Non-Invasive Evaluation of Retinal Vascular Alterations in a Mouse Model of Optic Neuritis Using Laser Speckle Flowgraphy and Optical Coherence Tomography Angiography

Abstract

Optic neuritis, a characteristic feature of multiple sclerosis (MS), involves the inflammation of the optic nerve and the degeneration of retinal ganglion cells (RGCs). Although previous studies suggest that retinal blood flow alterations occur during optic neuritis, the precise location, the degree of impairment, and the underlying mechanisms remain unclear. In this study, we utilized two emerging non-invasive imaging techniques, laser speckle flowgraphy (LSFG) and optical coherence tomography angiography (OCTA), to investigate retinal vascular changes in a mouse model of MS, known as experimental autoimmune encephalomyelitis (EAE). We associated these changes with leukostasis, RGC injury, and the overall progression of EAE. LSFG imaging revealed a progressive reduction in retinal blood flow velocity and increased vascular resistance near the optic nerve head in the EAE model, indicating impaired ocular blood flow. OCTA imaging demonstrated significant decreases in vessel density, number of junctions, and total vessel length in the intermediate and deep capillary plexus of the EAE mice. Furthermore, our analysis of leukostasis revealed a significant increase in adherent leukocytes in the retinal vasculature of the EAE mice, suggesting the occurrence of vascular inflammation in the early development of EAE pathology. The abovechanges preceded or were accompanied by the characteristic hallmarks of optic neuritis, such as RGC loss and reduced visual acuity. Overall, our study sheds light on the intricate relationship between retinal vascular alterations and the progression of optic neuritis as well as MS clinical score. It also highlights the potential for the development of image-based biomarkers for the diagnosis and monitoring of optic neuritis as well as MS, particularly in response to emerging treatments.

Keywords: biomarker; laser speckle flowgraphy (LSFG); multiple sclerosis (MS); optic neuritis; optical coherence tomography angiography (OCTA); retinal blood flow; retinal vascular inflammation; retinal vasculature.

Figure 1

Illustration of the analysis of MBR and waveform values from LSFG. (A) Representative composite color maps using the MBR as measured with LSFG with a 200-pixel diameter encircling the optic nerve head. (B) Enlargement of the optic nerve head area. The red color indicates high MBR, and the blue color indicates low MBR. (C) LSFG Analyzer Software automated binarization of the image into vessels (white area) and tissue (black area). (D) Fluctuations of MBR throughout a 118-frame scan with a calculation of waveform parameters: beat strength (BS), beat strength over MBR (BOM), Max MBR, Min MBR, and Max–Min MBR. (E) The total retinal artery and vein analyzer (TRAVA) automated the calculation of the total retinal flow index divided by the vessel width (TRFI/w) for the top four highest velocity arteries only (top four in bright red).

Figure 2

Evaluation of the EAE mice model. The EAE model was induced in WT mice. Body weight (A) and EAE clinical score (B) were measured at 0, 7, 10, 14, 21, and 28 dpi. n = 5–7. (C) Representative images of RBPMS-positive RGCs from the control and EAE mice at 28 dpi. The bar graph represents the quantification of the RGC number. Scale bar = 50 µm. n = 9–14; * p < 0.05, *** p < 0.001, **** p < 0.0001.

Figure 3

EAE impairs retinal vascular function. The EAE model was induced in WT mice. Retinal vasculature was examined via OCTA at 14 dpi, and representative OCTA images were shown. Vessel density, vessel branch points (number of junctions), and total vessel length were quantified with AngioTool. n = 14–16; ** p < 0.01, *** p < 0.001. SVC: superficial vascular complex; ICP: intermediate capillary plexus; DCP: deep capillary plexus.

Figure 4

EAE impairs retinal blood flow. The EAE model was induced in WT mice. Retinal blood flow was examined with LSFG at 7, 10, 14, and 28 dpi. (A) Representative composite color maps using the MBR via LSFG with a 200-pixel diameter encircling the optic nerve head. (B) Enlargement of the optic nerve head area. The red color indicates high MBR, and the blue color indicates low MBR.

Figure 5

EAE decreases retinal blood flow velocity and increases vascular resistance. LSFG images were analyzed using the LSFG Analyzer software (version 3.00) for the MBR in MV, MT, and MA (A), waveform parameters BS, BOM, Max MBR, Min MBR, and Max–Min MBR in the overall area of the ONH (B); and TRFI/w (C). n = 9–16; * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. MV: MBR in the vessel area; MT: MBR in the tissue area; MA: MBR in all areas.

Figure 6

Visual acuity in the EAE mice. The EAE model was induced in WT mice. Visual acuity was evaluated at 7, 10, 14, and 28 dpi. n = 9–14; **** p < 0.0001.

Figure 7

Leukocyte attachment is increased in the retina of the EAE mice. (A,B) SLO imaging was performed right before and 7 days after EAE induction, and stationary leukocytes in the retinas were quantified. (C–E) Leukostasis was performed 14 days after EAE induction. Representative images of leukostasis in peripheral and central retinas from the control and EAE mice were shown. Images are enlarged from the square of C. White arrows indicate adherent leukocytes. The bar graph represents the number of adherent leukocytes per retina. Scale bar = 50 µm. n = 6; ** p < 0.01, *** p < 0.001.