This protocol describes how high-resolution imaging techniques such as spectral domain optical coherence tomography and scanning laser ophthalmoscopy can be utilized in small rodents, using an ophthalmic imaging platform system, to obtain information on retinal thickness and microglial cell distribution, respectively.
Spectral domain optical coherence tomography (SD-OCT) and scanning laser ophthalmoscopy (SLO) are extensively used in experimental ophthalmology. In the present protocol, mice expressing green fluorescent protein (gfp) under the promoter of Cx3cr1 (BALB/c-Cx3cr1gfp/gfp) were used to image microglia cells in vivo in the retina. Microglia are resident macrophages of the retina and have been implicated in several retinal diseases1,2,3,4,5,6. This protocol provides a detailed approach for generation of retinal B-scans, with SD-OCT, and imaging of microglia cell distribution in Cx3cr1gfp/gfp mice with SLO in vivo, using an ophthalmic imaging platform system. The protocol can be used in several reporter mouse lines. However, there are some limitations to the protocol presented here. First, both SLO and SD-OCT, when used in the high-resolution mode, collect data with high axial resolution but the lateral resolution is lower (3.5 µm and 6 µm, respectively). Moreover, the focus and saturation level in SLO is highly dependent on parameter selection and correct alignment of the eye. Additionally, using devices designed for human patients in mice is challenging due to the higher total optical power of the mouse eye compared to the human eye; this can lead to lateral magnification inaccuracies7, which are also dependent on the magnification by the mouse lens among others. However, despite that the axial scan position is dependent upon lateral magnification, the axial SD-OCT measurements are accurate8.
In experimental ophthalmology, examination of retinal pathology is usually evaluated using histological techniques. However, histology requires animal euthanization and may cause alteration to the actual properties of the tissue. SD-OCT and SLO are routinely used in clinical ophthalmology for diagnostic purposes and for the monitoring of several retinal diseases such as diabetic macular edema9, anterior ischemic optic neuropathy10, or retinitis pigmentosa11. SD-OCT and SLO are non-invasive techniques that generate high-resolution images of the retina, which are visualized through the dilated pupil without further intervention. SD-OCT provides information of retinal structure and retinal thickness by collecting backscattering data to create cross-sectional images of the retina, while SLO collects fluorescence data to produce stereoscopic high-contrast images of the retina. Nowadays, both techniques are increasingly used in experimental ophthalmology using small rodents12,13,14,15 (or even zebrafish16,17) and can provide both qualitative and quantitative information12,17,18,19,20,21.
Accumulation of endogenous fluorophores like lipofuscins or the formation of drusen in the retina can be visualized by SLO as auto fluorescent signal. This feature makes SLO a valuable technique for diagnosis and monitoring of retinal diseases such as age-related macular degeneration or retinitis pigmentosa22,23. In experimental ophthalmology, auto fluorescence imaging (AF) can be used for the detection of specific cell types in reporter mouse lines. For example, mice heterozygous for the expression of gfp under the promoter of Cx3cr124 are advantageous for in vivo visualization of microglial cells in the normal retina and for the investigation of microglia/macrophage dynamics in retinal disease21. Microglia are the resident macrophages of the retina, which play a crucial role on tissue homeostasis and tissue repair upon injury1,25,26. Microglia activation in the retina has been reported in retinal injury, ischemia, and degeneration, suggesting a role of these cells in retinal disease2,3,4,5,6.
The aim of the present protocol is to describe a relatively simple method for retinal imaging and measurement of retinal thickness using SD-OCT, and for visualization of gfp positive microglia cells in the Cx3cr1gfp/gfp mouse retina using SLO (Heidelberg Spectralis HRA+OCT system). This protocol can be utilized for imaging and thickness measurements of healthy or diseased retinas in various mouse lines. Additionally, morphometric analyses can be performed for the identification and quantification of microglia numbers and microglia activation in the retina using SLO21. Microglia cells are associated with degenerative diseases in the central nervous system (CNS), including the retina27,28,29. Thus, by combining the two methods used in the present protocol, correlation of microglia distribution and retinal degeneration can be made, which can facilitate monitoring disease severity or the effectiveness of therapeutic approaches in vivo.
In all procedures, BALB/c adult male and female mice who express gfp under the promoter of Cx3cr1 were used24. Mice were treated according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and all procedures were approved from the Swiss government according to the Federal Swiss Regulations on Animal Welfare. Mice were anesthetized by a subcutaneous injection of medetomidine hydrochloride (0.75 mg/kg) and ketamine (45 mg/kg). Proper anesthesia was confirmed by monitoring the respiratory rate and the animal's reflex against a tail pinch. At the end of the experiments, mice were euthanized with CO2 inhalation.
NOTE: Perform each imaging session as quickly as possible (maximum 20 min), since cataract formation following anesthesia may hamper retinal visualization30.
1. System Configuration
2. Mouse Preparation
3. SD-OCT
4. Auto Fluorescence Imaging
5. Anesthesia Reversal
6. Manual Retinal Thickness Measurement from SD-OCT Images
Using the protocol presented here, SD-OCT scans and SLO images were obtained from Cx3cr1gfp/gfp mice in the same imaging session. Figure 3 includes representative SD-OCT single scans obtained with a 30 ° or a 55 ° lens (Figure 3A) and representative SLO images obtained with a 55 ° or a 102 ° lens, where gfp positive microglia cells are visualized. Higher reflectivity of the choroid is observed in the SD-OCT scans obtained with the 30 ° compared to the 55 ° lens. However, retinal architecture is clear in both the 30 ° and the 55 ° images (Figure 3, right magnified images). After manual corrections of retinal boundaries (inner limiting and base membrane), a good correlation of retinal thickness measurements was observed between the 30 o and the 55 ° lens when measuring in the same distance from the optic nerve head (Figure 3C; Pearson correlation = 0.967, p < 0.0001). Previous studies have shown that retinal thickness measurements on SD-OCT scans correlate well with ex vivo measurements on histological retinal sections15,31 and thus SD-OCT can be used as a non-invasive technique for retinal thickness measurements. Using SLO, we could identify gfp positive microglia cells as hyperreflective signal in the auto fluorescent images, however the wide field 102 ° lens was able to cover a larger fundus area compared to the 55 ° lens. SLO can also be utilized for the monitoring of retinal microglia cells distribution and activation in reporter mice with a functional fractalkine receptor (Cx3cr1gfp/+ mice) in normal or diseased retina12,21,32. The combination of SD-OCT with SLO in the same imaging session can provide information about microglia distribution and activation upon retinal degeneration and could be useful for monitoring retinal diseases in mice21.
Figure 1: Representative image of the device used in the present study. To image the right eye of the mouse, the mouse is positioned on the left part of the custom-made platform with its right orbit facing the lens and its body lying prone on the left part of the platform. The height of the platform can be adjusted by turning clockwise (to lower the platform) or counterclockwise (to lift the platform up) the "chin rest adjustment". The filter lever must be positioned to "A" to allow IR+OCT and AF imaging. With the focus knob (not visible on the figure), the experimenter can change the focus of the scanner. The micromanipulator can be used to correctly illuminate and align the retina for high quality images. The objective can be moved up, down, right, or left by the camera handle. Please click here to view a larger version of this figure.
Figure 2: Illustration of the device`s control panel. (A) On the left part of the control panel display the acquisition modes can be selected (infrared, IR; auto fluorescence, AF or indocyanine green angiography, ICGA; combined or not with optical coherence tomography, OCT). On the middle part of the panel the acquisition setting can be chosen (intensity of the IF image, section or volume scans). The field of view is determined by the loaded lens. With the sensitivity knob the brightness of the IR or AF image can be adjusted. By pressing the sensitivity knob, the software will start averaging the images obtained in IR or AF to provide a high-quality image. (B) Illustration of the acquisition pattern menu. (C) Summary of functions used in the software with the corresponding icon. Please click here to view a larger version of this figure.
Figure 3: OCT scans and auto fluorescent images of Cx3cr1gfp/gfp mouse retinas. (A) B-scans of the mouse retina obtained with a 55 ° (upper panel) or a 30 ° lens (bottom panel). Higher magnifications of the yellow boxes are presented on the right panel. (B) Images of gfp positive microglia cells in the retinas of Cx3cr1gfp/gfp mice obtained with a 55 ° and a 102 ° lens (left and right panels, respectively). (C) Retinal thickness measurements, measured on the same distance from the optic nerve head, on SD-OCT scans derived from the 30 ° and 55 ° lens. A good correlation was observed between the two lenses (Pearson correlation = 0.967). Scale bars: 200 µm, 100 µm on magnified SD-OCT images (A, right panel). Please click here to view a larger version of this figure.
The present article demonstrates a protocol for the acquisition of retinal B-scans and imaging of gfp positive microglia distribution in the mouse retina in the same imaging session. SD-OCT and SLO are increasingly used in animal models of retinal disease to provide information of retinal alterations over time10,14,17,18,21. With this protocol, Cx3cr1gfp/gfp or Cx3cr1gfp/+ mouse retina can be non-invasively imaged and information of retinal thickness can be obtained. This is critically important in retinal disease experimental models, where disease severity changes over time or when therapeutic treatments are tested. Moreover, the protocol describes how reporter mouse lines can be utilized for the in vivo imaging of fluorescent cells in the retina.
However, the protocol presented here has some limitations. First, imaging of the mouse retina in devices designed for humans may be challenging due to the higher total optical power of the mouse eye, which can lead to lateral magnification inaccuracies7. However, the axial SD-OCT measurements are accurate8. Additionally, follow-up examination in mice using the “set reference” option of the software is rather difficult due to the not fixed position of the mouse on the platform and thus longitudinal imaging of mouse retinas can be challenging. Longitudinal alignment of the eye can be achieved by positioning the optic nerve in the center of the infrared image and then acquiring the desired images. In order to correct for eye rotation, manual alignments of the SD-OCT en face images can be performed after acquisition and apply to the thickness measurements as described elsewhere33. Besides the imaging system used in the present protocol, other imaging systems have also been developed that provide a greater angle and can image a larger retinal area without the use of the contact lens. Optos for example can cover a larger retinal surface but with greater image variability compared to the device used in the present study34. Despite these limitations, the present protocol can be utilized in several experimental models of retinal disease. The information that can be obtained includes retinal thickness alterations (swelling, atrophy), detection of retinal detachment, or the presence of hyperreflective dots and fluorophore-labeled cells and the increase or decrease of their numbers.
Combination of the two methods in mice heterozygous for the expression of Cx3cr1 (Cx3cr1gfp/+), which have a functional fractalkine receptor, enables the experimenter to simultaneously detect in vivo alterations of retinal integrity in real time and possible microglia accumulation in the vicinity of retinal abnormalities (e.g., retinal swelling, retinal atrophy) during retinal disease or injury. Moreover, by applying morphometric analysis, microglia activation state can be assessed based on the soma size and complexity of microglia processes21. In retinal diseases, such as retinal vein occlusion, activation and accumulation of microglia/macrophages on retinal atrophic sites have been observed27. Moreover, in mice defective for the Nr2e3 gene (Nuclear Receptor Subfamily 2 Group E Member 3), a mouse model of the recessively inherited enhanced S-cone sensitivity syndrome (ESCS), the outer nuclear layer folding (rosettes) can be observed at early postnatal days.
Interestingly, microglia cells can be found inside the rosettes and it has been suggested that microglia might contribute to the degeneration of photoreceptors observed in the retinas of these mice35,36. Since activation of microglia have been implicated in the course of a variety of retinal diseases and sometimes is accompanied by alterations in retinal architecture10,27,37,38, the combination of SD-OCT and SLO can reveal correlations between microglia activation and retinal degeneration. Despite some challenges, the present protocol can provide such information longitudinally, and thus it would be valuable for the correlation of retinal damage with microglia distribution/activation and for the monitoring of therapeutic treatment effectiveness21.
The authors have nothing to disclose.
This work was supported by a grant of the Swiss National Science Foundation (SNSF; #320030_156019). The authors received nonfinancial support from Heidelberg Engineering GmBH, Germany.
Spectralis Imaging system (HRA+OCT) | Heidelberg Engineering, Germany | N/A | ophthalmic imaging platform system |
Heidelberg Eye Explorer | Heidelberg Engineering, Germany | N/A | Version 1.9.13.0 |
78D standard ophthalmic non-contact slit lamp lens | Volk Optical Inc., Ohio, USA | V78C | |
Spectralis wide angle 55° lens | Heidelberg Engineering, Germany | 50897-002 | |
ultra widefield 102° lens | Heidelberg Engineering, Germany | 50117-001 | |
medetomidine hydrochloride 1 mg/mL (Domitor) | Provet AG, Lyssach, Switzerland | Swissmedic Nr. 50'590 – ATCvet: QN05CM91 | anesthetic/analgesic |
ketamine 50mg/ml (Ketalar) | Parke-Davis, Zurich, Switzerland | 72276388 | anesthetic |
tropicamide 0.5% + phenylephrine HCl 2.5% (Augentropfen mix) | ISPI, Bern, Switzerland | N/A | pupil dilation |
Omnican Insulin-50 0.5 ml G30 0.3 x 12mm | B. Braun Mesungen AG, Carl-Braun-Straße, Germany | 9151125 | |
hydroxypropylmethylcellulose (Methocel 2%) | OmniVision, Neuhausen, Switzerland | N/A | |
+4 dpt rigid gas permeable contact lens | Quantum I, Bausch + Lomb Inc., Rochester, NY | N/A | Base Curve: 7.20 to 8.40 mm Diameter: 9.00 / 9.60 / 10.20 mm Power: -25.00 to +25.00 Diopters |
balanced salt solution (BSS) | Inselspital, Bern, Switzerland | N/A | |
silicon forceps | N/A | N/A | |
atipamezole 5 mg/mL (Antisedan) | Provet AG, Lyssach, Switzerland | N/A | α2 adrenergic receptor antagonist |
GraphPad Prism 7 | GraphPad Software, Inc, San Diego, CA, USA | N/A | statistical analysis software |