The pericytes in retinal vasculature were examined by immunofluorescent staining with platelet-derived growth factor receptor β after retro-orbital injection of fluorescent tomato lectin. The labeled retina was further treated with the tissue-clearing method and whole mounted for visualizing the three-dimensional views of pericytes surrounding retinal vasculature under a confocal microscope.

Retinal pericytes are essential for vascular development and stability of the retina, playing a key role in maintaining the integrity of the retinal vasculature. To provide a detailed view of the morphological characteristics of pericytes, this study described a new approach combining the retro-orbital injection of a fluorescent agent, immunofluorescent-staining, and tissue-clearing treatment. Firstly, the fluorescent tomato lectin was injected into the retro-orbital sinus of the live mouse to label the retinal vasculature. After 5 min, the mouse was sacrificed, and its intact retina was carefully removed from the retinal cup and immunofluorescently stained with platelet-derived growth factor receptor β to reveal the pericytes. Additionally, the stained retina was treated with tissue-clearing reagent and whole mounted on the microscope slide. Through these approaches, the retinal vasculature and pericytes were clearly observed in the transparent retina. Under a confocal microscope, we obtained more opportunities to take high-resolution images for further reconstructing and analyzing the morphological characteristics of pericytes along the retinal vascular tree in a three-dimensional view. Methodologically, this protocol offers an effective approach for visualizing pericytes within the retinal vasculature, providing valuable insights into their role under both physiological and pathological conditions.

Retinal pericytes are embedded within the basement membrane along the abluminal side of vascular endothelial cells, displaying a mesh-like pattern of short-range processes wrapped around the retinal blood vessels¹. This intimate contact between pericytes and the retinal vasculature contributes to maintaining retina environmental homeostasis². Although numerous studies have provided morphological evidence of pericytes in retinal vasculature, most of our understanding of the morphological characteristics of retinal pericytes comes from conventional histological techniques³.

In line with these studies, we designed this study to effectively showcase more morphological details of pericytes in the retinal microvasculature by combining traditional histological techniques with modern cutting-edge scientific methods. The approach here integrates three key techniques: retro-orbital injection of fluorescent tomato lectin in live mice, immunofluorescent staining with platelet-derived growth factor receptor β (PDGFR-β), and the solvent-based tissue clearing strategy. Retro-orbital injection of fluorescent tomato lectin has been proven effective and reliable for observing mouse retinal vasculature in both in vivo and in vitro settings^4,5. PDGFR-β serves as a commonly used biomarker for the immunofluorescent staining of pericytes⁶. To capture high-resolution images of pericytes within the thicker and whole-mount retinas, the tissue-clearing strategy enhances histological transparency^2,7,8.

While each of these techniques has been applied individually in retinal studies, their combined compatibility for investigating pericytes in mouse retinal vasculature remains uncertain. Since the average thickness of the mouse retina is approximately 180-220 µm⁹, antibody labeling and confocal imaging at this thick tissue without clearing treatment often result in high background signals, complicating the observation of the spatial relationship between pericytes and blood vessels¹⁰. Through the approaches described here, we aim to provide a straightforward method for visualizing pericytes within the retinal vasculature in a three-dimensional (3D) view under a confocal microscope. This enhanced cellular information from pericytes in the retinal vasculature could improve our understanding of retinal environmental homeostasis and underscore potential strategies for vascular protection, thereby enhancing functional outcomes in retinal vascular disorders.

For this study, three young adult male C57BL/6 mice (8-10 weeks old, weighing 20-25 g) were used. They were housed in a 12 h light/dark cycle with controlled temperature and humidity and allowed free access to food and water. This study was approved by the Ethics Committee of the Institute of Acupuncture and Moxibustion, China Academy of Chinese Medical Sciences (approval No. 2023-03-14-04) and conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (National Academy Press, Washington, D.C., 1996).

1. In vivo retro-orbital injection of fluorescent tomato lectin

1. Inject 50 mg/kg pentobarbital intraperitoneally to anesthetize the mouse. Assess the depth of anesthesia by evaluating the absence of a response to a toe pinch stimulus.
2. Place the mouse in the right lateral recumbency with its head facing to the left.
3. Press two fingers gently on the peri-orbital area to expose the mouse's left eye to allow the easiest administration of the injection into the left retro-orbital sinus of the animal¹¹.
4. Use a 1 mL syringe equipped with a 27G needle to gently pierce about 2-3 mm into the mouse's orbital venous sinus with the bevel on the needle facing forward at a 45° angle.
5. Inject 0.1 mL (1 mg/mL) of fluorescent tomato lectin into the mouse's orbital venous sinus.

2. Perfusion and enucleation of eyes

1. At 5 min after retro-orbital injection, inject tribromoethanol solution (250 mg/kg) intraperitoneally to deeply anesthetize the mouse. Subsequently, euthanize the animal via exsanguination and cardiac perfusion. Once breathing stops, open the thoracic cavity with surgical scissors¹².
2. Once breathing stops, open the thoracic cavity with surgical scissors. Sterilize surgical instruments in advance and operate in a fume hood. Use a 24G needle and insert it 2 mm into the left cardiac ventricle through the left ventricular apex. Perform perfusion at a rate of 3 mL/min with 20 mL of 0.9% physiological saline followed by 20 mL of 4% formaldehyde in 0.1 M phosphate buffer (PB, pH 7.4).
3. Place the sacrificed mouse on its side and remove the skin covering the eyes with scissors. Enucleate the eyes with scissors and forceps.
4. Cut the optic nerve and surrounding tissues, then lift out the eye and transfer to a 12-well plate for post-fixing in 4% formaldehyde in 0.1 M PB for 2 h.
5. Cryoprotect the tissue in 25% sucrose in 0.1 M PB (pH 7.4) at 4 °C until it sinks to the bottom of the solution.
   NOTE: Since ocular tissue dehydrates well in a sucrose solution, the retina can be more easily and completely detached, facilitating subsequent comprehensive studies.

3. Dissection of retinas

1. Transfer the eyes into a Petri dish containing 0.1 M PB (pH 7.4) using a plastic Pasteur pipette.
2. Pierce the edge of the cornea with sharp scissors, cut around the cornea and iris, and discard.
3. Use forceps to remove the lens and vitreous humor. Then, pull the cup-shaped retina away from the center of the eye with fine forceps. Rinse the retina with 0.1 M PB (pH 7.4) in a clean Petri dish.

4. Immunofluorescent-staining and tissue-clearing of retinas

1. Carefully remove the cup-shaped retina and incubate it in a 2% Triton X-100 solution in 0.1 M PB (pH 7.4) overnight at 4 °C.
2. Transfer the retina into the blocking solution (0.1 M PB + 1% Triton-X 100 + 0.2% sodium azide + 10% normal donkey serum) and rotate at 72 rpm on the shaker overnight at 4 °C.
3. Transfer the retina into the solution containing the primary antibodies of goat anti-PDGFR-β (10 µg/mL) in dilution buffer (0.1 M PB + 0.2% Triton-X 100 + 0.2% sodium azide + 1% normal donkey serum) in a microcentrifuge tube and rotate on the shaker for 2 days at 4 °C.
4. Wash the retina 2x with washing buffer (0.1 M PB + 0.2% Triton-X 100) at room temperature, then keep them in washing buffer overnight at 4 °C on the shaker.
5. The next day, transfer the retina into the mixed solution containing the 1 mg/mL secondary antibody of donkey anti-goat 488 and 0.2 ng/mL fluorescent nuclear 4',6-diamidino-2-phenylindole (DAPI) in a microcentrifuge tube and rotate at 72 rpm on the shaker for 5 h at 4 °C.
6. Wash the retina in a 6-well plate with washing buffer for 1 h, 2x, at room temperature. Keep the retina in washing buffer on the shaker overnight at 4 °C. Perform imaging and analysis as per step 5 to obtain the before tissue clearing views.
7. Transfer the retina to the tissue-clearing reagent and gently rotate it at 60 rpm on the shaker for 1 h at 37 °C.
8. After tissue clearing treatment, rinse the cup-shaped retina with 0.1 M PB (pH 7.4) in a clean Petri dish.
9. Make 4 radial incisions reaching approximately 2/3 of the radius of the retina using spring scissors to create a petal shape.
10. Flatten and mount the retina on a microscope slide, and then remove any excess PB with a small piece of absorbent paper. Keep the inside of the retina facing up on the slide.
11. Circle the mounted retina with a spacer and fill the gap with fresh tissue-clearing reagent and a coverslip.

5. Imaging and analyses

1. Obtain the outside views of the retina before (after completing step 4.6) and after tissue-clearing treatment.
2. Scan the montage views of labeled retina with the panoramic tissue slice scanner. Use a 2x objectives lens with NA of 0.08.
3. Take the higher magnification images by confocal imaging system equipped with the following objectives lenses: 10x lens with NA of 0.40 and 40x lens with NA of 0.95. Set the confocal pinhole at 152 µm (10x objective lens) and 105 µm (40x objective lens). Set the spatial resolution of image capture at 1024 x 1024 pixels (10x objective lens) and 640 x 640 pixels (40x objective lens).
4. Capture 50 images (Z-stack) in 2 µm frames from the labeled retina. Integrate all images in a single in-focus under the projection/topography mode. For the setup, click Set Start Focal Plane > Set End focal plane > Set step size > Choose Depth Pattern > Image Capture > Z series.
5. Capture images using the following excitation and emission wavelengths for different fluorescence signals: Green fluorescence signals at 499 nm and 519 nm; red fluorescence signals at 591 nm and 618 nm; and blue fluorescence signals at 401 nm and 421 nm.
6. Reconstruct the images in the three-dimensional pattern by importing Z-stack confocal images into an analysis software by selecting Add New Surfaces > Choose Volume settings > Choose Source Channel > Adjust display > Adjust Surface Angle > Snapshot.

From the outside views of the whole-mount retina, the tissue transparency of the whole-mount retina was increased using the tissue-clearing treatment compared to that of the retina without tissue-clearing (Figure 1).

For the histological examination on the whole-mount retina, the retinal vasculature was labeled in red with fluorescent tomato lectin through the retro-orbital injection, and retinal pericytes are shown in green with PDGFR-β (Figure 2, Figure 3, and Figure 4). Although this labeling could be observed on the sample before and after tissue clearing, as a comparison, the background noise of the retinal sample was reduced by the tissue-clearing treatment (Figure 2). The treatment also made the labeled retinal pericytes and vasculature clearer under the panoramic tissue slice scanner (Figure 2) and laser scanning confocal microscope (Figure 3).

The gradational distribution of retinal pericytes is also demonstrated here along the retinal vascular tree (Figure 3). From the central part of the retina to its peripheral region, pericytes are located on the basement membrane of the retinal vasculature, including the vascular trunk and branches and the capillaries (Figure 3).

Based on the above observation, the spatial correlation of pericytes and retinal vasculature was further assessed with the higher-magnified Z-stack images from the cleared retinal sample (Figure 4). With the aid of the image processing system, we further provided the 3D views to demonstrate that pericytes were tightly wrapped around the retinal vasculature in different diameters (Figure 4). Figure 4A0-D0 represents the hierarchical branching of retinal vessels from highest to lowest. Since pericytes are more abundantly distributed in capillaries, the PDGFRβ staining in Figure 4D0 shows stronger intensity and signal compared to Figure 4A0-C0.

In contrast to the mouse retina without tissue-clearing treatment (Figure 1A, Figure 2A, Figure 3A0-D0), the present approach has effectively reduced the retinal background signals for demonstrating the spatial relationship between pericytes and blood vessels from the whole-mount retina (Figure 1B, Figure 2B, Figure 3A1-D1).

Figure 1: Comparison of optical transparency of the whole mount retina without and with tissue clearing treatment. (A, B) The outside views of the whole mount retina before (A) and after (B) tissue clearing treatment. Please click here to view a larger version of this figure.

Figure 2: Comparison of labeling features of retinal pericytes and vasculature from the whole mount retina without and with tissue clearing treatment. (A, B) Representative montage views of the whole mount retina before (A) and after (B) tissue clearing treatment were taken under the Virtual Slide System. The vasculature is in red (fluorescent tomato lectin), pericytes are in green (PDGFR-β), and the cellular nucleus is in blue (DAPI). Please click here to view a larger version of this figure.

Figure 3: Comparison of the imaging resolution of the retinal pericytes gradationally along the retinal vascular tree before and after tissue-clearing treatment. (A-D) The representative images from the labeled retina before tissue clearing treatment (A0-D0) and after tissue clearing treatment (A1-D1) show the distribution of pericytes surrounding the retinal vasculature in order of vascular trunk (A0, A1), the first order branches (B0, B1), the second order branches (C0, C1), and the capillary (D0, D1). The vasculature is in red (fluorescent tomato lectin), pericytes are in green (PDGFR-β), and the cellular nucleus is in blue (DAPI). Please click here to view a larger version of this figure.

Figure 4: Spatial correlation of pericytes with retinal vasculature in different diameters. (A-D) Representative high-resolution images from the labeled retina along the retinal vascular tree showing the distribution of pericytes surrounding the retinal vasculature in different diameters, including vascular trunk (A0, A1), the first order branches (B0, B1), the second order branches (C0, C1), and the capillary (D0, D1). A1-D1: Corresponding images of panels A0-D0 were reconstructed to show the spatial correlation of pericytes with retinal vasculature in different diameters in the three-dimensional views and further demonstrate the magnified views of local regions in panels A1-D1 (arrows) with insets. The vasculature is in red (fluorescent tomato lectin), pericytes are in green (PDGFR-β), and the cellular nucleus is in blue (DAPI). Please click here to view a larger version of this figure.

In this study, we detailed the technical process for demonstrating the morphology of pericytes in retinal vasculature using retro-orbital injection of fluorescent tomato lectin, immunofluorescent examination with PDGFR-β, and solvent-based tissue clearing. The results show that these techniques are highly compatible for highlighting pericytes in the whole-mount retina. The combination of these methodologies offers an effective approach for visualizing the morphological characteristics of pericytes in retinal vasculature from a detailed 3D perspective.

One of the key techniques, retro-orbital injection in mice, is proven to be a simple and optimal method for delivering vascular labeling agents, outperforming tail vein and intraperitoneal injection methods for observing mouse retinal vasculature^13,14. Although the principle and mechanism behind vascular labeling through retro-orbital injection is not fully understood, it is evident that this method robustly labels retinal vasculature with fluorescent tomato lectin. The fluorescent labeling remains stable through both histochemical processing and tissue clearing. Therefore, beyond its use in in vivo imaging, retro-orbital injection with fluorescent tomato lectin effectively labels the blood vessel lumen, offering a clear view of the retinal vasculature for in vitro and ex vivo histological examination.

To label the retinal pericytes, we utilized immunofluorescent staining with PDGFR-β¹⁵. Considering the stained retina underwent further treatment with tissue clearing reagent to reduce tissue light-scattering under microscopic morphological examination, the retina was incubated in the staining solution for a longer time to enhance the penetration of antibody^16,17.

The whole-mount retina is relatively transparent, allowing an overview of the retinal vasculature¹⁸. However, due to the light-scattering with its thickness, it limits light penetration and high-resolution imaging¹⁹. With the aid of tissue-clearing techniques, it has become possible to view the pericytes in the whole-mount retina from a high-resolution perspective^19,20. However, these methods have not yet been widely applied to the investigation of mouse retinal pericytes. In this study, the solvent-based tissue-clearing approach significantly enhanced our ability to visualize pericytes in the retinal vasculature, enabling us to obtain high-resolution Z-stack images using a laser scanning confocal microscope.

Pericytes, distributed along the microvascular tree, play essential roles in the complex spatial and temporal distribution of blood within the retinal capillary network²². Their highly plastic morphology and adaptive contractile capabilities make pericytes ideal for regulating capillary blood flow in response to local neuronal activity under physiological and pathological conditions²³. As a key player of the blood-retina barrier or retinal neurovascular coupling, retinal pericytes are also drawn attention in the maintenance of the neovascularization, angiogenesis, and leukocyte transmigration²⁴.

As a technical limitation, it should be kept in mind that, in addition to PDGFR-β, other markers such as NG-2 and CD13 have also been used to identify pericytes^22,24,25. Therefore, selecting a proper pericyte-specific marker to demonstrate the pericytes unambiguously is still a challenge. This study provides further evidence that PDGFR-β is a better candidate for revealing the morphological features of pericytes.

In summary, we demonstrated that the combination of retro-orbital injection of fluorescent tomato lectin, immunofluorescent staining, and tissue clearing enabled the acquisition of high-resolution images, revealing the detailed morphology of pericytes in the mouse retinal vasculature under physiological conditions. These findings suggest that this method could be valuable for future investigations into the alterations of retinal pericytes under pathological conditions.

No conflicts of interest.

This study was supported by the Beijing Natural Science Foundation (No. 7244480), the CACMS Innovation Fund (No. CI2021A03407), the National Natural Science Foundation of China (No. 82004492), and the Fundamental Research Funds for the Central public welfare research institutes (Nos. ZZ-YQ2023008, ZZ14-YQ-032, ZZ-JQ2023008, ZZ-YC2023007).