NRF2/ARE-mediated antioxidant response to glaucoma: Role of astrocytes and retinal ganglion cells

doi:10.21203/rs.3.rs-1750335/v1

Download PDF

Research Article

NRF2/ARE-mediated antioxidant response to glaucoma: Role of astrocytes and retinal ganglion cells

https://doi.org/10.21203/rs.3.rs-1750335/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: Glaucoma, the second leading cause of irreversible blindness worldwide, is associated with age and sensitivity to intraocular pressure (IOP). We use an inducible mouse model, in which we detect increased levels of reactive oxygen species (ROS). In response, the retina elicits an endogenous antioxidant response mediated by Nuclear factor erythroid 2-Related Factor 2 (NRF2), a transcription factor that binds to the antioxidant response element (ARE). Blocking this pathway results in earlier and greater glaucoma pathology. In this study, we sought to determine if this endogenous antioxidant response occurs within the retinal ganglion cells (RGCs) or astrocytes, and if overexpression of NRF2 in RGCs is neuroprotective.

Methods: We used Nrf2 ^fl/fl mice and cell-type specific adeno-associated viruses (AAVs) expressing Cre, we altered Nrf2 levels in either the RGCs or astrocytes. Then, we quantified the endogenous antioxidant response, visual function and histology after IOP elevation.

Results: Knockdown of Nrf2 in either cell type reduced the endogenous antioxidant response, accelerated and exacerbated visual function deficits and increased axon degeneration in the optic nerve. Overexpression of Nrf2 in the RGCs resulted in lower superoxide levels, preservation of the photopic negative response (PhNR) amplitude, and decreased axon degeneration after IOP elevation.

Conclusions: Together, these data suggest that the endogenous Nrf2/ARE-mediated antioxidant response is either driven by both cell types and/or there is extensive ROS-mediated communication between these cell-types in glaucoma. Regardless, RGC-directed Nrf2 overexpression is a viable therapeutic approach for the treatment of glaucoma.

glaucoma

Nrf2

oxidative stress

retinal ganglion cell

optic nerve

antioxidant

astrocyte

Glaucoma is a neurodegenerative disease characterized by the progressive degeneration and eventual death of retinal ganglion cells (RGCs), the neurons necessary for the processing and transmitting of visual information from the retina to the brain^1–4. It results in irreversible blindness and affects almost 80 million people worldwide, with that number projecting to increase to 111 million by 2040^4,5. The major risk factors for developing glaucoma are age, genetics, and sensitivity to intraocular pressure (IOP). Current treatments for glaucoma aim to decrease IOP. However, poor patient compliance with IOP lowering drops combined with lack of responsiveness to IOP lowering drugs and surgeries in some patients leads to permanent vision loss in many patients. Thus, further research aimed at protecting RGCs, independent of IOP, is an approach that needs to be explored^1,3,6.

While the underlying molecular cause of RGC degeneration and death in glaucoma is complex, accumulation of reactive oxygen species (ROS) plays a role^7–10. One of the key ways that cells respond to oxidative stress is through activation of the transcription factor, Nuclear factor erythroid 2-Related Factor 2 (Nrf2). When activated, Nrf2 binds to the Antioxidant Response Element (ARE), an enhancer element the is upstream of many antioxidant genes. Under homeostatic conditions, Nrf2 remains in the cytoplasm, where it is sequestered by its repressor protein, Kelch-like ECH associated protein 1 (Keap1) and is constantly targeted for degradation via ubiquitination. When Keap1 is oxidized, it undergoes a conformational change that releases Nrf2, allowing it to accumulate and translocate into the nucleus of the cell where it can bind to the ARE, and activate antioxidant gene transcription. Nrf2 can also be released from Keap1 upon phosphorylation from upstream signaling molecules, resulting also in nuclear translocation^11–14.

Nrf2 affects the degree and onset of neurodegeneration in many models, including models of RGC injury and glaucoma^14,15−20. We previously showed that Nrf2 is phosphorylated by the PI3K/Akt pathway at 2 weeks post-IOP elevation²¹. Inhibiting this pathway prevented increases in ARE-driven transcripts, showing that the retina endogenously responds to increased IOP via activation of the NRF2-ARE pathway through the PI3K/Akt pathway.¹⁴

In this study we sought to determine whether the NRF2/ARE pathway is activated in our mouse model of glaucoma. We used Nrf2^fl/fl mice and novel AAVs targeting either RGCs or astrocytes to deliver Cre to each cell type separately. We then quantified functional and molecular changes in both groups after IOP elevation. Finally, we demonstrate the efficacy of AAV2/2.Nrf2 gene therapy.

Construct development and AAV production:

AAV.ARE reporter: Synthetic DNA gBlocks were purchased from IDT (Integrated DNA Technologies, Coralville, IA) and contained either wild-type human thioredoxin (Trx) ARE or mutant Trx ARE (oligonucleotide M4)²² positioned upstream of the mini-TK promoter. The ARE promoters drive expression of nuclear-targeted tdTomato²³ and the SV40 promoter drives expression of HA-tagged ZsGreen (Takara Bio, San Jose, CA). All components were cloned into an AAV2 backbone (Cell Biolabs, San Diego, CA). The wild-type Trx plasmid was packaged into AAV2 at SignaGen (Fredrick, MD).

AAV.Sncg.Cre, AAV.Vim.Cre and respective control constructs: AAVs were generated that express Cre recombinase and tdTomato. The promoters included either the 1 kb human vimentin (Vim) promoter that was derived from Addgene plasmid #29114 (a gift from Elizabeth Simpson)²⁴ or the 0.66 kb mouse gamma-synuclein (Sncg) promoter (a gift from Yang Hu)²⁵. Astrocytes were targeted with the combination of Vim promoter and a modified AAV6 capsid (ShH10 with an additional Y455F mutation). RGCs were targeted with the combination of Sncg promoter and AAV2 capsid. Control AAVs contained tdTomato without Cre. AAVs were packaged in-house using triple transfection of HEK cells.

AAV.CMV.eGFP and AAV.CMV.Nrf2 constructs: pAAV.CMV.Nrf2 was purchased from Addgene (Watertown, MA; plasmid #67636) and packaged into AAV2/2 at SignaGen (Fredrick, MD). pAAV.CMV.eGFP was purchased from Addgene (plasmid #67634) and packaged into AAV2/2 at SignaGen.

Table 1

Plasmids used to produce AAVs in this study.
Plasmid	AAV nomenclature
pAAV.pTrx-ARE(WT)-TnSV0HA-zG	AAV2/2.ARE
pAAV.pCMV.Nrf2	AAV2/2.Nrf2
pAAV.pCMV.eGFP	AAV2/2.eGFP
pAAV.pSncg.Cre.IRES.tdTomato	AAV2/2.Sncg.Cre
pAAV.pVimentin.P2A.tdTomato	AAV2/6m.Vim.Cre
pAAV.pSncg.IRES.tdTomato	AAV2/2.Sncg.tdTom
pAAV.pVimentin.P2A.tdTomato	AAV2/6m.Vim.tdTom

ARPE-19 cells: ARPE-19 cells were purchased from ATCC (Manassas, VA) and grown as previously described.²⁶ Cells in complete media were transfected with plasmid DNA mixed with FuGENE HD (Promega, Madison, WI; #E2311) at a ratio of 1:4 (1 mg DNA: 4 ml FuGENE). Eight-well chamber slides and 6-well plates were transfected with 200 ng and 1 mg DNA per well, respectively. One day after transfection, the media was replaced with serum-free (SF) medium and incubated for 24 h. Cells were treated with SF medium containing either 5 mM sulforaphane (MedChemExpress, Monmouth Junction, NJ; #HY-13755) or vehicle (0.025% DMSO) and incubated for an additional 24 h. Cells in chamber slides were fixed with Histochoice, washed with PBS, and coverslips mounted with ProLong Gold (Thermo Fisher). For immunoblots, 10 mg of PBS-soluble protein was analyzed per lane. Blots were probed with rabbit anti-RFP (Rockland Immunochemicals, Inc., Limerick, PA; #600-401-379) and mouse anti-HA (Cell Signaling Technologies, Danvers, MA; #2367).

Mice

C57Bl/6 J (Jackson Labs, Bar Harbor, ME), B6.129X1-Nfe2l2^tm1Ywk/J or C58BL/6-Nfe2l2^tm1.Sred/SbisJ mice (Jackson Labs, Bar Harbor, ME) were group-housed, maintained on a 12-h light-dark cycle, and provided food and water ad libitum. An equal distribution of 2–3 month old male and female mice were used for this project.

Microbead Occlusion

IOP was bilaterally elevated using the well-characterized MOM ^27–29. We injected 2µl of 15-µm diameter FluoSpheres polystyrene microbeads into the anterior chamber of anesthetized mice (Thermo Fisher, Waltham, MA) as previously described^27–29. Additional mice received bilateral injections of an equivalent volume of lactated Ringer's saline solution as controls. Briefly, 1.5 mm outer diameter/1.12 mm inner diameter filamented capillary tubes (World Precision Instruments, Sarasota, FL) were pulled using a P-97 horizontal puller (Sutter Instrument Company, Novato, CA), and the resulting needles were broken using forceps to an inner diameter of ~ 100 µm. Microbeads were loaded and injected using a microinjection pump (World Precision Instruments, Sarasota, FL). Mice were anesthetized with isoflurane and dilated using topical 1% tropicamide ophthalmic solution (Patterson Veterinary, Devens, MA), and 2 µl (~ 2,000 microbeads) were injected. The needle was maintained in the injection site for 20 seconds before retraction to reduce microbead efflux. Mice were given topical 0.3% tobramycin ophthalmic solution (Patterson Veterinary, Devens, MA) following injection.

IOP measurements

IOP was measured immediately prior to microbead injection and biweekly thereafter using the Icare TonoLab rebound tonometer (Colonial Medical Supply, Franconia, NH) as previously described^27,28,30. Mice were anesthetized using isoflurane, and 10 measurements were acquired from each eye within 2 minutes of induction of anesthesia.

AAV injections

For experiments with an endpoint of 5 weeks post-IOP elevation, viruses were intravitreally injected one week prior to MOM injections. For experiments with an endpoint of 2 weeks post-IOP elevation, viruses were intravitreally injected two weeks prior to MOM injections. All vectors used in this study were injected with 1ul of virus solution at a concentration of 1 x 10⁹GC/ul.

In vivo electrophysiology

Mice were dark adapted overnight, dilated with 1% tropicamide for 10 minutes and anesthetized with 20/8/0.8 mg/kg ketamine/xylaxine/urethane according to previously published methodology.^36–39 Mice were placed on the heated surface of the ERG system to maintain body temperature. Corneal electrodes with integrated stimulators (Celeris System, Diagnosys LLC, Lowell, MA) were placed on eyes that were lubricated with GenTeal drops. Subdermal platinum needle electrodes were placed in the snout and back of the head at the location of the visual cortex. A ground electrode was placed in the back of the mouse. For VEPs, mice were exposed to 50 flashes of 1Hz, 0.05 cd.s/m² white light with a pulse frequency of 1 flash. For ERGs, mice were exposed to flashes of 1 Hz, 1 cd.s/m² white light with a pulse frequency of 1. For photopic negative ERGs (PhNR), mice were exposed to 20 continuous flashes of white light on a green background with a pulse frequency of 2. Each experimental group had 12–16 eyes.

Dihydroethidum Fluorescence

A dye that fluoresces in the presence of superoxide and, to a lesser extent, hydrogen peroxide, dihydroethidum (DHE), was utilized for these studies as previously described ³¹. Mice were anesthetized with 2.5% isofluorane and intravitreally injected with 1 µl (0.5uM) of DHE (ThermoFisher Scientific, Waltham, MA) diluted in phosphate-buffered saline (PBS) using a 30-gauge Hamilton syringe. Just prior to imaging, mice were anesthetized with ketamine/xylazine and eyes were dilated with 1% tropicamide. Thirty minutes after DHE injection, fluorescence was imaged on a Micron IV retinal imaging microscope (Phoenix Research Labs, Pleasanton, CA) using an FF02-475/50 nm excitation filter (Semrock, Inc. Rochester, NY) and ET620/60X emission filter (Chroma Technology Corp., Bellows Falls, VT). The average intensity of the fluorescence throughout the retina was quantified using ImageJ (Rasband, W.S., 2018). For each experimental group, 6–8 eyes were analyzed. For purposes of comparison, the DHE fluorescence measured in wildtype mice at 2-weeks (wks) (shown in Fig. 2) was also used in Fig. 6 to compare to Nrf2 KO mice. Notably, the DHE fluoresence was measured for each group during the same imaging session.

Tissue collection

For western blots and qPCR, retinas were collected and flash frozen from mice euthanized by anesthetic overdose and cervical dislocation. For immunohistochemistry and optic nerve histology, tissue was collected and incubated in 4% paraformaldehyde until use at 4^oC.

Protein assay: Protein concentrations were determined from 10 µl of retina homogenates with the Pierce BCA Protein Assay Kit (cat#: 23225, ThermoFisher Scientific, Waltham, MA). BSA was used as the protein standard. Absorbance was measured with the plate reader SpectraMax M2 (Molecular Devices, San Jose, CA).

Western blot: Single retinas were sonicated in lysis buffer (PBS, EDTA and Halt protease inhibitor) and centrifuged for 30 minutes at 4^oC. 4x Laemmli buffer (Bio-rad, cat# 1610747) containing ß-mercaptoethanol was added to the samples and heated for 5 minutes at 95^oC. Known amounts of protein (10–20 µg/retina) or protein ladder (cat#1610375, Bio-rad, Hercules, CA) were loaded in 4–20% polyacrylamide gels (Bio-Rad #456–1095). Proteins were transferred onto nitrocellulose using the Bio-Rad trans blot turbo transfer system. Membranes were blocked in 2% BSA in TBS overnight at 4^oC. Membranes were incubated in primary antibody (see Table 2) at room temperature with rocking for 2 hours. Membranes were incubated with secondary antibody (IRDye 800CW Donkey anti-rabbit, #926-32213 or IRDye 680CW Donkey anti-mouse, #926-68022,1:5000 in 1% BSA/TBS) at room temperature for 1 hour. After washing, blots were imaged with a Bio-Rad ChemiDoc system. Band density was quantified by scanning the blot using Adobe Photoshop. Each band was selected with the same frame and set measurements were used to obtain the gray mean value for each. Band intensity measurements from protein of interest were divided by band intensity measurements of loading control (b-actin). Each experimental group consisted of 5 retinas.

Quantitative PCR: Retinas were extracted from euthanized mice and placed immediately onto dry ice and stored at ⁻80^oC until homogenized by hand using 1.5ml-capacity pestles (cat#46C911, Grainger, Nashville, TN). RNA was extracted using a Qiagen RNeasy kit (Valencia, CA) as previously described.³² RNA concentration and purity were measured on a spectrophotometer. First-strand complementary DNA (cDNA) was synthesized from 250 ng of RNA from each sample using the Superscript III First-Strand synthesis system and oligo-dT20 primers (Invitrogen, Waltham, MA). Quantitative PCR (qPCR) was performed using Power SYBR green master mix (Applied Biosystems, Waltham, MA). All primer sequences were obtained from previous studies; we assessed the following: Prdx6, Gpx1, Ho-1 and Sod3 (see Table 3). All qPCR was performed in triplicate using an Applied Biosciences 7300 real-time PCR system (Waltham, MA). The amplification threshold was set using system software. Relative changes in gene expression were determined using Actin as the internal control. Each experimental group had 4–5 retinas.

Immunohistochemistry: Eyes were embedded in paraffin and sectioned at 10 microns according to previously published methods using the Vanderbilt Vision Research Center histology core^21,33,34. Slides were then warmed on a slide warmer at a medium setting (about 40 ^oC) for 30 min. Slides were then placed in a rack and went through a series of deparaffinization steps: xylene (10 min), 100% ethanol (10 min), 100% ethanol (5 min), 95% ethanol (5 min), 80% ethanol (5 min), 60% ethanol (5 min), 40% ethanol (5 min). Slides were then placed in coplin jar covered with sodium citrate solution and boiled for 30 min (2.94g of tri-sodium citrate dehydrate in 1L of DI water, adjusted to pH of 6.0 and then added 0.5ml of Tween 20). Following boiling, slides were washed twice in 1x PBS for 5 min. Then, slides incubated in sodium borohydride solution (0.05g sodium borohydride dissolved in 50ml DI water, made fresh every time) at room temperature. Slides were then placed in blocking buffer (500mL 1x PBS, 1.25mL Triton-X, 1.25mL Tween 20, 0.5g sodium citrate, 11.25g glycine, 5g BSA) and 5% normal donkey serum (cat #: D9663, Millipore Sigma, Darmstadt, Germany) for 1 hr at room temperature. Slides were washed once with 1xPBS and placed in primary antibody diluted in staining buffer (500mL 1x PBS, 1.25mL Triton X, 1.25 mL Tween 20, 5g BSA) overnight at 4^oC in a humidified chamber. The following day, slides were twice washed with 1x PBS for 5 min each. Secondary antibody was diluted in staining buffer and was added to the slides for 2 hrs at room temperature at 1:200 dilution after spinning for 10 min at 13,000g. After 2 hrs, slides were washed twice in 1x PBS for 5 min each. Then, slides were coverslipped with Vectashield containing DAPI (cat#: H-1200-10, Vector Laboratories, Burlingame, CA) and sealed with nail polish. Slides were imaged on a Nikon Eclipse epifluorescence microscope (Nikon, Melville, NY). All images were collected from the same retinal region with identical magnification, gain and exposure settings. Fluorescence intensity was quantified via ImageJ as previously described^21,33. A rectangle was selected around the region of interest, channels were split for multiple antibodies, threshold was adjusted, noise was de-speckled and fluorescence intensity was measured. Fluorescence intensity was normalized to saline-injected mice. Each experimental group included 5 eyes.

Optic nerve counts: Optic nerves were post-fixed in glutaraldehyde followed by Resin 812 embedding and Araldite 502 (cat#: 14900 and 10900 respectively, Electron Microscopy Sciences, Hatfield, PA) according to previously published protocols^14,34−36. Leica EM-UC7 microtome was used to collect 1 mm thick sections of the optic nerves. Sections were then stained with 1% paraphenylenediamine and 1% toluidine blue and were imaged on a Nikon Eclipse Ni-E microscope using 100x oil immersion objective (Nikon Instruments, Melville, NY). The optic nerves were montaged into a 5 x 5 image using the Nikon Elements software to scan a large image. We used the Counting Array and Better Cell Counter plugins to ImageJ, which creates a grid of nine squares overtop the montaged optic nerve. We manually counted healthy and degenerating axons, which are color-coded by the plugins. Degenerative axon profiles were identified by dark paraphenylenediamine staining due to collapsed myelin or loose myelin (onioning) surrounding the axon. A grid was used to avoid bias, by always counting in the same squares, using a cross configuration. Twenty percent of the optic nerve cross-sectional area was counted and the total was multiplied by five to estimate total and degenerating axons within the nerve. Each experimental group included 4–5 nerves.

Data Analysis

All statistical analyses were performed using GraphPad Prism software (La Jolla, CA). A one-way ANOVA with a Bonferroni post hoc test (a = 0.05) was used to analyze western blot quantification, IHC fluorescence quantification, ON quantification data, and ERG/VEP latencies and amplitudes. A one-way ANOVA and Dunnett’s multiple comparisons post hoc test (a = 0.05) were used to analyze the qPCR results. Means and standard deviation were calculated for each data set.

Table 2

Antibodies used in this study
Antibody	Company	Catalog Number	Species	Dilution for western blot	Dilution for IHC
Nrf2	Abcam	137550	Rabbit	1:1000	1:200
pNrf2	ThermoFisher	PA5-67520	Rabbit	1:1000	1:200
ß-actin	Cell Signaling	E4D9Z	Mouse	1:1000	N/A
RFP (for tdTomato)	ThermoFisher	MA5-15257	Mouse	1:200	1:200
Prdx6	Abcam	133348	Rabbit	1:1000	N/A
Gpx1	ThermoFisher	PA5-26323	Rabbit	1:500	N/A
SOD3	Abcam	80946	Rabbit	1:1000	N/A
ß-tubulin	Sigma	T8678	Mouse	N/A	1:300

Table 3

qPCR primers used in this study
Gene	Forward Primer	Reverse Primer
Prdx6	TTG ATG ATA AGG GCA GGG AC	CTA CCA TCA CGC TCT CTC CC
Gpx1	GGTTCGAGCCCAATTTTACA	CCCACCAGGAACTTCTCAAA
Ho-1	CCTTCCCGAACATCGACAGCC	GCAGCTCCTCAAACAGCTCAA
SOD3	AGGTGGATGCTGCCGAGAT	TCCAGACTGAAATAGGCCTCAAG

Overexpression of Nrf2 in GCL neurons in Nrf2 KO mice

We previously demonstrated that Nrf2 KO mice with elevated IOP lack an endogenous retinal antioxidant response and exhibit earlier onset axon degeneration and visual dysfunction²¹. To determine if overexpression of Nrf2 only in the GCL neurons would be sufficient to activate the antioxidant response, we intravitreally injected AAV2/2.Nrf2 or AAV2/2.eGFP into Nrf2 KO mice. This caused an expected increase in total NRF2 levels in the retina (Fig. 1A). As expected, DHE fluorescence, an in vivo indicator of superoxide levels, was elevated in AAV2/2.eGFP treated Nrf2 KO mice regardless of IOP elevation (p = 0.355; n = 3 eyes/group; Fig. 1B, C)²¹. Treatment with AAV2/2.Nrf2 decreased DHE fluorescence regardless of IOP elevation (Fig. 1B, C). In both the saline-injected control mice and microbead-injected mice, treatment with AAV2/2.Nrf2 increased PRDX6 and GPX1 levels (p = 0.0276 and p = 0.0141 respectively, for saline-injected controls; p = 0.0121 and p = 0.1376, respectively for microbead-injected mice; n = 4–5 retinas/group; Fig. 1D, E). There was no difference in SOD3 levels between the saline and microbead injected groups (data not shown).

We also assessed if transduction of the RGCs with AAV2/2.Nrf2 was sufficient to decrease glaucoma pathology. We previously published that the PhNR and VEP amplitudes were reduced in Nrf2 KO mice and in wildtype microbead-injected mice compared to their relevant controls¹⁴. Here we found that the AAV2/2.Nrf2 treated, saline injected Nrf2 KO mice had a greater PhNR amplitude than the AAV2/2.eGFP treated saline group (n = 6 or 8 eyes/group respectively; p = 0.0357; Fig. 1F, G). AAV2/2.Nrf2 treated, microbead injected mice also had a greater PhNR amplitude than AAV2/2.eGFP treated, microbead injected mice (n = 6 or 8 eyes/group respectively; p = 0.039; Fig. 1F, G). There was no difference in PhNR amplitude between saline or microbead injected mice from either group¹⁴. Treatment with AAV2/2.Nrf2 preserved the VEP N1 amplitude in Nrf2 KO mice injected with saline or microbeads (p = 0.0015 and p = 0.0146, respectively; n = 6 or 8 eyes/group respectively; Fig. 1F, H).

At 2 weeks post-IOP elevation, AAV2/2.eGFP Nrf2 KO mice had fewer axons than their saline-injected controls (p = 0.0054; n = 4 nerves/group; Fig. 1I, J), in agreement with our previous findings¹⁴. Further, AAV2/2.Nrf2 treatment preserved the optic nerve axons after IOP elevation. AAV2/2.Nrf2 microbead injected Nrf2 KO mice showed no additional axon loss when compared to AAV2/2.Nrf2 saline injected Nrf2 KO mice (p = 0.1304; n = 3–4 nerves/group; Fig. 1I, J). Microbead-injected, AAV2/2.eGFP treated Nrf2 KO mice had fewer total axons and more degenerative axons in comparison to AAV2/2.Nrf2 treated Nrf2 KO mice (p = 0.0263 and p = 0.009 respectively; n = 3–4 nerves/group; Fig. 1I-K). The same trend occurred for the microbead-injected Nrf2 KO mice: AAV2/2.Nrf2 treated mice had more axons and fewer degenerative axons in comparison to the AAV2/2.eGFP mice (p < 0.0001 and p = 0.001 respectively; n = 3–4 nerves/group; Fig. 1I, K). There was no difference in the number of degenerative axons between the AAV2/2.Nrf2 injected groups (p = 0.2191; n = 3–4 nerves/group; Fig. 1I, K).

Activation of the ARE in RGCs at 2 weeks post-IOP elevation

As an independent approach to determine if elevated IOP can activate the ARE in the RGCs, we designed an ARE reporter construct and a control with an inactive ARE (M4) (Fig. 2A, B). We tested the constructs in ARPE19 cells and quantified fluorescence after exposure to sulforaphane, an NRF2 inducer (Fig. 2C). ARE activation, based on increased TdTomato fluorescence, was detected in transfected cells treated with sulforaphane, but not in those that did not receive sulforaphane nor those that were transfected with the plasmid carrying the inactive M4 ARE (Fig. 2C). Cells transfected with WT ARE and treated with either DMSO or sulforaphane produced both tdTomato and ZsGreen, with more intense tdTomato in the cells treated with sulforaphane (Fig. 2D). However, in the ARPE19 cells transfected with M4 ARE, there was no tdTomato band (Fig. 2D).

We then packaged the ARE reporter construct into AAV2 and intravitreally injected the vector into wildtype mice two weeks prior to inducing elevated IOP. tdTomato fluorescence was increased in both the 1-week and 2-week post-IOP elevation group in comparison to saline-injected controls (p = 0.0015 and p = 0.0025, respectively; n = 4 or 5 eyes/group; Fig. 2E, F). Ultimately, the 2-week post-IOP elevation group had more tdTomato fluorescence than the 1-week post-IOP elevation group (p = 0.029; n = 4 eyes/group). The greater ARE activation at 2-weeks matched our previously findings of peak activation of the endogenous antioxidant response (NRF2/ARE pathway).¹⁴ Ultimately, the use of this construct in vivo illustrated that the endogenous antioxidant response in RGCs to increased IOP can be elicited by the RGCs.

Knockdown of NRf2 in either RGCs or astrocytes decreases endogenous antioxidant response

In order to explore which cell type, RGCs or astrocytes, were responsible for the endogenous antioxidant response after IOP elevation, we knocked down Nrf2 in each cell type separately. Intravitreal injection of AAV2/2.pSncg.Cre.IRES.tdTomato (AAV2/2.Sncg.Cre; Fig. 3A) into Nrf2^fl/fl mice resulted in RGC-specific expression (Fig. 3B). Intravitreal injection of AAV2/6m.pVim.Cre.p2A.tdTomato (AAV2/6m.Vim.Cre; Fig. 3A) in Nrf2^fl/fl mice resulted in astrocyte-specific expression (Fig. 3D). While vimentin is also expressed in Müller glia in the retina^35–38, we validated that the vector primarily transduced astrocytes rather than Müller glia by assessing morphology and localization of GFAP and tdTomato positive cells in retina cross-sections (Fig. 3D). Controls were Nrf2^fl/fl mice injected with the same serotype AAV and promoters, delivering tdTomato instead of Cre.

To confirm that the vectors decreased NRF2 activation, we quantified the ratio of phosphorylated to total NRF2 (Fig. 3F,G). There was less phosphorylated NRF2 in both the AAV2/2.Sncg.Cre and the AAV2/6m.Vim.Cre treated mice than in their respective controls (p = 0.0233 and p = 0.0035, respectively; n = 5 retinas/group).

We then quantified levels of ARE-driven transcripts that we detected previously as being elevated during the endogenous antioxidant response¹⁴. We found decreased Prdx6 in RGC-specific Nrf2 knockdown mice in comparison to astrocyte-specific Nrf2 knockdown mice, normalized to tdTomato controls (p = 0.0294; n = 4 or 5 retinas/group respectively; Fig. 3E). There was no difference in Gpx1, Sod3, or Ho-1 levels if Nrf2 is knocked down in either astrocytes or RGCs (p = 0.0607, p = 0.6651, and p = 0.533, respectively; n = 4 or 5 retinas/group).

We also quantified the endogenous antioxidant response at the protein level (Fig. 3F, H-J). There was less PRDX6 levels in the AAV2/2.Sncg.Cre retinas in comparison to AAV2/2.Sncg.tdTomato controls (p = 0.0395; n = 5 retinas/group; Fig. 3F, H). There was no significant difference in PRDX6 levels between the AAV2/6.Vim.tdTomato and AAV2/6.Vim.Cre groups (p = 0.1036; n = 5 retinas/group). There was no statistically significant difference in GPX1 levels among any of the groups, however, the AAV2/2.Sncg.Cre group had trending decreases in GPX1 levels (p = 0.0988; n = 5 retinas/group; Fig. 3F, I). There was less SOD3 in both the RGC-specific and the astrocyte-specific knockdown groups in comparison to their respective controls (p = 0.0415 and p = 0.049, respectively; n = 4 retinas/group; Fig. 3F, J). Ultimately, knockdown of Nrf2 in either RGCs or astrocytes is sufficient to dampen the retina’s endogenous antioxidant response to elevated IOP.

Knockdown of Nrf2 in either RGCs or astrocytes decreases visual function and accelerates axon degeneration

Nrf2 knockdown in both the RGCs and the astrocytes decreased PhNR amplitude compared to respective tdTomato controls (p = 0.0153 and p < 0.0001, respectively; n = 16 eyes/group; Fig. 4A, B). There was no difference in PhNR amplitude between the two Nrf2 knockdown groups at 2 weeks post-IOP elevation (p = 0.226; Fig. 4A, B). At 5 weeks post-IOP elevation, there was no difference in PhNR amplitude between the control and AAV2/6.Vim.Cre groups (p = 0.5182, data not shown). At 5 weeks, RGC-specific knockdown of Nrf2 caused a further reduction in PhNR amplitude than controls (data not shown).

There were no differences in VEP amplitude between any of the experimental groups at 2-weeks post-IOP elevation in agreement with previous findings from our lab and others (data not shown)^21,27,29. At 5-weeks post-IOP elevation, there was an equal decrease in VEP amplitude in the mice injected with either AAV2/2.Sncg.tdTomato or AAV2/2.Sncg.Cre (p = 0.8786; n = 16 eyes/group; Fig. 4A, C). Mice injected with AAV2/6m.Vim.Cre had a decreased VEP N1 amplitude in comparison to controls (p = 0369; n = 14 eyes/group; Fig. 4A, C).

We previously detected earlier and greater axon degeneration in total Nrf2 KO mice after IOP elevation¹⁴. At 2-weeks post-IOP elevation, AAV2/2.Sncg.Cre injected mice had fewer total axons and more degenerative axons than AAV2/2.Sncg.tdTomato controls (p < 0.0001 and p = 0.0003, respectively; n = 5 nerves/group; Fig. 4D-F). The same pattern occurred with the astrocyte-specific Nrf2 knockdown group: AAV2/6.Vim.Cre injected mice had fewer total axons and more degenerative axons than AAV2/6.Vim.tdTomato controls (p < 0.0001 and p < 0.0001; n = 5 nerves/group; Fig. 4D-F). These data show that even a cell-type specific knockdown of Nrf2 in either neurons or astrocytes accelerated and exacerbated axon degeneration following MOM at 2 weeks post-IOP elevation.

At 5 weeks post-IOP elevation, axon degeneration in the optic nerve was even more severe in all groups (Fig. 4D, G-H). AAV2/2.Sncg.Cre mice had fewer total axons and more degenerative axons than their AAV2/2.Sncg.tdTomato controls (p < 0.0001 for both measures; n = 5 nerves/group; Fig. 4D, G-H). AAV2/6.Vim.Cre mice also had fewer total axons and more degenerative axons than their AAV2/6.Vim.tdTomato controls (p = 0.001 and p = 0.0002, respectively; n = 5 nerves/group; Fig. 4D, G). There was no difference in the total number of axons between the knockdown groups (p = 0.4538, n = 5 nerves/group; Fig. 4D, G).

Overexpression of Nrf2 in GCL neurons increases endogenous antioxidant response of the retina at 2 weeks and 5 weeks post-IOP elevation

We sought to determine if overexpression of NRF2 in GCL neurons in wildtype mice would enhance the retina’s endogenous response to elevated IOP. Treatment with AAV2/2.Nrf2 increased total retina levels of NRF2 in both saline and microbead-injected mice (p = 0.0028 and p = 0.0184; n = 5 retinas/group; Figs. 5A,B). We confirmed localization of the increased NRF2 in the GCL neurons by co-labeling with NeuN (Fig. 5C). DHE fluorescence was increased in the microbead-injected AAV2/2.eGFP mice in comparison to their saline-injected controls (p = 0.0012; n = 5 eyes/group; Fig. 5D, E). Treatment with AAV2/2.Nrf2 prevented the IOP-induced increase in DHE fluorescence

(p < 0.0001; n = 5 eyes/group; Fig. 5D, E).

Levels of antioxidants at the mRNA and protein level were increased by treatment with AAV2/2.Nrf2. Prdx6, Nrf2, Gpx1 and Txn1 were increased after elevation of IOP and treatment with AAV2/2.Nrf2 in comparison to the AAV2/2.eGFP controls (p < 0.0001, p = 0.0035, p = 0.0467, and p = 0.0237, respectively; n = 5 retinas/group; Fig. 5F). This was above and beyond the endogenous increase in these antioxidant proteins we previously reported²¹. We also quantified antioxidants on the protein level (Fig. 5G, H). In AAV2/2.eGFP controls at 2 weeks post-IOP elevation, PRDX6, GPX1 and SOD3 were increased in microbead-injected mice in comparison to saline-injected controls (p = 0.0364, p = 0.0377, and p = 0.014 respectively; n = 5 retinas/group; Fig. 5G, H). This was consistent with our previous studies, which indicated that the retina endogenously responds to increased ROS by upregulating expression of these proteins²¹. Mice that received AAV2/2.Nrf2 had an even greater increase in these antioxidant proteins in both saline and microbead groups than AAV2/2.eGFP controls. PRDX6, GPX1 and SOD3 levels were increased in AAV2/2.Nrf2 treated saline-injected controls in comparison to their AAV2/2.eGFP treated counterparts (p = 0.0043, p = 0.0124, and p = 0.0032, respectively; n = 5 retinas/group). Similarly, PRDX6, GPX1 and SOD3 levels were all increased in AAV2/2.Nrf2 treated microbead-injected mice in comparison to AAV2/2.eGFP controls (p = 0.0153, p = 0.0096 and p = 0.0013, respectively; n = 5 retinas/group; Fig. 5G, H). In the 5-week post-IOP elevation group, there was no difference in GPX1 or SOD3 levels in any of the experimental groups, although both were increased compared to the AAV2/2.eGFP controls (data not shown). There was no difference in PRDX6 levels between saline or microbead AAV2/2.Nrf2 mice (n = 5 retinas/all groups; Fig. 5J, K).

At the 5-week timepoint, in AAV2/2.eGFP mice, DHE fluorescence was increased in microbead-injected mice in comparison to saline-injected controls (p = 0.0112; n = 5 eyes/group; Fig. 5I). In contrast, both saline and microbead injected AAV2/2.Nrf2 mice had similar DHE levels as the saline-injected AAV2/2.eGFP mice (p = 0.1783; n = 6 eyes/group; Fig. 5I). Overexpression of Nrf2 in GCL neurons of WT mice protects against IOP-induced increased ROS and increased the retina’s antioxidant response at both 2 and 5 weeks post-IOP elevation.

Overexpression of Nrf2 in GCL neurons protects against visual function deficits and slows axon degeneration at 2 and 5 weeks post-IOP elevation

At 2-weeks post-IOP elevation, in AAV2/2.eGFP mice, as we have previously published, the PhNR amplitude was decreased in comparison to their saline-injected controls (p = 0.0043; n = 14 eyes/group; Fig. 6A). Overexpression of Nrf2 in GCL neurons using AAV2/2.Nrf2 prevented this decrease (p = 0.922; n = 16 eyes/group; Fig. 6A). There were no differences in the latency of the PhNR in any of the experimental groups (data not shown). As expected, that there were no differences in VEP N1 amplitudes at 2 weeks post-IOP elevation (p = 0.9707; n = 16 eyes/group; Fig. 6B).

At 5-weeks post-IOP elevation, the microbead-injected AAV2/2.eGFP mice had a lower amplitude of the PhNR in comparison to their saline-injected controls (p = 0.0043; n = 10 eyes/group; Fig. 6F). Microbead-injected AAV2/2.Nrf2 mice also had a lower amplitude than their saline-injected controls (p = 0.0325; n = 12 eyes/group; Fig. 6F). However, AAV2/2.Nrf2 still preserved the amplitude of the PhNR better than AAV2/2.eGFP controls; microbead-injected AAV2/2.Nrf2 mice had greater amplitudes than their AAV2/2.eGFP controls (p = 0.0232; n = 10 or 12 eyes/group respectively; Fig. 6F). This pattern remained among the saline-injected groups (p = 0.0006; n = 10 or 12 eyes/group; Fig. 6F). There was no significant difference in the PhNR latencies of any of the four experimental groups (data not shown). In AAV2/2.eGFP mice, VEP N1 amplitude was decreased in the microbead-injected mice in comparison to their saline-injected controls (p = 0.0007; n = 10 eyes/group; Fig. 7G), while there was no difference in VEP N1 amplitude in either group injected with AAV2/2.Nrf2 (p = 0.606; n = 12 eyes/group; Fig. 6G). Additionally, the VEP N1 amplitudes of the AAV2/2.Nrf2 MOM group were greater than the amplitudes of the AAV2/2.eGFP MOM group (p = 0.016; n = 10 or 12 eyes/group respectively; Fig. 6G).

Both AAV2/2.eGFP and AAV2/2.Nrf2 treated, microbead-injected groups had fewer total axons than their respective saline-injected controls (p = 0.0335 and p = 0.0011; respectively; n = 5 nerves/group; Fig. 6H-I). Still, the total number of axons in the microbead-injected AAV2/2.Nrf2 group was greater than the microbead-injected AAV2/2.eGFP group (p = 0.0491; n = 5 nerves/group; Fig. 6H, I). Additionally, even the AAV2/2.eGFP treated, saline-injected mice had more degenerative axons than AAV2/2.Nrf2 treated, saline-injected mice (p = 0.0316; n = 5 nerves/group). AAV2/2.eGFP treated, microbead-injected mice had more degenerative axons than AAV2/2.Nrf2 treated, microbead-injected mice (p = 0.0005; n = 5 nerves/group; Fig. 6H, I).

We previously discovered that an Nrf2-mediated endogenous antioxidant response of the retina delays the onset of glaucomatous neurodegeneration.¹⁴ In this study, we sought to elucidate the roles of the RGCs and the astrocytes in this response and to assess the efficacy of AAV2/2.Nrf2 as a neuroprotective treatment.

Supplementation of Nrf2 in GCL neurons in the Nrf2 KO background activated the Nrf2/ARE pathway and decreased vision loss and pathology. Unfortunately, most of the vision loss and pathology was due to lack of Nrf2 and not to glaucoma, as shown by equal deficits in the saline and microbead-injected Nrf2 KO groups treated with the control AAV. The one exception was the extent of axon degeneration, which was increased in the microbead-injected mice. Nrf2 supplementation in the RGCs was sufficient to overcome total retinal loss of Nrf2 in terms of ROS levels, endogenous antioxidant response and RGC health.

To determine if elevated IOP caused activation of the ARE in the RGCs, we transduced the cells with an ARE reporter construct. Activation of the ARE was detected, supporting the hypothesis that the RGCs contribute to the endogenous antioxidant response of the retina to elevated IOP. The increased tdTomato fluorescence occurred concurrently with the previously characterized activation of Nrf2 at 2 weeks post-IOP elevation¹⁴.

As another approach, we used different AAV serotype and promoter combinations to specifically deliver Cre to either RGCs or astrocytes in the Nrf2^fl/fl mouse. The use of AAV2/2 and the gamma synuclein promoter provided specific expression in the RGCs as previously reported²⁵. The use of the AAV6 capsid with tyrosine mutations and a 1kb vimentin promoter resulted in specific expression in the astrocytes, but not in Müller cells. Either RGC-specific or astrocyte-specific knockdown of Nrf2 resulted in a decreased endogenous antioxidant response and increased glaucomatous pathology as well as an earlier onset of this pathology. The greater decrease in the VEP in mice with Nrf2 knocked down in the astrocytes suggests an important role of these cells. Conversely, our results also show that the RGCs contribute to the endogenous antioxidant response and that increased Nrf2 in these cells can enhance this response and provide neuroprotection. The data suggests that both of these cells contribute to the response but the amplification of this pathway in RGCs is sufficient to compensate for lack of Nrf2 elsewhere. It is possible that activation of the Nrf2/ARE pathway is important in both astrocytes and RGCs for the IOP-dependent antioxidant response, or these two cell types communicate so substantially that this in vivo approach was insufficient to isolate effects from the RGCs or the astrocytes independently. Future in vitro experiments may be needed to separate the two potential explanations.

We found that Nrf2 overexpression in GCL neurons, beginning prior to IOP elevation, counteracts the increase in ROS and protects RGCs at both 2 and 5 weeks post-IOP elevation. Our findings agree with previous publications showing an essential role for Nrf2 in neurons’ antioxidant response^18,19,39,40. In the future, it would be interesting to determine the therapeutic window for Nrf2 gene therapy. Additionally, the efficacy of Nrf2 activators should be tested. Pharmacological activation of Nrf2 has been shown to be neuroprotective in a model of ischemia/reperfusion retinal injury.¹⁸ Four Nrf2 activators are currently being explored in clinical trials, which could be used in glaucoma studies in the future.⁴¹ Overexpression or chronic activation of Nrf2 could also have negative side effects. In Drosophila, prolonged activation of Nrf2 can actually shorten lifespan.⁴² Additionally, many activators of Nrf2 work by modifying Keap1 so that it cannot degrade Nrf2 via disruption of its cysteine residues—these activators are often nonspecific and can have off-target effects.^41,43 Therefore, future studies investigating the downstream responses of activation of the Nrf2/ARE pathway might yield more specific therapeutic targets. For example, our work suggests that PRDX6 and SOD3 might contribute significantly to the endogenous antioxidant response, and therefore, exogenously increasing levels of these proteins might be beneficial in delaying or decreasing glaucoma pathogenesis. Previous studies have shown that antioxidants are effective in preserving RGCs and visual function in models of glaucoma^11,44,45. This study further supports this approach and provides insights into the best antioxidants to pursue for clinical translation.

Overall, our study elucidates the importance for NRF2 in both RGCs and astrocytes for the retina’s endogenous antioxidant response to glaucoma-induced ROS. Additionally, we show that overexpression of Nrf2 in GCL neurons of wildtype mice is a viable therapeutic approach for axon degeneration, RGC function and other glaucomatous pathology. We conclude that overexpression of Nrf2 in RGCs while all surrounding cells do not have Nrf2 is not sufficient to mitigate inevitable glaucomatous pathology.

IOP: intraocular pressure

ROS: reactive oxygen species

NRF2: Nuclear factor-erythroid factor 2-related factor 2

ARE: antioxidant response element

RGC: retinal ganglion cell

AAV: adeno-associated virus

PhNR: photopic negative response

KEAP1: Kelch-like ECH associated protein 1

Ethical approval and consent to participate: All experiments were approved by the Institutional Animal Care and Use Committee of Vanderbilt University Medical Center.

Consent for publication: Not applicable

Availability of supporting data: All data generated and analyzed during this study are included in this published article and its supplementary information files.

Competing interests: The funding sources had no involvement in the study design, or collection, analysis, or interpretation of the data. The authors confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its outcome.

Funding: This work was supported by DoD W81XWH-15-1-0096 (TR), DoD W81XWH-17-2-0055 (TR), NEI R01 EY022349 (TR, DC), NEI R01 EY017427 (DC), NEI U24 EY029893 (TR), NEI U24 EY029903 (DC), Potocsnak Discovery Grant in Regenerative Medicine, Ayers Foundation Regenerative Visual Neuroscience Pilot Grant, Ret. Maj. General Stephen L. Jones, MD Fund, Stein Innovation Award (DC), Stanley Cohen Innovation Fund (DC), NEI P30EY008126 (VVRC), T32 EY021833 (VVRC) and Research Prevent Blindness, Inc Unrestricted Funds (VEI).

Authors’ contributions: SN designed all in vivo experiments with TR and performed all in vivo experiments, performed some IOP measurements with AH, analyzed all in vivo ROS, western blot, IHC, and qPCR data and wrote manuscript. JB constructed all plasmids and AAVs that were not purchased from AddGene/packaged at SignaGen, completed transfection of ARPE19 cells, and assisted with writing the manuscript. EA completed all microbead and intravitreal injections with SN. PG completed all retina and optic nerve processing and sectioning. AH and JA completed optic nerve axon degeneration analysis and AH assisted with performing IOP measurements. DC edited the manuscript and assisted with experimental design. TR designed experiments with SN and worked with SN to write manuscript. All authors read and approved the final manuscript.

Acknowledgements: The authors would like to acknowledge Lauren Wareham and Rachael Cheung for guidance and assistance to IHC protocol.

Authors’ information:

Sarah Naguib: ¹Department of Ophthalmology & Visual Sciences, Department of Neuroscience, Vanderbilt University School of Medicine, Nashville,

TN; [email protected];

Jon Backstrom: ²Vanderbilt Eye Institute, Vanderbilt University Medical Center, Nashville, TN;, [email protected]

Elisabeth Artis: ²Vanderbilt Eye Institute, Vanderbilt University Medical Center, Nashville, TN; [email protected]

Purnima Ghose: ²Vanderbilt Eye Institute, Vanderbilt University Medical Center, Nashville, TN; [email protected]

Ameer Haider: ¹Department of Ophthalmology & Visual Sciences, Department of Neuroscience, Vanderbilt University School of Medicine, Nashville, TN; [email protected]

John Ang: ¹Department of Ophthalmology & Visual Sciences, Department of Neuroscience, Vanderbilt University School of Medicine, Nashville, TN; [email protected]

David Calkins: ²Vanderbilt Eye Institute, Vanderbilt University Medical Center, Nashville, TN; [email protected]

Tonia Rex: ¹Department of Ophthalmology & Visual Sciences, Department of Neuroscience, Vanderbilt University School of Medicine, Nashville, TN; ²Vanderbilt Eye Institute, Vanderbilt University Medical Center, Nashville, TN; [email protected]

Calkins, D. J. Critical pathogenic events underlying progression of neurodegeneration in glaucoma. Progress in Retinal and Eye Research (2012). doi:10.1016/j.preteyeres.2012.07.001
Greco, A. et al. Emerging Concepts in Glaucoma and Review of the Literature. American Journal of Medicine 129, (2016).
Killer, H. E. & Pircher, A. Normal tension glaucoma: Review of current understanding and mechanisms of the pathogenesis /692/699/3161/3169/3170 /692/699/3161 review-article. Eye (Basingstoke) (2018). doi:10.1038/s41433-018-0042-2
Weinreb, R. N., Aung, T. & Medeiros, F. A. The pathophysiology and treatment of glaucoma: A review. JAMA - Journal of the American Medical Association (2014). doi:10.1001/jama.2014.3192
He, S., Stankowska, D. L., Ellis, D. Z., Krishnamoorthy, R. R. & Yorio, T. Targets of Neuroprotection in Glaucoma. J. Ocul. Pharmacol. Ther. 34, 85–106 (2018).
Goldberg, I. Relationship between intraocular pressure and preservation of visual field in glaucoma. Surv. Ophthalmol. 48, (2003).
Nita, M. & Grzybowski, A. The Role of the Reactive Oxygen Species and Oxidative Stress in the Pathomechanism of the Age-Related Ocular Diseases and Other Pathologies of the Anterior and Posterior Eye Segments in Adults. Oxidative Medicine and Cellular Longevity (2016). doi:10.1155/2016/3164734
Ghanem, A. A., Arafa, L. F. & El-Baz, A. Oxidative Stress Markers in Patients with Primary Open-Angle Glaucoma. Curr. Eye Res. 35, 295–301 (2010).
Chrysostomou, V., Rezania, F., Trounce, I. A. & Crowston, J. G. Oxidative stress and mitochondrial dysfunction in glaucoma. Current Opinion in Pharmacology 13, 12–15 (2013).
Zhao, J. et al. Oxidative stress in the trabecular meshwork (Review). International Journal of Molecular Medicine (2016). doi:10.3892/ijmm.2016.2714
Kansanen, E., Kuosmanen, S. M., Leinonen, H. & Levonen, A.-L. The Keap1-Nrf2 pathway: Mechanisms of activation and dysregulation in cancer $. (2013). doi:10.1016/j.redox.2012.10.001
Raghunath, A. et al. Antioxidant response elements: Discovery, classes, regulation and potential applications. Redox Biology 17, 297–314 (2018).
Nguyen, T., Huang, H. C. & Pickett, C. B. Transcriptional regulation of the antioxidant response element. Activation by Nrf2 and repression by MafK. J. Biol. Chem. 275, 15466–15473 (2000).
Tonelli, C., Chio, I. I. C. & Tuveson, D. A. Transcriptional Regulation by Nrf2. Antioxid. Redox Signal. 29, 1727–1745 (2018).
Cheng, J., Liang, J. & Qi, J. Role of nuclear factor (Erythroid-derived 2)-like 2 in the age-resistant properties of the glaucoma trabecular meshwork. Exp. Ther. Med. 14, 791–796 (2017).
Batliwala, S., Xavier, C., Liu, Y., Wu, H. & Pang, I. H. Involvement of Nrf2 in ocular diseases. Oxidative Medicine and Cellular Longevity (2017). doi:10.1155/2017/1703810
Nakagami, Y. Nrf2 Is an Attractive Therapeutic Target for Retinal Diseases. (2016). doi:10.1155/2016/7469326
Xu, Z. et al. Neuroprotective role of Nrf2 for retinal ganglion cells in ischemia-reperfusion. J. Neurochem. 133, 233–241 (2015).
Liu, X. F. et al. The Nrf2 signaling in retinal ganglion cells under oxidative stress in ocular neurodegenerative diseases. International Journal of Biological Sciences 14, 1090–1098 (2018).
Himori, N. et al. Critical role of Nrf2 in oxidative stress-induced retinal ganglion cell death. J. Neurochem. 127, 669–680 (2013).
Naguib, S., Backstrom, J. R., Gil, M., Calkins, D. J. & Rex, T. S. Retinal oxidative stress activates the NRF2/ARE pathway: An early endogenous protective response to ocular hypertension. Redox Biol. (2021). doi:10.1016/j.redox.2021.101883
Kim, Y.-C., Yamaguchi, Y., Kondo, N., Masutani, H. & Yodoi, J. Thioredoxin-dependent redox regulation of the antioxidant responsive element (ARE) in electrophile response. Oncogene 22, 1860–1865 (2003).
Campbell, E. A. et al. Structure of the bacterial RNA polymerase promoter specificity sigma subunit. Mol. Cell 9, 527–539 (2002).
Portales-Casamar, E. et al. A regulatory toolbox of MiniPromoters to drive selective expression in the brain. Proc. Natl. Acad. Sci. U. S. A. 107, 16589–16594 (2010).
Q, W. et al. Mouse γ-Synuclein Promoter-Mediated Gene Expression and Editing in Mammalian Retinal Ganglion Cells. J. Neurosci. 40, (2020).
Backstrom, J. R., Sheng, J., Wang, M. C., Bernardo-Colón, A. & Rex, T. S. Optimization of S. aureus dCas9 and CRISPRi Elements for a Single Adeno-Associated Virus that Targets an Endogenous Gene. Mol. Ther. - Methods Clin. Dev. 19, 139–148 (2020).
Sappington, R. M., Carlson, B. J., Crish, S. D. & Calkins, D. J. The microbead occlusion model: A paradigm for induced ocular hypertension in rats and mice. Investig. Ophthalmol. Vis. Sci. (2010). doi:10.1167/iovs.09-3947
Calkins, D. J., Lambert, W. S., Formichella, C. R., McLaughlin, W. M. & Sappington, R. M. The microbead occlusion model of ocular hypertension in mice. in Methods in Molecular Biology 1695, 23–39 (Humana Press Inc., 2018).
Bond, W. S. et al. Virus-mediated EpoR76E therapy slows optic nerve axonopathy in experimental glaucoma. Mol. Ther. (2016). doi:10.1038/mt.2015.198
Hines-Beard, J., Bond, W. S., Backstrom, J. R. & Rex, T. S. Virus-mediated EpoR76E gene therapy preserves vision in a glaucoma model by modulating neuroinflammation and decreasing oxidative stress. J. Neuroinflammation 13, 39 (2016).
Bernardo-Colon, A. et al. Antioxidants prevent inflammation and preserve the optic projection and visual function in experimental neurotrauma. Cell Death Dis. 9, (2018).
Hines-Beard, J., Bond, W. S., Backstrom, J. R. & Rex, T. S. Virus-mediated EpoR76E gene therapy preserves vision in a glaucoma model by modulating neuroinflammation and decreasing oxidative stress. J. Neuroinflammation 13, 1–13 (2016).
Naguib, S., Bernardo-Colón, A., Cencer, C., Gandra, N. & Rex, T. S. Galantamine protects against synaptic, axonal, and vision deficits in experimental neurotrauma. Neurobiol. Dis. 134, (2020).
Bricker-Anthony, C., Hines-Beard, J. & Rex, T. S. Molecular changes and vision loss in a mouse model of closed-globe blast trauma. Investig. Ophthalmol. Vis. Sci. 55, 4853–4862 (2014).
O’Leary, L. A. et al. Characterization of Vimentin-Immunoreactive Astrocytes in the Human Brain. Front. Neuroanat. 0, 31 (2020).
Fernández-Sánchez, L., Lax, P., Campello, L., Pinilla, I. & Cuenca, N. Astrocytes and Müller Cell Alterations During Retinal Degeneration in a Transgenic Rat Model of Retinitis Pigmentosa. Front. Cell. Neurosci. 0, 484 (2015).
J, S., WW, F. & M, S. Immunocytochemical demonstration of vimentin in astrocytes and ependymal cells of developing and adult mouse nervous system. J. Cell Biol. 90, 435–447 (1981).
MJ, P.-A. et al. Vimentin isoform expression in the human retina characterized with the monoclonal antibody 3CB2. J. Neurosci. Res. 86, 1871–1883 (2008).
Wang, M., Li, J. & Zheng, Y. The potential role of nuclear factor erythroid 2-related factor 2 (Nrf2) in glaucoma: A review. Medical Science Monitor 26, e921514-1 (2020).
Himori, N. et al. Critical role of Nrf2 in oxidative stress-induced retinal ganglion cell death. J. Neurochem. 127, 669–680 (2013).
Robledinos-Antón, N., Fernández-Ginés, R., Manda, G. & Cuadrado, A. Activators and Inhibitors of NRF2: A Review of Their Potential for Clinical Development. Oxidative medicine and cellular longevity 2019, 9372182 (2019).
Pitoniak, A. & Bohmann, D. Mechanisms and Functions of Nrf2 Signaling in Drosophila. Free Radic. Biol. Med. 88, 302 (2015).
Cuadrado, A. et al. Therapeutic targeting of the NRF2 and KEAP1 partnership in chronic diseases. Nat. Rev. Drug Discov. 2018 184 18, 295–317 (2019).
Hines-Beard, J., Bond, W. S., Backstrom, J. R. & Rex, T. S. Virus-mediated EpoR76E gene therapy preserves vision in a glaucoma model by modulating neuroinflammation and decreasing oxidative stress. J. Neuroinflammation (2016). doi:10.1186/s12974-016-0499-5
Inman, D. M., Lambert, W. S., Calkins, D. J. & Horner, P. J. α-Lipoic Acid Antioxidant Treatment Limits Glaucoma-Related Retinal Ganglion Cell Death and Dysfunction. PLoS One 8, e65389 (2013).

Download PDF

Version 1

posted

You are reading this latest preprint version

NRF2/ARE-mediated antioxidant response to glaucoma: Role of astrocytes and retinal ganglion cells

Status:

Version 1

Abstract

Figures

Background

Materials And Methods

Construct development and AAV production:

Results

Overexpression of Nrf2 in GCL neurons in Nrf2 KO mice

Activation of the ARE in RGCs at 2 weeks post-IOP elevation

Knockdown of NRf2 in either RGCs or astrocytes decreases endogenous antioxidant response

Knockdown of Nrf2 in either RGCs or astrocytes decreases visual function and accelerates axon degeneration

Discussion

Conclusions

Abbreviations

Declarations

References

Status:

Version 1