Hyperhomocysteinemia-induced VCID results in visual deficits, reduced neuroinflammation and vascular alterations in the retina

Abstract

Over recent years, the retina has been increasingly investigated as a potential biomarker for dementia. A number of studies have looked at the effect of Alzheimer's disease (AD) pathology on the retina and the associations of AD with visual deficits. However, while OCT-A has been explored as a biomarker of cerebral small vessel disease (cSVD), studies identifying the specific retinal changes and mechanisms associated with cSVD are lacking. Using our model of hyperhomocysteinemia-induced cSVD, we aimed to identify the effects of cSVD on visual sensitivity and cognition, retinal glial and vascular cells, and neuroinflammatory and cardiovascular gene expression changes. We placed C57Bl6/SJL mice on a HHcy-inducing diet, a model that has been well characterized to have vascular pathologies in the brain similar to pathologic cSVD. After 14 weeks on diet, mice underwent the Visual-Stimuli 4-arm Maze to identify visual deficits. Whole mount retinas were stained for vessels, microglia and astrocytes to identify glial and vascular changes. Finally, neuroinflammatory and cardiovascular gene expression was measured using NanoString's nCounter system. Ultimately, HHcy led to visual changes that specifically affected the reaction to blue and white light, slightly decreased vascular volume and significantly decreased interaction of microglia with the vasculature, as well as downregulation of inflammatory and vascular genes. These changes provide novel insights and reproduce some prior observations. These studies highlight retinal changes in association with cSVD and serve as a precaution when interpreting vision-dependent cognitive testing of cSVD models.

Keywords: Astrocytes; Hyperhomocysteinemia; Microglia; Retina; VCID; Vessels, Vision.

Fig. 1

Vision and cognition are impaired by HHcy. (A) Lux conditions for the Red High and Equal conditions in the ViS4M. (B) Percent entries for each colored arm in the Red High and Equal conditions. (C) Duration spent in each arm for the Red High and Equal conditions. (D) Percent alternations in the Red High and Equal conditions. Depiction of Paths 1, 2, 3 (E) and the percent of each path taken in the Red High and Equal conditions (F). (G) Bidirectional transitions in the Red High and Equal conditions. (H) Chord diagrams showing unidirectional transitions in the Red High and Equal conditions. I) Graphed unidirectional transitions in the Red High and Equal conditions. * indicates P < 0.05, ** indicates P < 0.01

Fig. 2

Rod and cone gene expression are increased in HHcy. (A) Gene expression for rod specific markers. (B) Gene expression for cone specific markers. Data are shown as a fold change from control mice. * indicates P < 0.05, ** indicates P < 0.01

Fig. 3

HHcy does not alter retinal microglia or astrocytes. Representative images of IBA-1 staining in the inner (A) and outer (B) retina of mice on HHcy or control diet. Scale bar in images = 100 μm. (C) Quantification of microglial staining in the retina. Representative images of GFAP staining in the inner (D) and outer (E) retina of mice on HHcy or control diet. Scale bar in images = 100 μm. (F) Quantification of astrocyte staining in the retina

Fig. 4

Vessels and glial coverage of the vasculature are reduced by HHcy. Representative images of IBA-1 (purple), GFAP (red) and Ly6c (green) staining in the inner (A) and outer (B) retina of mice on HHcy or control diet. Scale bar in images = 100 μm. Quantification of the number of vessels (C), total vessel volume (D), average vessel volume (E), and major axis length (F). Quantification of the number of combined astrocytes and microglia touching vessels (G), microglia alone touching vessels (H), and astrocytes alone touching vessels (I). Quantification of combined coverage of astrocytes and microglia of vessels (J), microglia alone coverage of vessels (K), and astrocytes alone coverage of vessels (L). * indicates P < 0.05

Fig. 5

HHcy reduces neuroinflammatory and vascular gene expression. (A) Volcano plots for neuroinflammatory and cardiovascular genes. Data are shown as a fold change from control mice. (B) STRING-db network connections of significant genes with a fold change above 0.25 or below − 0.25.