Dysregulation of neuroprotective lipoxin pathway in astrocytes in response to cytokines and ocular hypertension​

Abstract

Glaucoma leads to vision loss due to retinal ganglion cell death. Astrocyte reactivity contributes to neurodegeneration. Our recent study found that lipoxin B₄ (LXB₄), produced by retinal astrocytes, has direct neuroprotective actions on retinal ganglion cells. In this study, we aimed to investigate how the autacoid LXB₄ influences astrocyte reactivity in the retina under inflammatory cytokine-induced activation and during ocular hypertension. The protective activity of LXB₄ was investigated in vivo using the mouse silicone-oil model of chronic ocular hypertension. By employing a range of analytical techniques, including bulk RNA-seq, RNAscope in-situ hybridization, qPCR, and lipidomic analyses, we discovered the formation of lipoxins and expression of the lipoxin pathway in rodents (including the retina and optic nerve), primates (optic nerve), and human brain astrocytes, indicating the presence of this neuroprotective pathway across various species. Findings in the mouse retina identified significant dysregulation of the lipoxin pathway in response to chronic ocular hypertension, leading to an increase in 5-lipoxygenase (5-LOX) activity and a decrease in 15-LOX activity. This dysregulation was coincident with a marked upregulation of astrocyte reactivity. Reactive human brain astrocytes also showed a significant increase in 5-LOX. Treatment with LXB₄ amplified the lipoxin biosynthetic pathway by restoring and amplifying the generation of another member of the lipoxin family, LXA₄, and mitigated astrocyte reactivity in mouse retinas and human brain astrocytes. In conclusion, the lipoxin pathway is functionally expressed in rodents, primates, and human astrocytes, and is a resident neuroprotective pathway that is downregulated in reactive astrocytes. Novel cellular targets for LXB₄'s neuroprotective action are inhibition of astrocyte reactivity and restoration of lipoxin generation. Amplifying the lipoxin pathway is a potential target to disrupt or prevent astrocyte reactivity in neurodegenerative diseases, including retinal ganglion cell death in glaucoma.

Keywords: Astrocyte reactivity; Glaucoma; Lipoxin B4; Lipoxygenase; Neurodegeneration; Optic nerve; Retina.

Fig. 1

Expression of lipoxin pathway in mouse retina and optic nerve. (A) Representative images of RNAscope ISH (red) (indicated with white arrows) of retina and optic nerve using mRNA probes for Alox5, Alox15, and Fpr2. Nuclei were labeled with DAPI to distinguish nuclear layers of the retina. All images were obtained using a 40x objective. Scale bar = 50 μm. (n = 3). Labels: ganglion cell layer (GCL), inner nuclear layer (INL), outer nuclear layer (ONL). (B) Bar graphs represent the quantification of RNAscope images performed by Qupath software. (C) Bar graphs represent Ct values of lipoxin pathway genes in retinal astrocytes and optic nerves analyzed by qPCR (n = 3)

Fig. 2

Expression of lipoxin pathway in optic nerve of primates and primary human brain astrocytes. (A) RNA-seq of the macaque optic nerve showing lipoxin pathway genes (n = 2). (B) LC−MS/MS-based lipidomic quantification of endogenous lipid mediators in primate optic nerve (n = 2). (C) Representative immunofluorescence images of primary human brain astrocytes stained with GFAP, Pax2, Vimentin, and CD44. (D) LC−MS/MS-based lipidomic quantification of endogenous lipid mediators in human brain astrocytes (n = 2)

Fig. 3

Validation of the silicone oil model. (A) IOPs were increased after silicone oil (SO) injection (n = 20). (B) Representative confocal images of flat-mounted retina sections showing surviving RBPMS-positive (red) RGCs at 8 weeks. Scale bar = 50 μm. (C) Bar graph indicates % of peripheral RGC loss. (D) Average ERG waveforms of RGC responses (pSTR, -2.5 log cd.s.m^− 2) at 8 weeks for each group. (E) Representative OCT images of mouse retina at 8 weeks. GCC: ganglion cell complex, including RNFL, GCL, and IPL layers, indicated by double-ended yellow arrows. (F) Bar graph indicating the % reduction in GCC thickness at 8 weeks. (n = 4–6). ****p < 0.001. (Ctrl- Control; OHT- ocular hypertension)

Fig. 4

Regulation of lipoxin pathway in response to ocular hypertension. (A) The time course response of lipoxin pathway genes was analyzed by qPCR at 2, 4, and 8 weeks. The results were normalized to GAPDH and age-matched controls. n = 3–4 (2 retinas were pooled for each replicate). (B) LC‒MS/MS-based lipidomic quantification of endogenous lipid mediators in the mouse retina at 4 weeks, n = 3 (4 retinas were pooled for each replicate). *p < 0.05, **p < 0.01, **p < 0.001. (Ctrl- Control; OHT- ocular hypertension)

Fig. 5

Regulation of macroglia reactivity in response to ocular hypertension. (A) Representative immunofluorescence images of GFAP (green) at 2 weeks, and (B) GFAP (green) RBPMS (red) at 8 weeks of ocular hypertension. Nuclei were labeled with DAPI. All images were obtained using a 40x objective. Scale bar = 50 μm. Labels: ganglion cell layer (GCL), inner nuclear layer (INL), outer nuclear layer (ONL). (C) Representative images of a portion of flat-mounted retinas stained with GFAP (green) at 8 weeks. (D) Muller glia reactivity gene, GLAST, was analyzed by qPCR at 2 and 4 weeks. The results were normalized to GAPDH. (E) Astrocyte markers were significantly upregulated at 2, 4, and 8 weeks when analyzed by qPCR. The results were normalized to GAPDH and controls. n = 4 (2 retinas were pooled for each replicate). *p < 0.05, **p < 0.01, **p < 0.001. (Ctrl- Control; OHT- ocular hypertension)

Fig. 6

Effect of LXB₄treatment on astrocyte reactivity markers in the mouse retina. LXB₄ methyl ester treatment (n = 8, 5 ng/g IP; 0.5 ng/g topical) was started after one week of silicone oil injection every other day until the 4th week. Treatment with LXB₄ modulates astrocyte reactivity genes in the ocular hypertension model, n = 4 (2 retinas were pooled for each replicate). (Ctrl- Control; OHT- ocular hypertension). *p < 0.05, **p < 0.01, ***p < 0.001

Fig. 7

Effect of LXB₄ treatment on lipoxin pathway in human brain astrocytes. (A) Astrocyte reactivity genes analyzed by qPCR. The results were normalized to GAPDH. Data are presented as the mean ± SEM, n = 3–4. (B) Representative immunofluorescence images of LCN2 and 5-LOX. Nuclei were labeled with DAPI. (C) LC‒MS/MS-based lipidomic quantification of endogenous lipid mediators, n = 4–5. (Ctrl-untreated control astrocytes, RA- reactive astrocytes). *p < 0.05, **p < 0.01, ***p < 0.001