OCT imaging of rod mitochondrial respiration in vivo

Abstract

There remains a need for high spatial resolution imaging indices of mitochondrial respiration in the outer retina that probe normal physiology and measure pathogenic and reversible conditions underlying loss of vision. Mitochondria are involved in a critical, but somewhat underappreciated, support system that maintains the health of the outer retina involving stimulus-evoked changes in subretinal space hydration. The subretinal space hydration light-dark response is important because it controls the distribution of vision-critical interphotoreceptor matrix components, including anti-oxidants, pro-survival factors, ions, and metabolites. The underlying signaling pathway controlling subretinal space water management has been worked out over the past 30 years and involves cGMP/mitochondria respiration/pH/RPE water efflux. This signaling pathway has also been shown to be modified by disease-generating conditions, such as hypoxia or oxidative stress. Here, we review recent advances in MRI and commercially available OCT technologies that can measure stimulus-evoked changes in subretinal space water content based on changes in the external limiting membrane-retinal pigment epithelium region. Each step within the above signaling pathway can also be interrogated with FDA-approved pharmaceuticals. A highlight of these studies is the demonstration of first-in-kind in vivo imaging of mitochondria respiration of any cell in the body. Future examinations of subretinal space hydration are expected to be useful for diagnosing threats to sight in aging and disease, and improving the success rate when translating treatments from bench-to-bedside.

Keywords: Diffusion MRI; mitochondria; optical coherence tomography; photoreceptors.

Figure 1.

Working model outlining how the SRS volume changes hydration with dark and light conditions. (a) In the dark, cyclic nucleotide–gated ion channels (left most cartoon) are maintained in an open position via availability of cGMP (up arrow), causing increased ion pumping and associated increase (up arrow) in mitochondrial (second cartoon) respiration. In turn, rods produce more waste water (# of droplets) and CO₂ that together lower (down arrow) the pH of the SRS. Water + CO₂ make an acidified environment that upregulates water efflux co-transporters (faucet cartoon) on RPE causing a reduction in SRS volume (indicated by smaller font). A smaller SRS volume is measured by a thinner ELM-RPE thickness. (b) In the light, cGMP is hydrolyzed reducing its concentration and causing cyclic nucleotide-gated channels to close, reversing the events in (a) and producing a larger SRS volume than in the dark with a thicker ELM-RPE region. Each step in this signaling pathway has been experimentally demonstrated (see text for details). (A color version of this figure is available in the online journal.)

Figure 2.

Diffusion MRI measures dark–light differences in the SRS volume in vivo. (a) Typical MRI image of a mouse eye (left-most image) indicating that the shape of isotropic water movement (as measured by diffusion MRI) changes from circular/spherical in the absent of barriers (e.g. in the vitreous) to more constrained/oblong in the extracellular fluid surrounding rod photoreceptors. Thus, a decrease in SRS volume is predicted to decrease isotropic water movement as water encounters more barriers. (b) Summary of retinal water mobility (i.e. apparent diffusion coefficient [ADC]) profiles as a function of retinal depth during dark (black, dark lightbulb) and 20 min of ∼500 lux light (pink, yellow light bulb) for wildtype C57BL/6J mice. A representative OCT image is presented at the top of the graph (after aligning the vitreous-retina and retina-choroid boundaries) to provide a guide for assigning a particular ADC value to a particular retinal laminae; indicated are the outer nuclear layer (ONL), external limiting membrane (ELM), inner segments (IS), outer segments (OS), and retinal pigmented epithelium (RPE). Profiles were spatially normalized to whole retinal thickness for each mouse (0% = vitreous/retina border; 100% = retina/choroid border). Data are means ± SEM. Black horizontal line = region with significant differences (P < 0.05) between profiles. This graph highlights (i) that in the light water mobility is greatest in the SRS relative to other parts of the retina, and (ii) that a dark-evoked decrease in ADC is localized to the SRS region (e.g. 92–100% retinal depth), as predicted by the signaling pathway in Figure 1 and microelectrode studies in the literature.

Figure 3.

Diffusion MRI data supporting signaling pathway factors in Figure 1 that contribute to the SRS volume changes in light and dark. Summary of paired data (filled = dark, open = light) of wildtype (WT) mice) showing a nice reduction in ADC in the dark, mice without the phototransduction protein transducing (GNAT1^−/− mice) do not show a light–dark difference, acidified SRS by acetazolamide (AZM) do not show a light–dark difference, and vehicle or an anti-oxidant (α-lipoic acid, LPA) treated two-month diabetic mice (STZ, STZ + LPA, respectively) show the presence of oxidative stress (which can acidify neurons). Red horizontal line, P < 0.05.

Figure 4.

Modifiers of the ELM-RPE thickness. (a) Light vs. dark: Zoomed-in region of the ELM-RPE in a representative mouse that was exposed to either 5 h of lab light or was overnight dark adapted; brackets provide a visual guide to highlight dark-evoked shrinkage of the ELM-RPE thickness. Quantitation of this light–dark change is provided in the first two bar graphs in (b) (also in Figure 5). (b) Pharmacology: Bar graph summary comparing dark (D, black bar) and light (L, white bar) adapted ELM-RPE thicknesses to a light-adapted mouse given the phosphodiesterase 6 inhibitor sildenafil (green bar) 1 h prior to examination. Data are means ± 95% confidence intervals; data for each mouse also shown as individual points. Black horizontal lines = significant differences between groups (P < 0.05). (A color version of this figure is available in the online journal.)

Figure 5.

Evidence that mitochondria drives changes in ELM-RPE thickness. (a) Light vs. dark: Summary of outer retina thickness, measured from OCT images in light (L) and dark (D) for mice with relatively inefficient mitochondria respiration (C57BL/6J [B6]) and with relatively more efficient mitochondrial respiration (129S6/SvEvTac [S6]). This means that in the light, S6 mice have a lower basal level of mitochondria activity than B6 mice resulting in lower waste water production and thus shorter ELM-RPE thickness and smaller light–dark difference. (b) Pharmacology: Paired differences (before-after DNP) are presented to account for changes within mice. Individual data points (= number of eyes examined; one eye per mouse) represent the replicate average for each mouse to illustrate animal-to-animal variation. Specific stimulation of mitochondria with a protonophore again resulted smaller ELM-RPE in light-adapted S6 mice compared to that in B6 mice, in agreement with the light-differences in (a). In all graphs, horizontal range bar indicates the region with significant differences (P < 0.05); error bars represent 95% confidence intervals.