Calbindin 2-specific deletion of arginase 2 preserves visual function after optic nerve crush

Abstract

We previously found that global deletion of the mitochondrial enzyme arginase 2 (A2) limits optic nerve crush (ONC)-induced neuronal death. Herein, we examined the cell-specific role of A2 in this pathology by studies using wild type (WT), neuronal-specific calbindin 2 A2 KO (Calb2^cre/+ A2 ^f/f), myeloid-specific A2 KO (LysM^cre/+ A2^f/f), endothelial-specific A2 KO (Cdh5^cre/+ A2^f/f), and floxed controls. We also examined the impact of A2 overexpression on mitochondrial function in retinal neuronal R28 cells. Immunolabeling showed increased A2 expression in ganglion cell layer (GCL) neurons of WT mice within 6 h-post injury and inner retinal neurons after 7 days. Calb2 A2 KO mice showed improved neuronal survival, decreased TUNEL-positive neurons, and improved retinal function compared to floxed littermates. Neuronal loss was unchanged by A2 deletion in myeloid or endothelial cells. We also found increased expression of neurotrophins (BDNF, FGF2) and improved survival signaling (pAKT, pERK1/2) in Calb2 A2 KO retinas within 24-hour post-ONC along with suppression of inflammatory mediators (IL1β, TNFα, IL6, and iNOS) and apoptotic markers (cleavage of caspase3 and PARP). ONC increased GFAP and Iba1 immunostaining in floxed controls, and Calb2 A2 KO dampened this effect. Overexpression of A2 in R28 cells increased Drp1 expression, and decreased mitochondrial respiration, whereas ABH-induced inhibition of A2 decreased Drp1 expression and improved mitochondrial respiration. Finally, A2 overexpression or excitotoxic treatment with glutamate significantly impaired mitochondrial function in R28 cells as shown by significant reductions in basal respiration, maximal respiration, and ATP production. Further, glutamate treatment of A2 overexpressing cells did not induce further deterioration in their mitochondrial function, indicating that A2 overexpression or glutamate insult induce comparable alterations in mitochondrial function. Our data indicate that neuronal A2 expression is neurotoxic after injury, and A2 deletion in Calb2 expressing neurons limits ONC-induced retinal neurodegeneration and improves visual function.

Fig. 1. Arginase 2 expression is increased in retinal neurons after optic nerve crush.

A Schematic representation of time-points for analyses after ONC. B, C Immunofluorescence imaging of WT retinal sections for A2 immunostaining (in red) showed increases in A2 in response to ONC at 6 h and 7 days after injury (arrows). n = 4. Scale bar = 40 μm.

Fig. 2. Calbindin2 tdTomato imaging of retinal neurons.

A Breeding strategy for cre-lox-mediated expression of the tdTomato fluorescent protein. The recombination allows for excision of the STOP codon, thus leading to tdTomato expression under the CAG (CMV enhancer, chicken beta-Actin promoter, and rabbit beta-Globin splice acceptor site) promoter in cells expressing Calb2-cre. B Immunofluorescence imaging of Calb2/td retina sections showed strong colocalization of tdTomato expression with Calbindin2 (Calb2) and NeuN in GCL and INL. C, D Immunofluorescence imaging of Calb2/td retina sections showed strong colocalization of tdTomato expression in Calbindin1 (Calb1)-positive amacrine and horizontal cells as well as in Choline acetyltransferase (ChAT)-positive amacrine cells. n = 3. Scale bar = 40 μm.

Fig. 3. Calbindin 2-specific A2 deletion is neuroprotective and preserves visual function after ONC.

Neuronal-specific (Calb2^cre/+ A2^f/f), myeloid-specific (LysM^cre/+ A2^f/f) and endothelial cell-specific (Cdh5^cre/+ A2^f/f) A2 KO, and floxed littermate mice were subjected to ONC injury and retinas were collected for analysis on day 14. A Confocal imaging of retinal flatmounts labeled with the neuronal marker, NeuN and (B) quantification showing a significant increase in NeuN positive cells in Calb2 A2 KO retina compared to A2 floxed control. C, D Myeloid-specific or endothelial cell-specific A2 KO did not show protection against ONC-induced neuronal loss. Data are presented as percent of A2^f/f sham. n = 5; scale bar = 40 μm. Data are presented as mean ± SEM. Retinal functional changes were measured by OptoMotry responses (OMR) and Pattern ERG in Calb2 A2 KO and floxed littermate mice at day 14 post-ONC. E, F Both visual acuity and contrast sensitivity were improved in Calb2 A2 KO mice as compared with the A2 floxed controls. n = 7–8. Data are presented as mean ± SEM. G Quantification of P1-N2 amplitude in the Pattern ERG showed a marked improvement in the Calb2 A2 KO mice as compared with the floxed controls. n = 5. Data are presented as mean ± SEM.

Fig. 4. Calbindin 2-specific A2 deletion increases neurotrophic factor expression and promotes pro-survival signaling.

A, B Quantitative RT-PCR showed a trend towards suppression of the neurotrophic factors BDNF and FGF2 at 24 h after ONC in the A2 floxed control retinas, whereas both were markedly increased in the Calb2 A2 KO retinas. n = 4. Data are presented as mean ± SEM. C Western blotting and (D, E) quantification of Akt and ERK1/2 phosphorylation in whole retinal lysates at 24 h after injury showed significant activation of pro-survival signaling in the Calb2 A2 KO retinas as compared with the A2 floxed controls. n = 5. Data are presented as mean ± SEM.

Fig. 5. Calbindin 2-specific A2 deletion limits ONC-induced cell death.

A Retinal cell death was measured by TUNEL immunolabeling of retina cross sections in Calb2 A2 KO and floxed control mice at day 4 post-ONC. C Quantification showed increased TUNEL positive cells in the GCL (arrows) at 4 days after ONC. Calb2 A2 KO mice also showed fewer TUNEL positive RGC neurons compared to A2 floxed controls. n = 4. Scale bar = 40 μm. Data are presented as mean ± SEM. B Western blotting and (D, E) quantification on whole retinal lysates at 5 days after ONC showed significant reductions in levels of cleaved Caspase 3 and PARP in the Calb2 A2 KO retinas as compared with the A2 floxed controls. n = 5. Data are presented as mean ± SEM.

Fig. 6. Calbindin 2-specific A2 deletion reduces ONC-induced activation of Müller cells and microglia/macrophage and prevents inflammation.

Immunolabeling for (A) GFAP and (B) Iba1 in retina cross sections together with quantification (C, D) showed increased activation of Müller cells and microglia/macrophage, respectively, in A2 floxed control retinas at 4 days after ONC. This was ameliorated in the Calb2 A2 KO retinas. n = 4. Scale bar = 40 μm. Data are presented as mean ± SEM. E–H Quantitative RT-PCR showed upregulation of proinflammatory mediators - IL1β, TNFα, IL6, and iNOS at 5 days after ONC in the A2 floxed control retinas. Calb2 A2 KO significantly blocked this increase. n = 4. Scale bar = 50 μm Data are presented as mean ± SEM.

Fig. 7. A2 expression induces Drp1-mediated mitochondrial dysfunction in vivo and in vitro.

A, C WT and A2KO mice were subjected to ONC for 6 h. Western blotting and quantification of Drp1 in whole retinal lysates at 6 h after injury showed significant increase in Drp1 in the WT retinas compared with the sham. Global deletion of A2 significantly suppressed ONC-induced Drp1 upregulation. n = 4. Data are presented as mean ± SEM. B, D Retinal neuronal R28 cells were differentiated and transduced with Ad-RFP or Ad-A2 at increasing multiplicities of infection (5, 10, 20 MOI). Western blot analyses with quantification showed increased expression of Drp1 with increasing expression of A2 compared to R28 transduced with increasing RFP expression. n = 4. Data are presented as mean ± SEM. E, F A2 or RFP overexpressing R28 cells were treated with 1 µM of ABH or vehicle for 16 h. Western blot analyses with quantification showed ABH treatment significantly reduced Drp1 expression in A2 overexpressing cells. n = 4. Data are presented as mean ± SEM.

Fig. 8. A2 overexpression impairs mitochondrial respiration in vitro.

A2 or RFP were overexpressed in retinal neuronal R28 cells, followed by treatment with vehicle or 1 µM ABH or 5 mM glutamate or vehicle for 16 h. Oxygen consumption rate (OCR) was measured using Seahorse Xfe96. A–D A2 overexpression decreased OCR, along with decreased basal respiration, maximal respiration, and ATP production. Treatment with ABH in A2 overexpressing cells significantly improved these parameters. *P < 0.05; **P < 0.01; ****P < 0.0001. n = 12. Data are normalized to protein content and presented as mean ± SEM. E–H A2 overexpression and/or L-glutamate treatment decreased OCR. Basal respiration, maximal respiration, and ATP production showing a decreased OCR with A2 overexpression and/or L-glutamate treatment group compared to RFP. ****P < 0.0001. n = 11. Data are normalized to protein content and presented as mean ± SEM.