, 20 μm.

Figure 2. Multi–photon excitation of the RPE in an intact 6–month–old WT mouse eye at different wavelengths of excitation light

(a) Graph showing fluorescence as a function of excitation light wavelength. Fluorescence intensities were obtained as mean pixel values from the area covered by at least 20 RPE cells in focus and were normalized to the maximum value at 720 nm. Blue arrows (from left to right) indicate fluorescence in response to 730, 850 and 910 nm excitation. (b) Zoomed–in two–photon image of RPE cells obtained with 730 nm excitation. Blue arrows denote retinosomes (white) clustering along plasma membranes; edges of one RPE cell are highlighted in yellow. Dark, mostly double nuclei are indicated by pink arrows. (c) Images of RPE cells at different excitation wavelengths indicated in each panel. All images were obtained with 10 mW of laser power. Photomultiplier tube (PMT) gain and offset were kept constant for images in the top row. Then the gain was readjusted and kept constant for the images in the bottom row. Scale bar, 38 μm.

Figure 3. Visualization of retinosomes by three–photon excitation spectroscopy in the intact 7–week–old Rpe65^−/− mouse eye

(a) Graph showing fluorescence intensity as a function of excitation light wavelength; fluorescence intensity was calculated as a mean pixel value from the area covered by at least 20 RPE cells in focus during imaging and normalized to the maximum value at 720 nm excitation. (b) Emission spectra from the RPE of WT and Rpe65^−/− mouse eyes excited with laser light at 730 nm and 910 nm. (c) Logarithmic plot of fluorescence intensity as a function of excitation power shows evidence of three–photon excitation at an excitation wavelength of 910 nm. The mean pixel value was proportional to the 2.83^rd power. (d) At an excitation wavelength of 730 nm, the mean pixel value was proportional to the 2.12^th power of the laser power. In (c) and (d) data points are shown as black circles. Modeling of the pure second power dependence is shown by the blue dotted line and modeling of the third power dependence is indicated by the green dashed line. Insets depict areas containing retinosomes from which fluorescence was quantified.

Figure 4. Two–photon excitation of 6–week–old Abca4^−/−Rdh8^−/− (dko) intact mouse eye

(a) Fluorescence intensity as a function of excitation wavelength, normalized to maximum emission at 720 nm excitation. (b) Emission spectra from RPE of WT and dko mice excited with laser light at 730 nm and 910 nm. Blue arrow points to 580 nm. (C) Images of dko mouse RPE at different excitation wavelengths. All images were obtained with 10 mW of laser power; PMT gain and offset were kept constant. Scale bar 50 μm. (d) Zoomed–in TPM image of unstained dko mouse RPE. Fluorescent granules are distributed uniformly throughout cells with clearly visible boundaries. Green color was arbitrarily chosen to make the image details more visible. Scale bar 25 μm.

Figure 5. Age–dependent changes in fluorophore accumulation in mouse eyes

(a) Two–photon excitation fluorescence intensity of WT mouse retina drops with increasing wavelengths of light excitation; the decline of fluorescence emission is more rapid in younger than in older mice. Fluorescence emission values were normalized to that observed with 720 nm excitation. (b) Ratios of fluorescence excited with 910 nm light relative to fluorescence excited with 730 nm light increased with age in WT mice. Abca4^−/−Rdh8^−/− (dko) mice displayed higher fluorescence ratio than WT mice. (c) Increasing levels of all–trans–retinyl esters determined by normal phase HPLC were found in the eyes of WT mice at the ages of six weeks, three months and six to seven months. Rpe65^−/− mouse eyes exhibited more than a 15–fold increase in all–trans–retinyl ester accumulation as compared to age–matched WT mouse eyes. No all–trans–retinyl esters were detected in Lrat^−/− mouse eyes (n=5 for each group). (d) A2E was extracted from mouse eyes and quantified by reverse phase HPLC. Control WT mice were evaluated at six weeks, three months and six to seven months of age and exhibited increased A2E amounts in an age–dependent manner. A2E levels in 3–month–old Abca4^−/−Rdh8^−/− mouse eyes were markedly higher than in age–matched and older WT controls. No A2E was detected in the eyes of 3–month–old Rpe65^−/− and Lrat^−/− mice (n=5 for each group). In (c) and (d) data columns are shown with one standard deviation bars.