DDIT3 (CHOP) contributes to retinal ganglion cell somal loss but not axonal degeneration in DBA/2J mice

Glaucoma is an age-related neurodegenerative disease characterized by the progressive loss of retinal ganglion cells (RGCs). Chronic ocular hypertension, an important risk factor for glaucoma, leads to RGC axonal injury at the optic nerve head. This insult triggers molecularly distinct cascades governing RGC somal apoptosis and axonal degeneration. The molecular mechanisms activated by ocular hypertensive insult that drive both RGC somal apoptosis and axonal degeneration are incompletely understood. The cellular response to endoplasmic reticulum stress and induction of pro-apoptotic DNA damage inducible transcript 3 (DDIT3, also known as CHOP) have been implicated as drivers of neurodegeneration in many disease models, including glaucoma. RGCs express DDIT3 after glaucoma-relevant insults, and importantly, DDIT3 has been shown to contribute to both RGC somal apoptosis and axonal degeneration after acute induction of ocular hypertension. However, the role of DDIT3 in RGC somal and axonal degeneration has not been critically tested in a model of age-related chronic ocular hypertension. Here, we investigated the role of DDIT3 in glaucomatous RGC death using an age-related, naturally occurring ocular hypertensive mouse model of glaucoma, DBA/2J mice (D2). To accomplish this, a null allele of Ddit3 was backcrossed onto the D2 background. Homozygous Ddit3 deletion did not alter gross retinal or optic nerve head morphology, nor did it change the ocular hypertensive profile of D2 mice. In D2 mice, Ddit3 deletion conferred mild protection to RGC somas, but did not significantly prevent RGC axonal degeneration. Together, these data suggest that DDIT3 plays a minor role in perpetuating RGC somal apoptosis caused by chronic ocular hypertension-induced axonal injury, but does not significantly contribute to distal axonal degeneration.


Introduction
Glaucoma is an age-related neurodegenerative disease characterized by the death of retinal ganglion cells (RGCs), the output neurons of the retina. An important risk factor for glaucomatous RGC death is elevated intraocular pressure (IOP), which leads to RGC axonal injury at the lamina cribrosa [1][2][3][4][5] (termed the glial lamina in mice 5 ). This insult is thought to trigger molecular signaling within RGCs that regulates somal degeneration proximal to the site of injury and axonal degeneration distal to the site of injury [6][7][8][9][10] . Identifying the molecular signaling pathways that lead from ocular hypertensive injury to RGC death is critical for understanding the pathobiology of glaucoma. To date, a mechanism important in both proximal and distal RGC degeneration has not been identified. The pro-apoptotic molecule BAX was shown to be required for RGC somal death but not axonal degeneration after chronic ocular hypertension and acute optic nerve injury 6,11,12 . Thus, the mechanism triggered by ocular hypertension that regulates glaucomatous neurodegeneration must ultimately converge upon BAX induction.
The adaptive response to endoplasmic reticulum (ER) stress (known as the unfolded protein response or the integrated stress response) has been implicated as a driver of neuronal death in many neurodegenerative diseases, including glaucoma [13][14][15][16][17] . After prolonged and unresolved ER stress, the unfolded protein response has been shown to promote apoptosis via induction of DNA damage inducible transcript 3 (DDIT3, also known as CHOP). DDIT3 has been shown to act as a pro-apoptotic transcription factor; DDIT3 promoted transcription of proapoptotic Bbc3 18,19 , Bim 18,19 , Gadd34 20 , Dr5 19,21 , and Ero1α 22 genes and inhibited transcription of the prosurvival gene Bcl2 19,20,23,24 . DDIT3 was also shown to be important for the translocation of activated BAX from the cytosol to the mitochondria 25,26 ; allowing the intrinsic apoptotic cascade to ensue. Therefore, as a pro-apoptotic transcription factor upstream of BAX, DDIT3 may be an important regulator of RGC death after glaucomatous insult.
DDIT3 has been shown to regulate RGC death in glaucoma and various other neurodegenerative diseases 27,28 . DDIT3 was expressed by RGCs after glaucomarelevant insults, including optic nerve crush [13][14][15]29 and the microbead model of acute ocular hypertension 13,14 . In addition, Ddit3 was upregulated in both the retinas and optic nerve heads (ONHs) of mice with chronic ocular hypertension prior to the onset of glaucomatous neurodegeneration [30][31][32] . Ddit3 deficiency or silencing was protective to RGC somas after mechanical axonal injury (optic nerve crush) 14,17,33 and the microbead model of acutely induced ocular hypertension 14,33 . Interestingly, despite not appearing to have a major role in RGC axonal degeneration after optic nerve crush 17 , DDIT3 deficiency lessened axonal degeneration in an acute ocular hypertension model 33 . This protection, though minor, appeared roughly equal to the level of somal protection, suggesting that in some cells, Ddit3 deficiency completely protected the RGC after an ocular hypertensive injury 33 .
DDIT3 appears to be an important mediator of RGC viability after glaucoma-relevant injuries. However, the role of DDIT3 in glaucomatous neurodegeneration has not been tested in a model of stochastic, age-related ocular hypertension. Here, we critically tested the role of DDIT3 in RGC axonal degeneration and somal loss in an inherited, age-related mouse model of chronic ocular hypertension. We found DDIT3 played a minor role in RGC somal death but not axonal degeneration in the DBA/2J (D2) mouse model of chronic, age-related ocular hypertension 3,5,34-36 .

Materials and methods
Mice DBA/2J (D2) mice and mice with a null allele of Ddit3 37 (B6.129S(Cg)-Ddit3 tm2.1Dron /J) were obtained from the Jackson Laboratory (Stock numbers 000671 and 005530, respectively). The Ddit3 null allele was backcrossed to the D2 background 10 times (>99% D2). After this backcross was completed, the D2.Ddit3 colony was maintained by D2.Ddit3 +/− × D2.Ddit3 +/− intercrossing. D2.Ddit3 +/+ environment-matched littermates were used as genetic controls for D2.Ddit3 −/− mice, and each genotype group included roughly equal numbers of females and males (D2.Ddit3 +/+ : 30 female, 34 male; D2.Ddit3 −/− : 29 female, 31 male). Mice were fed chow and water ad libitum and were housed on a 12-h light-to-dark cycle. All experiments were conducted in adherence with the Association for Research in Vision and Ophthalmology's statement on the use of animals in ophthalmic and vision research and were approved by the University of Rochester's University Committee on Animal Resources.

Retina processing for plastic sectioning
As previously described 9 Optic nerve processing for plastic sectioning and grading Optic nerves were harvested and processed as previously described 6,9,38 . In brief, optic nerves were fixed in situ in 2.5% glutaraldehyde, 10% formalin in 1× PBS for 24 h at 4°C. Nerves were dissected from the brain and were incubated in 1% osmium for 2 h at room temperature. Otherwise, nerves were processed identically to eyes as described above. 1.5 μm cross sections were cut and collected on microscope slides. Nerve sections were stained with 1% paraphenylenediamine (PPD) in absolute methanol for 15 min, and washed with 100% ethanol for 10 min. PPD stains the myelin sheath of all axons but differentially darkly stains the axoplasm of dying axons. A masked observer used a validated grading scale to assess the level of glaucomatous damage of each optic nerve. As previously described 5,6,9,40 , nerves with <5% axons damaged or lost (consistent with axonal loss associated with normal aging) were judged to have no/early damage, nerves judged to have moderate damage had 5-50% axonal damage or loss (averaging~30% loss) often with localized areas of gliosis, and nerves with >50% axonal damage or loss, often with large areas of glial scaring, were judged to have severe damage. A masked observer selected optic nerves with the most axonal damage (judged to have <5% axonal survival) for assessment of RGC somal survival.

Controlled optic nerve crush
Controlled optic nerve crush (CONC) was performed as previously described 6,38,39 . Briefly, mice were anesthetized with intraperitoneal 100 mg/kg ketamine and 10 mg/kg xylazine. Analgesic 2 mg/kg meloxicam was administered subcutaneously prior to surgery. The optic nerve was exposed and crushed immediately behind the eye with self-closing forceps for 5 s. Sham surgery was performed on the contralateral eye, where the optic nerve was exposed but not crushed. Antibiotic ointment was applied to the eyes following the procedure. Eyes were harvested 5 and 14 days post-CONC.

IOP measurement
As previously described 9,35,38 , IOPs were measured by a masked observer using Tonolab (Colonial Medical Supply, Franconia, NH, USA) according to manufacturer's instructions 3-5 min after intraperitoneal administration of anesthetic 100 mg/kg ketamine and 10 mg/kg xylazine.
Retinas were then washed and incubated for 24 h at 4°C in secondary antibodies (Table 1) diluted in 1× PBS. Retinas were washed and mounted on microscope slides ganglion cell layer-up in Flourogel in TRIS buffer (Electron Microscopy Sciences, 17985-11).

Cell quantification
As previously described 38,39 , cCASP3+ RBPMS+ cells were quantified using eight 20x fields per retina, and RBPMS+ cell counts were assessed using eight 40x fields per retina. Images were taken approximately 220 μm from the peripheral edge of the retina and were equally spaced from each other. The manual cell counter plug-in in ImageJ was utilized for cell quantification. Retinal imaging and cell quantifications were performed by a masked observer. Cell quantifications were normalized to the total area measured and reported as cells/mm 2 .

Statistical analysis
Data were analyzed using GraphPad Prism8 software. Comparisons between two groups (cCASP3+ RBPMS+ cells/mm 2 after CONC between genotypes, Fig. 1a and %RGC survival in retinas with severe optic nerves between genotypes, Fig. 4) were analyzed using an unpaired two-tailed student's t test. Comparisons across more than two groups (RGCs/mm 2 14 days after sham and CONC procedures between genotypes, Fig. 1b) or two groups across multiple timepoints (IOP measurements at multiple timepoints between genotypes, Fig. 2b) were analyzed using a two-way ANOVA followed by a Sidak post hoc test. For these statistical tests, multiplicity adjusted P values are reported. The comparison of the percent of optic nerves at each grade between genotypes ( Fig. 3b) was analyzed using a Chi-square test. P values of <0.05 were considered statistically significant. Throughout the manuscript, results are reported as mean ± standard error of the mean (SEM).
Ddit3 deletion did not alter D2-associated endophenotypes The ONH is an important site in the pathobiology of glaucoma. In ocular hypertensive DBA/2J mice, the ONH is likely the site of an early critical axonal injury 4,5 . To ensure Ddit3 deletion did not cause any developmental ONH or retinal abnormalities in D2 mice, D2.Ddit3 −/− and D2.Ddit3 +/+ ONH and retinal morphologies were assessed at 1.5-3 M. Ddit3 deletion caused no gross morphological ONH or retinal abnormalities in D2 mice as judged by semi-thin sections (Fig. 2a).
ER stress has been implicated in regulating IOP elevation for some genetic causes of glaucoma 45,46 . Since RGC degeneration in D2 mice depends on age-related IOP elevation 36,47-51 , it was important to determine if Ddit3 deficiency altered IOP elevation in D2 mice. IOP was assessed at 5, 7.5, 9, 10.5, and 12 M (Fig. 2b). As a population, IOP was not elevated at 5 and 7.5 M. Both genotypes had significant IOP elevations at 9, 10.5, and 12 M compared to baseline IOPs taken at 5 M. D2.Ddit3 −/− mice had similar IOPs to D2.Ddit3 +/+ mice at each timepoint measured, thus, Ddit3 deletion did not alter the stereotypic IOP profile of D2 mice.
Ddit3 deletion did not significantly prevent RGC axonal degeneration in a model of chronic ocular hypertension DDIT3 has been implicated in driving both RGC somal and axonal degeneration after acute axonal injury 52  Ddit3 −/− mice. Optic nerve damage was graded as "no/ early", "moderate", or "severe" using a validated grading scale 5,6,9,40 (Fig. 3a, see "Materials and methods" for grading details). Neither D2.Ddit3 +/+ nor D2.Ddit3 −/− optic nerves exhibited any signs of axonal degeneration at 1.5-3 M (Fig. 3b). At 12 M, D2.Ddit3 −/− mice had similar levels of optic nerve damage compared to D2.Ddit3 +/+ controls (Fig. 3b). Therefore, Ddit3 deletion did not provide protection to RGC axons in D2 mice, suggesting that DDIT3 is likely not a critical regulator of axonal degeneration in a model of chronic age-related ocular hypertension.

Discussion
Chronic ocular hypertension is an important risk factor for the development of glaucomatous neurodegeneration. Ocular hypertension is thought to injure RGCs as they exit the eye at the lamina cribrosa [1][2][3][4][5] . Axonal injury is thought to trigger both RGC somal and axonal degeneration pathways [6][7][8][9][10] . ER stress, specifically DDIT3, has been implicated as a driver of RGC death after glaucomarelevant injuries [13][14][15]17,33 . Importantly, DDIT3 was shown to regulate both RGC axonal degeneration and somal apoptosis in models of mechanical axonal injury and acutely induced ocular hypertension 33 . In the present work, the role of DDIT3 in age-related, chronic ocular hypertension-induced RGC death was investigated. While Ddit3 deletion in D2 mice provided mild protection to RGC somas, it did not significantly prevent RGC axonal degeneration. These data suggest DDIT3 has a minor role in regulating RGC somal death after axonal injury induced by chronic ocular hypertension. Therefore, the molecular process triggered by ocular hypertension that governs both RGC somal apoptosis and axonal degeneration remains unknown.
DDIT3 played a minor role in RGC somal degeneration in ocular hypertensive D2 mice; Ddit3 deletion protected 20% of RGC somas in retinas with severe RGC axonal degeneration. The pro-apoptotic molecule BAX was shown to be required for RGC somal degeneration after CONC 6,54 and in ocular hypertensive D2 mice 6 , however, BAX did not regulate RGC axonal degeneration in these models 6 . DDIT3 is important in the translocation of BAX from the cytosol to the mitochondria during prolonged ER stress 25,26 . However, because Ddit3 deletion only protected~20% of RGC somas in D2 mice, another mechanism must work in tandem with DDIT3 to induce BAX. The mitogen-activated protein kinase effector and transcription factor JUN was shown to be an important regulator of ocular hypertension-induced RGC somal apoptosis. In fact, Jun deficiency protected~2.5 times more RGC somas compared to Ddit3 deficiency in 12 M D2 mice with severe optic nerve degeneration 9 . Interestingly, JUN and DDIT3 were shown to additively contribute to RGC somal apoptosis after CONC; dual deletion of Jun and Ddit3 conferred 75% somal protection 120 days post-CONC (Jun and Ddit3 deletion alone allowed 48% and 25% protection at this timepoint, respectively) 17 . Therefore, identifying the ocular hypertension-induced upstream regulator of both JUN and DDIT3 may be an important step in determining an upstream mechanism driving glaucomatous RGC death. Further, the role of both Jun and Ddit3 in glaucomatous neurodegeneration should be tested in a model of agerelated chronic ocular hypertension.
Previous reports have shown that DDIT3 deficiency lessened both RGC axonal degeneration and somal loss in the microbead model of acute IOP elevation and after CONC 33 . However, we report no difference in RGC axonal degeneration between D2.Ddit3 +/+ and D2.Ddit3 −/− mice at 12 M. This result is consistent with our previous report that Ddit3 deletion did not protect from loss of RGC axonal conductance after CONC in C57BL/6J mice 17 , unlike manipulation of molecules known to protect axons from degeneration (Wld S and Sarm1 10 ). The differences between these results could perhaps be explained by the nature and/or duration of the insults. In the microbead model of acute ocular hypertension, optic nerves had only moderate neurodegeneration (~29% axonal loss) after 8 weeks 33  Ddit3 +/+ controls with equally severe optic nerves (% survival ± SEM; D2.Ddit3 +/+ : 11.0 ± 1.8%, D2.Ddit3 −/− : 28.8 ± 1.4%, n ≥ 10 per genotype, P < 0.001, two-tailed t test). RGC counts were normalized to young (1.5-2.5 M) control RGC counts of the respective genotype, see Fig. 1b. Error bars, SEM. Scale bar, 100 μm possible that DDIT3 deficiency can delay axonal degeneration after an ocular hypertensive injury, but not prevent degeneration after long term ocular hypertensive insult or severe mechanical injury. It is also conceivable that the differences in findings are explained by the age-related nature of the DBA/2J disease, as acute induction of ocular hypertension was performed on young animals 33 . Finally it is possible that there is a small number of axons surviving in the D2. Ddit3 −/− optic nerve that were not detected using a grading system. Regardless, our findings suggest that DDIT3 does not play a major role in axonal degeneration in an age-related, chronic ocular hypertension model of glaucoma.
In conclusion, the role of DDIT3 in glaucomatous neurodegeneration was tested in the DBA/2J (D2) inherited model of chronic, age-related ocular hypertension. DDIT3 deficiency did not alter retinal or optic nerve morphology, nor did it alter the IOP profile of the D2 model. In this model, DDIT3 did not contribute to RGC axonal degeneration, but it was responsible for~20% of RGC somal apoptosis. Future work should focus on the roles of both JUN and DDIT3 together in perpetuating glaucomatous RGC death and should elucidate upstream regulators of both JUN and DDIT3 after glaucoma-relevant injury.