Tissue Hypoxia and Associated Innate Immune Factors in Experimental Autoimmune Optic Neuritis

Abstract

Visual loss in acute optic neuritis is typically attributed to axonal conduction block due to inflammatory demyelination, but the mechanisms remain unclear. Recent research has highlighted tissue hypoxia as an important cause of neurological deficits and tissue damage in both multiple sclerosis (MS) and experimental autoimmune encephalomyelitis (EAE) and, here, we examine whether the optic nerves are hypoxic in experimental optic neuritis induced in Dark Agouti rats. At both the first and second peaks of disease expression, inflamed optic nerves labelled significantly for tissue hypoxia (namely, positive for hypoxia inducible factor-1α (HIF1α) and intravenously administered pimonidazole). Acutely inflamed nerves were also labelled significantly for innate markers of oxidative and nitrative stress and damage, including superoxide, nitric oxide and 3-nitrotyrosine. The density and diameter of capillaries were also increased. We conclude that in acute optic neuritis, the optic nerves are hypoxic and come under oxidative and nitrative stress and damage. Tissue hypoxia can cause mitochondrial failure and thus explains visual loss due to axonal conduction block. Tissue hypoxia can also induce a damaging oxidative and nitrative environment. The findings indicate that treatment to prevent tissue hypoxia in acute optic neuritis may help to restore vision and protect from damaging reactive oxygen and nitrogen species.

Keywords: hypoxia inducible factor-1α; multiple sclerosis; nitric oxide; oxidative stress; peroxynitrite; superoxide.

Figure 1

Identification of EAE-ON and EAE-NON. Examples of fluorescent images of optic nerve sections labelled with (a) ED1 and (c) IBA1 in an IFA animal (IFA; (a-i,c-i)) and in an inflamed optic nerve of an animal with EAE (EAE-ON; (a-ii,c-ii)). Both (b) ED1 and (d) IBA1 showed significantly greater labelling in EAE-ON than in EAE-NON and IFA nerves. Each data point in (b,d) represents one nerve. (e,f) show inflamed nerves (EAE-ON). In such nerves, the labelling for ED1 and more intense IBA1 commenced from the orbital end of the nerve (e), extending in some nerves towards the chiasm and involving the chiasm (f). Mean ± SEM, one-tailed independent t-test, * p < 0.05, ** p < 0.01, *** p < 0.001, bar = 100 µm.

Figure 2

The inflamed optic nerve was hypoxic. Photomicrographs of optic nerve sections labelled for (a) HIF1α and (b) pimonidazole in tissue from an IFA animal (IFA; (a-i,b-i)), and tissue representing EAE-NON (a-ii,b-ii) and EAE-ON (a-iii,b-iii) from animals immunised for EAE. The EAE-ON nerves showed significantly more intense labelling for (c) HIF1α compared with nerves from IFA animals, which was more intense in the most inflamed nerves (d). A reduction in absolute tissue oxygenation was confirmed in the EAE-ON nerves by significantly greater labelling for (e) pimonidazole compared with nerves from IFA animals. Each data point in (c–e) represents one nerve. Mean ± SEM, one-tailed independent t-test and linear regression (in EAE-ON group; (d)), * p < 0.05, ** p < 0.01, bar = 50 µm.

Figure 3

Vascular changes in the inflamed optic nerve. Fluorescent images of sections of optic nerves labelled for (a) GLUT1 in nerves from an IFA animal (IFA; (a-i)) and nerves identified as EAE-NON (a-ii) and EAE-ON (a-iii). The inflamed nerve showed a tendency for vessels to be less aligned with the length of the nerve, perhaps due to displacement by inflammatory cells. (b) Vessel number appeared significantly increased in the EAE-ON group, accompanied by significant (c) dilation compared with the EAE-NON and IFA groups. Each data point in (b,c) represents one nerve. Mean ± SEM, one-tailed independent t-test, ns: not significant, ** p < 0.01, *** p < 0.001, bar = 100 µm.

Figure 4

The inflamed optic nerve showed increased labelling for the presence of suspected superoxide (DHE), nitric oxide (iNOS) and peroxynitrite (3NT). Sections of optic nerves showing fluorescence resulting from (a) DHE and labelled for (c) iNOS; (e) 3NT in animals ‘immunised’ with IFA (a-i,c-i,e-i) and those immunised for EAE but without optic neuritis (EAE-NON; (a-ii,c-ii,e-ii)) or with optic neuritis (EAE-ON; (a-iii,c-iii,e-iii)). Inflamed optic nerves from animals with EAE showed significantly more intense fluorescence from (b) DHE and labelling for (b) iNOS, (c) 3NT and (f) DHE. The labelling for iNOS was highest for EAE-ON at peak of disease (day 2 after disease onset), but was significantly reduced when examined on day 4 (d). On day 2, the labelling for iNOS was positively correlated with the magnitude of inflammation (g), but labelling for 3NT was absent in the most inflamed nerves (h). Each data point in (b,d,f–h) represents one nerve. Mean ± SEM, one-way ANOVA, ns: not significant, * p < 0.05, ** p < 0.01, *** p < 0.001, bar = (a,c) 100 µm or (e,g) 50 µm.