Quantitative Retinal Protein Analysis after Optic Nerve Transection Reveals a Neuroprotective Role for Hepatoma-Derived Growth Factor on Injured Retinal Ganglion Cells

Adam Hollander; Philippe M. D'Onofrio; Mark M. Magharious; Meghan D. Lysko; Paulo D. Koeberle

doi:10.1167/iovs.11-8350

Abstract

Purpose.: Retinal ganglion cell (RGC) degeneration is an important cause of visual impairment and can be modeled by optic nerve transection, which causes the death of 90% of RGCs within 14 days postaxotomy. We performed a proteomic study to identify and quantify proteins in the rat retina after optic nerve transection. Our goal was to isolate potential targets for therapeutic intervention to prevent RGC degeneration.

Methods.: iTRAQ proteomics was used to analyze adult rat retinas at 1, 3, 4, 7, 14, and 21 days postaxotomy. Hepatoma-derived growth factor (HDGF), a target identified by iTRAQ, was delivered by intraocular injections. Wortmannin or PD98059 were coadministered with HDGF to determine if the protective effects of HDGF are dependent on PI3 kinase or MAP kinase activity, respectively.

Results.: At a false-discovery rate of 5%, 216 proteins were identified by iTRAQ proteomics, 71 of which showed changes in expression (<0.7× or >1.3×) at one time point after injury: 52 proteins had expression peaks, whereas 19 showed downward expression spikes. Levels of GAPDH did not change after axotomy. Among these differentially expressed proteins was HDGF. HDGF delivery significantly increased RGC survival compared with control treatments, and increased Akt phosphorylation in the retina at 24 hours after intraocular injection. RGC rescue by HDGF was dependent on both MAP kinase and PI3 kinase activity in the retina.

Conclusions.: We have identified numerous proteins that are differentially regulated at key time points after axotomy, and how the temporal profiles of their expression parallel RGC death. Using these data, we showed that HDGF is a potent neuroprotective factor for injured adult RGCs.

Introduction

Retinal ganglion cell (RGC) degeneration is a hallmark of glaucoma, traumatic and hereditary optic neuropathy, and diabetic retinopathy.^1–4 Optic nerve transection is a reproducible model of adult mammalian RGC degeneration.^5–10 The temporal profile of RGC death after axotomy is well established: 90% of injured RGCs die within 14 days of optic nerve transaction.^5,11–13 Injured RGCs survive for 4 days postaxotomy, after which they die rapidly, resulting in 50% cell death by day 7, with only 10% remaining at day 14.^5,11–13 The rate of DNA fragmentation in RGCs peaks at 7 days postaxotomy.⁷ There are two phases of RGC death after axotomy: a rapid phase over the first 2 weeks postinjury, and a protracted phase during which very little cell death occurs over the following months.¹² Although some of the triggers and mediators of RGC degeneration have been identified, most of this process remains unclear.

The role of several apoptotic effectors in RGC death has previously been investigated: caspase-3 and -9, Bax, Bcl-2, Bcl-X, and p38 mitogen-activated protein kinase (MAPK) regulate RGC apoptosis.^8,14–17 Triggers involved in RGC apoptosis include the loss of trophic support,^8–10 glutamate excitotoxicity,^18,19 inflammatory processes,^20–22 genetically encoded immune components,^23–25 voltage-gated potassium channel activity,^22,26 and dependence receptor function.²⁷ To widen our understanding of the mechanisms that underlie RGC degeneration, it is necessary to identify and quantify proteins en masse, at critical time points along the temporal profile of RGC degeneration.

Proteome profiling, using isobaric tags for relative and absolute quantification (iTRAQ), permits the study of wide-scale changes in protein expression. We used mass spectrometry (MS) to evaluate retinal protein expression after iTRAQ labeling, revealing changes in the retinal proteome profile after RGC axotomy. Through this, we identified and quantified 216 proteins (false-discovery rate [FDR] of 5%) at six time points after optic nerve transection. One of the targets that we identified (hepatoma-derived growth factor [HDGF]) has previously been shown to antagonize apoptosis in cell lines.²⁸ We tested the effect of HDGF on RGC survival after axotomy, and identified intracellular pathways that contribute to HDGF neuroprotection. These findings and future studies on the protein targets revealed by this work will aid in developing novel therapeutics for retinal degeneration.

Materials and Methods

Optic Nerve Transection

Animal experiments were conducted in accordance with the ARVO Statement for the use of animals in ophthalmic and visual research. Optic nerve transections were performed as previously described.^7,29,30 All animals were pathogen-free female Sprague-Dawley rats. Rats were housed in a controlled environment and monitored daily. For surgical procedures, animals were anesthetized with isoflurane (Baxter, Mississauga, ON; 2.5%, 0.8 L/min O₂), and placed in a stereotaxic frame with a gas anesthesia mask to maintain sedation. The nerve was accessed through an incision in the superior rim of the orbit, following retraction of the overlying rectus muscles. Meningeal coverings of the optic nerve were dissected longitudinally to avoid disrupting the blood supply to the retina. The optic nerve was lifted out of the meningeal coverings, and cut within 2 mm of the back of the eye. Postoperative discomfort was minimized with intraperitoneal injections of ketoprofen (5 mg/mL, dosage for rats: 0.1 mL/100 g body weight). Following recovery under a heat lamp, animals were returned to their normal housing.

Intraocular Injections

Separate groups of animals, not used for any of the iTRAQ analyses, were prepared to test the effect of HDGF on RGC survival, and determine which intracellular pathways are required for HDGF neuroprotection. Each animal received an intraocular injection of HDGF (1 μg/μL dissolved in PBS) at 3 days postaxotomy (n = 4). A total of 4 μL of HDGF solution was injected into each eye. To examine the effects of blocking phosphatidylinositol-3 kinase (PI3K) activity on HDGF neuroprotection, animals received HDGF + Wortmannin (2 μL of 20 mM Wortmannin dissolved in dimethyl sulfoxide [DMSO]) at 3 days postaxotomy (n = 4). Wortmannin is a selective inhibitor of PI3K with a half maximal inhibitory concentration (IC₅₀) of 5 nM.^31,32 The effects of MAPK pathway inhibition on HDGF neuroprotection were studied by intraocular delivery of HDGF + PD98059 (2 μL of 20 mM PD98059 dissolved in DMSO) at 3 days postaxotomy (n = 4). PD98059 is a specific inhibitor of MAPK activation by MapK/ERK Kinase 1 (MEK1), with an IC₅₀ of 4 μM.³³ Control intraocular injections consisted of 2 μL Wortmannin (20 mM) dissolved in DMSO (n = 4), or 2 μL PD98059 (20 mM) dissolved in DMSO (n = 4), PBS (4 μL; n = 4), or DMSO (2 μL; n = 3).

Intraocular injections were performed as previously described.^29,30 A hydraulic injection system consisting of a pulled glass micropipette, coupled to a 10-μL Hamilton syringe through polyetheretherketone tubing, was used to deliver solutions into the vitreous chamber. Following injection, the cornea was covered with ophthalmic ointment to prevent desiccation and animals were returned to their normal housing.

Tissue Processing for Quantifying RGC Survival

Animals were killed at 14 days postaxotomy and eyes were enucleated. After removing the cornea and lens, the eye cup was fixed in 4% paraformaldehyde for 1 hour, and then rinsed in PBS for 15 minutes. The retina was removed from the eye cup, flat-mounted, and coverslipped using a 50:50 solution of PBS:glycerol. Fluorogold-stained RGCs were visualized using an Andor iXon 885+ electron-multiplying charge-coupled device camera (Andor Technology, Belfast, Northern Ireland) mounted on a Leica DM LFSA microscope (Leica Microsystems, Concord, Canada), with a Sutter Lambda XL light source (Quorum Technologies, Guelph, Canada) for illumination. RGC densities were measured at 1/6 (inner), 1/2 (middle), and 5/6 (outer) retinal eccentricities (defined distances from the center of the retina) of retinal quadrants. Results were grouped by retinal eccentricity, and expressed as the mean number of RGCs/mm² ± SEM. One-way ANOVA followed by post hoc analysis using Tukey's Multiple Comparisons were used to determine statistical significance between experimental and control samples.

Western Blotting

Retinal lysates from a separate group of animals were used for Western blots, independent of the samples used for iTRAQ analysis. Western blots on whole retinas were performed as previously described.³⁴ Total protein fractions were separated by SDS-PAGE (10% acrylamide) and immunoblotted after semidry electrotransfer to nitrocellulose membranes (0.2 μm pore size). The primary antibody for detecting HDGF was a rabbit polyclonal (1:250, Santa Cruz, Santa Cruz, CA). Primary antisera for intracellular messengers consisted of rabbit-anti-phospho-Akt (1:1000; 4060, Cell Signaling Technology, Pickering, ON), rabbit-anti-Akt (1:1000; 9272, Cell Signaling Technology), and rabbit-anti-phospho-MAPK (1:1000; 9101, Cell Signaling Technology), mouse-anti-MAPK (1:1000; 4696, Cell Signaling Technology). Primary antisera were detected with a 1:2000 dilution of secondary antibody (horseradish peroxidase conjugated, cross reacted against rat serum antigens; Jackson Immunoresearch, West Grove, PA). Chemiluminescent immunoreactive complexes were visualized using a Bio-Rad Fluor-S Max imager (Bio-Rad, Mississauga, ON). Loading was verified by visualizing protein bands with Ponceau S dye, and reprobing blots with a rabbit antisera directed against glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (1:2000; 2118, Cell Signaling Technology).

Retinal Processing and 8-plex iTRAQ Proteomics

iTRAQ proteomics was performed using an 8-plex procedure so that all experimental samples were processed in the same MS run, together with the normal (unlesioned) retina sample that was used as the baseline standard. All protein quantifications for experimental samples were performed relative to the normal (unlesioned) sample.

Animals were anesthetized with isoflurane and killed by cervical dislocation at 1, 3, 4, 7, 14, or 21 days postaxotomy (n = 8 for each time point). Normal-unlesioned retinas (n = 8) were used as the baseline standard for relative quantitation by iTRAQ proteomics. Eyes were enucleated, and the cornea and lens were carefully removed in ice-cold PBS. The retinas were then removed from the eye cup, placed in a 1.5-mL centrifuge tube, rapidly frozen on dry ice, and maintained at −80°C until processing.

iTRAQ was performed at the Ontario Cancer Biomarker Network facilities in Toronto, ON, Canada.³⁵ Frozen retinas were thawed on ice and 250 μL sonication buffer was added to each retina. The sonication buffer was made up with RIPA buffer (Thermo Scientific, Ottawa, Canada) and Protease Inhibitor cocktail I (Calbiochem; EMD-Millipore, Billerica, MA). Each sample was alternately sonicated and cooled five times. Retinal lysates from the same time point (n = 8) were then combined and incubated at 4°C for 30 minutes. Thus, each data point in the iTRAQ runs represented a sum of eight retinal samples in order to minimize variations due to experimental conditions or intra-animal variability. The entire iTRAQ analysis via MS was then run in duplicate. The combined sonicated samples were spun at 14,000 rpm for 30 minutes. To ensure that constituents of the protease inhibitor cocktail did not affect trypsin digestion for iTRAQ, the lysed protein mixture was subjected to protein extraction/concentration using a Proteospin Kit (Norgen, Thorold, ON, Canada), which removes salts, detergents, and small molecules, including the protease inhibitors from the protein fraction. Protein concentrations were measured using a Bio-Rad DC protein assay kit.

iTRAQ labeling and trypsin digestion were executed according to vendor's instructions. The 8-Plex iTRAQ labeling kit was purchased from Applied Biosystems (Burlington, Canada), and trypsin was purchased from Promega (Madison, WI). An amount of 30 μg protein from each time point was used for labeling. iTRAQ-labeled and trypsin-digested samples were then pooled and subjected to strong cation exchange (SCX) fractionation.

Sample fractionation and liquid chromatography (LC)/MS/MS analysis were performed as previously published.³⁵ Twenty SCX fractions were collected, and 15 SCX fractions were analyzed for each LC/MS/MS run. Half of each collected SCX fraction was analyzed in each run. Protein-pilot 3.0 (AB Sciex, Foster City, CA) identified 216 proteins with more than one peptide, and a ProtScore of greater than 1.3 (95% confidence) based on the Swissprot-Uniprot database of the Rattus norvegicus proteome.The FDR was calculated using a decoy protein database with Proteomics System Performance Evaluation Pipeline (PSPEP) software (AB Sciex) as previously described.³⁶ PANTHER analysis (GO cellular components) was performed on all identified proteins to compare the relative amounts of different subclasses of proteins to the to the known R. norvegicus proteome. For the preparation of Table 1 protein, the results of the two iTRAQ runs were combined. Only proteins that were identified in both iTRAQ runs were included in the list, thereby reducing the chances of a false-positive identification from approximately 1:20 to approximately 1:400 using the FDR of 5%. The mean relative protein level at each time point and a 95% confidence interval (CI) was then calculated, and statistically analyzed relative to the mean ±95% CI for GAPDH at each time point, using ANOVA followed by Tukey's post hoc comparisons. If protein quantification data were not available for both iTRAQ runs, the results of a single run were included in the table.

Table 1.

View Table

Proteins That Are Differentially Expressed after Optic Nerve Transection

Table 1.

Proteins That Are Differentially Expressed after Optic Nerve Transection

UniProt Accession	UniProt ID	Entrez-Gene Name	Avg. %Cov (95)	No. of Pep	Day 1 Avg.	Day 3 Avg.	Day 4 Avg.	Day 7 Avg.	Day 14 Avg.	Day 21 Avg.
Unchanged proteins
P04797	G3P_RAT	Glyceraldehyde-3-phosphate dehydrogenase	45.6	37	1.030	1.088	1.035	1.056	1.052	1.118
		Glyceraldehyde-3-phosphate dehydrogenase			0.168	0.044	0.068	0.327	0.285	0.367
P07323	ENOG_RAT	Gamma-enolase	39.6	16	1.100	1.047	1.047	0.981	1.022	0.998
					0.634	0.781	0.526	0.346	0.525	0.858
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
Upregulated proteins
Day 1 maxima
P00507	AK1A1_RAT	Alcohol dehydrogenase [NADP+]	5.8	1	2.120	0.967	1.031	1.302
					1.523	0.792	0.590	0.074
					P < 0.001	NS P > 0.05	NS P > 0.05	P < 0.05
P16036	H4_RAT	Histone H4	43.7	6	1.485	1.054	1.063	1.173	1.132	1.056
					0.203	0.878	0.518	0.836	0.083	0.007
					P < 0.001	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
P51635	HINT1_RAT	Histidine triad nucleotide-binding protein 1	54.8	6	1.479	0.841	1.042	0.904	1.063	0.909
		Histidine triad nucleotide-binding protein 1			6.006
					P < 0.05
P62804	H2B1_RAT	Histone H2B type 1	40.8	9	1.460	1.404	1.209	1.115	1.366	1.352
					1.231	1.067	0.973	1.493	1.337	1.906
					P < 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
P62959	MPCP_RAT	Phosphate carrier protein, mitochondrial	3.4	1	1.445	1.310	1.277	1.154	1.174	1.424
		Phosphate carrier protein, mitochondrial			3.987	2.139	2.468	3.179	1.465	1.647
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
Q00715	AATM_RAT	Aspartate aminotransferase, mitochondrial	12.3	3	1.372	0.902	0.925	1.136	0.861	1.359
		Aspartate aminotransferase, mitochondrial			3.020	0.912	0.692	0.783	0.816	2.830
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
Q6LED0	HDGF_RAT	Hepatoma-derived growth factor	14.3	3	1.334	1.245	1.168	1.386	0.892	1.254
Q8VHK7	ATPA_RAT	ATP synthase subunit alpha, mitochondrial	8.0	5	1.334	1.009	1.162	1.027	1.080	1.231
		ATP synthase subunit alpha, mitochondrial			0.014	0.047	0.704	0.543	0.599	0.312
					P < 0.001	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
Day 3 maxima
P26772	CH10_RAT	10-kDa heat-shock protein, mitochondrial	11.8	1	0.845	1.584	1.310	0.755	0.779	0.854
		10-kDa heat-shock protein, mitochondrial				2.928
						P < 0.01
P63055	PCP4_RAT	Purkinje cell protein 4	27.4	1	1.107	1.501	1.291	1.066	1.225	1.084
					1.194	0.984	0.597	0.298	1.982	0.466
					NS P > 0.05	P < 0.001	P < 0.01	NS P > 0.05	NS P > 0.05	NS P > 0.05
P18437	HMGN2_RAT	Non-histone chromosomal protein HMG-17	35.6	3	1.235	1.447	1.071	1.075	1.411	1.177
		Non-histone chromosomal protein HMG-17			0.566	0.822	0.506	0.503	0.459	0.360
					P < 0.01	P < 0.001	NS P > 0.05	NS P > 0.05	P < 0.001	NS P > 0.05
P35704	PRDX2_RAT	Peroxiredoxin-2	15.4	1	1.110	1.408	1.408	1.222	1.048	1.172
					0.597	0.791	0.965	1.247	0.491
					NS P > 0.05	P < 0.01	P < 0.001	NS P > 0.05	NS P > 0.05
P02770	ALBU_RAT	Serum albumin	32.8	20	0.947	1.397	0.937	0.843	0.784	0.763
					0.319	0.341	0.052	0.505	1.487	0.970
					NS P > 0.05	P < 0.01	NS P > 0.05	P < 0.05	P < 0.01	P < 0.001
P83868	TEBP_RAT	Prostaglandin E synthase 3	10.6	1	1.238	1.396	1.181	1.049	1.032	1.124
					0.342	1.947	1.301	1.575	0.949	0.217
					NS P > 0.05	P < 0.01	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
O35179	SH3G2_RAT	Endophilin-A1	7.4	2	0.915	1.381	0.864	1.104	0.904	0.887
					0.390	0.009	0.372	0.325	0.058	0.912
					NS P > 0.05	P < 0.001	P < 0.01	NS P > 0.05	P < 0.05	P < 0.001
O35814	STIP1_RAT	Stress-induced-phosphoprotein 1	8.5	3	1.212	1.359	1.124	1.306	1.280	1.221
					0.722	2.387	2.495	2.101	1.289	2.218
					NS P > 0.05	P < 0.001	P < 0.01	NS P > 0.05	P < 0.05	P < 0.001
P45592	COF1_RAT	Cofilin-1	23.5	3	1.077	1.330	1.163	1.204	0.984	0.942
					1.154	2.140	0.834	1.890	0.520	1.044
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
Q923W4	HDGR3_RAT	Hepatoma-derived growth factor–related protein 3	12.1	3	1.274	1.309	1.036	1.187	1.083	1.206
		Hepatoma-derived growth factor–related protein 3			0.016	0.334	0.053	0.122	0.968	0.349
					P < 0.001	P < 0.001	NS P > 0.05	P < 0.05	NS P > 0.05	NS P > 0.05
P15865	H12_RAT	Histone H1.2	24.2	8	1.237	1.293	1.100	1.065	1.113	1.068
					0.119	0.085	0.321	0.557	0.155	0.113
					P < 0.001	P < 0.001	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
Day 4 maxima
O35952	GLO2_RAT	Hydroxyacylglutathione hydrolase, mitochondrial	5.7	1	1.562	1.076	1.833	1.490	0.965	1.107
		Hydroxyacylglutathione hydrolase, mitochondrial			5.132	2.782	5.497	3.670	1.195	1.404
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
P07632	SODC_RAT	Superoxide dismutase [Cu-Zn]	23.4	3	1.319	0.966	1.718	1.568	1.554	1.412
					3.195	1.145	4.833	5.353	4.587	3.405
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
P26453	BASI_RAT	Basigin	4.6	1	1.185	1.189	1.422	1.211	1.358	1.157
					0.306	1.148	0.854	0.048	1.117	0.501
					NS P > 0.05	NS P > 0.05	P < 0.001	NS P > 0.05	P < 0.01	NS P > 0.05
Q5RKI0	WDR1_RAT	WD repeat-containing protein 1	2.3	1	1.038	0.737	1.381	1.056	1.082	1.030
					3.812	0.105	2.235	2.993	3.250	4.723
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
P13264	GLSK_RAT	Glutaminase kidney isoform, mitochondrial	2.4	1	1.413	1.127	0.914	1.911	1.275	1.557
		Glutaminase kidney isoform, mitochondrial			0.757	0.313	0.654	1.729	0.554	0.028
					P < 0.001	NS P > 0.05	NS P > 0.05	P < 0.001	P < 0.05	P < 0.001
Day 7 maxima
P97615	THIOM_RAT	Thioredoxin, mitochondrial	9.0	1	1.102	1.310		1.503	1.118
					0.442	1.137		1.207	2.003
					NS P > 0.05	NS P > 0.05		P < 0.01	NS P > 0.05
Q62951	DPYL4_RAT	Dihydropyrimidinase-related protein 4 (Fragment)	5.9	2	1.074	1.087	0.984	1.429	1.098	1.165
		Dihydropyrimidinase-related protein 4 (Fragment)			1.814	0.389	2.942	0.140	2.171	4.062
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
Q3T1J1	IF5A1_RAT	Eukaryotic translation initiation factor 5A-1	7.8	1	1.026	1.270	0.995	1.401	1.226	1.115
P50398	GDIA_RAT	Rab GDP dissociation inhibitor alpha	2.2	1	0.716	0.875	0.743	1.385	0.682	1.074
					0.866	0.814	3.353	0.484	1.521	0.941
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
Q6PEC4	SKP1_RAT	S-phase kinase-associated protein 1	22.7	2	0.971	0.939	0.998	1.371	0.895	0.942
					0.635	1.085	2.127	2.906	0.329	2.157
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
Q5XIT1	MARE3_RAT	Microtubule-associated protein RP/EB family member 3	9.6	3	1.101	0.823	1.067	1.318	1.052	0.978
		Microtubule-associated protein RP/EB family member 3			0.373	0.115	0.813	2.152	0.029	0.919
					NS P > 0.05	P < 0.05	NS P > 0.05	P < 0.05	NS P > 0.05	NS P > 0.05
P31044	PEBP1_RAT	Phosphatidylethanolamine-binding protein 1	58.8	12	1.112	0.933	1.283	1.296	1.193	1.166
		Phosphatidylethanolamine-binding protein 1			0.374	0.966	0.269	0.285	0.536	0.138
					NS P > 0.05	P < 0.05	P < 0.001	P < 0.001	P < 0.05	NS P > 0.05
Day 14 maxima
Q9ET62	IMPG1_RAT	Interphotoreceptor matrix proteoglycan 1	2.3	1	1.210	0.874	1.573	1.189	1.642	1.331
		Interphotoreceptor matrix proteoglycan 1			0.476	0.678	1.061	0.485	0.122	0.011
					P < 0.05	P < 0.01	P < 0.001	NS P > 0.05	P < 0.001	P < 0.01
P40307	PSB2_RAT	Proteasome subunit beta type-2	8.5	1	0.886	0.800			1.595
					1.000	1.000			2.000
					P < 0.001	P < 0.001			P < 0.001
P17132	HNRPC_RAT	Heterogeneous nuclear ribonucleoprotein C (Fragment)	32.3	4	1.232	1.252	1.492	1.065	1.497	1.253
P61980	HNRPK_RAT	Heterogeneous nuclear ribonucleoprotein K	3.2	1	1.256	1.009	1.053	1.369	1.389	1.112
		Heterogeneous nuclear ribonucleoprotein K			1.206	1.302	0.263	4.730	1.290	0.459
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
P13086	SUCA_RAT	Succinyl-CoA ligase (GDP-forming) subunit alpha, mitochondrial	8.3	2	1.040	1.347	0.963	1.022	1.372	0.755
					0.814	2.884	1.112	0.639	0.751	1.232
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
Q3SWU3	HNRDL_RAT	Heterogeneous nuclear ribonucleoprotein D-like	4.7	1	0.956	1.187	1.256	0.965	1.364	1.388
Q7M0E3	DEST_RAT	Destrin	21.2	3	1.123	1.168	1.107	1.090	1.319	1.047
					1.069	1.252	1.754	0.249	2.087	0.624
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
Q9EQS0	TALDO_RAT	Transaldolase	9.2	3	1.087	0.988	1.027	1.124	1.298	1.268
					1.072	0.408	1.698	0.067	1.044	0.879
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
Day 21 maxima
Q64598	H2A1F_RAT	Histone H2A type 1-F	10.8	1	2.796	2.303	3.829	0.708	3.902	5.368
P0C170	H2A1E_RAT	Histone H2A type 1-E	10.8	1	1.840	1.557	2.154	1.004	2.345	2.657
P35565	CALX_RAT	Calnexin	3.0	1	1.348	1.139	1.163	1.222	1.192	1.601
					0.657	0.317	0.060	0.233	0.380	1.493
					P < 0.001	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	P < 0.001
Q812D1	PSIP1_RAT	PC4 and SFRS1-interacting protein	3.2	1	1.066	1.148	1.010	1.184	1.116	1.536
Q01986	MP2K1_RAT	Dual-specificity mitogen-activated protein kinase kinase 1	3.1	1	1.176	1.218	1.164	1.403	1.267	1.422
		Dual-specificity mitogen-activated protein kinase kinase 1			1.683	0.377	0.417	1.565	0.118	1.041
					NS P > 0.05	NS P > 0.05	NS P > 0.05	P < 0.01	NS P > 0.05	P < 0.05
P15887	ARRS_RAT	S-arrestin	41.7	18	1.287	1.152	1.257	1.111	1.262	1.384
					0.486	0.855	1.018	0.419	0.273	0.512
					P < 0.01	NS P > 0.05	P < 0.01	NS P > 0.05	P < 0.05	P < 0.01
P24942	EAA1_RAT	Excitatory amino acid transporter 1	4.1	1	1.227	0.878	0.913	1.026	1.041	1.383
					0.724		0.308		0.360	1.936
					NS P > 0.05		NS P > 0.05		NS P > 0.05	P < 0.05
P50137	TKT_RAT	Transketolase	4.8	2	1.196	0.995	1.186	1.202	1.179	1.358
					0.241	0.103	0.135	0.225	0.343	0.127
					P < 0.001	P < 0.05	P < 0.001	P < 0.001	P < 0.01	P < 0.001
P61589	RHOA_RAT	Transforming protein RhoA	13.5	2	1.151	1.064	1.295	1.074	1.123	1.331
					0.461	0.331	1.162	0.184	0.724	0.128
					NS P > 0.05	NS P > 0.05	P < 0.01	NS P > 0.05	NS P > 0.05	P < 0.01
P84087	CPLX2_RAT	Complexin-2	19.4	2	1.121	1.209	1.279	1.049	1.259	1.327
					0.327	0.258	1.075	0.802	0.060	0.172
					NS P > 0.05	P < 0.05	P < 0.001	NS P > 0.05	P < 0.001	P < 0.001
Q6P9V9	TBA1B_RAT	Tubulin alpha-1B chain	30.9	15	1.169	0.978	1.032	1.019	0.966	1.317
P31000	VIME_RAT	Vimentin	4.4	1	1.155	1.104	1.119	1.299	1.153	1.305
					2.188	1.449	0.934	2.488	1.229	1.901
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
P50503	F10A1_RAT	Hsc70-interacting protein	10.3	3	1.160	1.266	1.207	1.205	1.205	1.294
					0.707	1.462	0.204	0.625	0.569	1.340
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
Downregulated proteins
Day 1 minima
Q66HR2	MARE1_RAT	Microtubule-associated protein RP/EB family member 1	6.7	1	0.492	0.858	0.799	0.909	0.819	0.896
		Microtubule-associated protein RP/EB family member 1			0.211	0.734	0.257	0.050	0.386	0.495
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05
P10818	CX6A1_RAT	Cytochrome c oxidase subunit 6A1, mitochondrial	14.4	1	0.516	0.841	0.734	0.834	0.610	1.127
		Cytochrome c oxidase subunit 6A1, mitochondrial			0.033	0.214	0.026	0.217	0.153	0.269
					P < 0.001	P < 0.001	P < 0.001	P < 0.001	P < 0.001	NS P > 0.05
Day 3 minima
P63164	RSMN_RAT	Small nuclear ribonucleoprotein-associated protein N	5.8	1	0.947	0.457	0.840	0.519	0.900	0.839
					3.187	0.224	2.645	0.371	0.161	1.593
					NS P > 0.05	P < 0.001	NS P > 0.05	P < 0.01	NS P > 0.05	NS P > 0.05
P0C5E9	CRBS_RAT	Beta-crystallin S	18.5	2	0.554	0.516	0.829	0.672	0.704	0.776
					0.269	0.529	0.341	0.230	0.580	0.960
					P < 0.001	P < 0.001	P < 0.01	P < 0.001	P < 0.001	P < 0.001
P25286	VPP1_RAT	V-type proton ATPase 116 kDa subunit a isoform 1	1.3	1	0.861	0.539	0.758	1.071	0.708	1.077
		V-type proton ATPase 116 kDa subunit a isoform 1			1.522	0.891	1.273	0.252	0.834	1.438
					NS P > 0.05	P < 0.001	NS P > 0.05	NS P > 0.05	P < 0.05	NS P > 0.05
Day 4 minima
P62882	GBB5_RAT	Guanine nucleotide-binding protein subunit beta-5	3.4	1	0.730	0.759	0.566	0.887	0.706	0.648
		Guanine nucleotide-binding protein subunit beta-5			0.588	0.364	0.196	0.235	0.669	0.059
					P < 0.001	P < 0.001	P < 0.001	P < 0.01	P < 0.001	P < 0.001
P28480	TCPA_RAT	T-complex protein 1 subunit alpha	2.2	1	0.865	0.748	0.655	0.809	0.927	0.845
					0.243	0.161	0.502	0.359	0.392	0.527
					P < 0.01	P < 0.001	P < 0.001	P < 0.001	P < 0.05	P < 0.001
P27881	HXK2_RAT	Hexokinase-2	3.2	2	1.138	0.752	0.670	0.860	0.928	0.913
					1.009	0.688	0.628	0.287	1.373	0.329
					NS P > 0.05	P < 0.01	P < 0.001	NS P > 0.05	NS P > 0.05	NS P > 0.05
Day 7 minima
P08082	CLCB_RAT	Clathrin light chain B	5.7	1	0.809	1.074	0.958	0.518	1.085	0.709
					1.005	0.916	0.608	0.090	0.602	0.596
					P < 0.05	NS P > 0.05	NS P > 0.05	P < 0.001	NS P > 0.05	P < 0.001
Q5U2Z3	NP1L4_RAT	Nucleosome assembly protein 1-like 4	2.6	1	0.840	0.722	0.922	0.580	0.816	1.012
P23928	CRYAB_RAT	Alpha-crystallin B chain	6.3	2	0.676	0.792	0.757	0.595	0.763	0.864
					0.169	0.044	0.105	0.199	0.180	0.355
					P < 0.001	P < 0.001	P < 0.001	P < 0.001	P < 0.001	P < 0.001
P61954	GBG11_RAT	Guanine nucleotide-binding protein G(I)/G(S)/G(O) subunit gamma-11	6.8	1	0.706	0.909	0.938	0.612	0.760	0.807
					0.306	0.073	0.207	0.657	0.760	0.426
					P < 0.001	P < 0.01	NS P > 0.05	P < 0.001	P < 0.001	P < 0.001
Q9Z2G8	NP1L1_RAT	Nucleosome assembly protein 1-like 1	2.6	1	0.880	0.905	0.854	0.643	0.779	1.016
P62697	CRBB2_RAT	Beta-crystallin B2	20.7	5	0.742	0.742	0.703	0.648	0.727	0.963
					0.316	0.163	0.285	0.984	0.017	0.189
					P < 0.001	P < 0.001	P < 0.001	P < 0.001	P < 0.001	P < 0.05
Q9R063	PRDX5_RAT	Peroxiredoxin-5, mitochondrial	6.6	1	0.803	0.767	0.953	0.673	0.897	0.810
					0.250	0.493	0.098	0.519	0.705	1.231
					P < 0.01	P < 0.001	NS P > 0.05	P < 0.001	NS P > 0.05	P < 0.001
Q68FY0	QCR1_RAT	Cytochrome b-c1 complex subunit 1, mitochondrial	2.1	1	0.707	0.804	0.791	0.688	0.692	0.788
		Cytochrome b-c1 complex subunit 1, mitochondrial			0.925	0.461	0.219	0.364	0.229	0.109
					P < 0.001	P < 0.001	P < 0.001	P < 0.001	P < 0.001	P < 0.001
Day 14 minima
Q6XVN8	MLP3A_RAT	Microtubule-associated proteins 1A/1B light chain 3A	11.6	1	0.741	0.689	0.600	1.057	0.593	0.765
		Microtubule-associated proteins 1A/1B light chain 3A			0.299		0.183	0.365	0.135	0.335
					P < 0.001		P < 0.001	NS P > 0.05	P < 0.001	P < 0.001
P24623	CRYAA_RAT	Alpha-crystallin A chain	11.7	3	0.623	0.693	0.673	0.636	0.612	0.686
					0.719	0.277	0.884	0.428	0.941	0.690
					NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	NS P > 0.05	P < 0.05
Q6AY84	SCRN1_RAT	Secernin-1	2.2	1	0.838	1.026	0.888	0.684	0.644	0.767
					0.499	0.797	0.332	0.004	0.162	1.085
					P < 0.05	NS P > 0.05	NS P > 0.05	P < 0.001	P < 0.001	P < 0.001

Changes in protein levels as quantified by isobaric tags for relative and absolute quantification (iTRAQ) proteomics at 1, 3, 4, 7, 14, or 21 days postaxotomy. The value at each time point was derived from a pooled sample of eight axotomized retinas, and represents the averaged relative fold-change with respect to the baseline sample of eight pooled normal retinas (value of 1) that was simultaneously processed in the same iTRAQ/ mass spectrometry (MS) run. The 95% confidence interval (CI) is presented below the average relative fold-change, and the P value of Tukey's post hoc comparisons relative to glyceraldehydes-3-phosphate dehydrogenase (GAPDH) at each time point is presented below the 95% CI. The iTRAQ experiment was performed in duplicate and the mean value for each time point was calculated for proteins that were identified in both iTRAQ runs: this reduces the chance of a false-positive identification from ∼ 1/20 to ∼ 1/400. Proteins were grouped according to the day-postaxotomy that had the greatest upregulation (maxima) or downregulation (minima) relative to normal. The 95% CI (± value listed below the mean fold-change for each time point) was calculated based on the two runs, and used to test for statistical significance against the corresponding mean ± 95% CI for GAPDH, a housekeeping gene whose expression remained constant after axotomy. ANOVA followed by Tukey's post hoc tests were used to assess statistical significance at each time point, and the P value for Tukey's test is listed below the 95% CI at each time point. Increases in mean protein levels of greater than 1.3× normal are marked in bold, and decreases of less than 0.7× normal are shown in italic. Avg., average; % Cov (95), percentage of protein covered by identified sequences at a 95% confidence level; No. of Pep, number of peptides identified for each protein. Uniprot, Universal Protein Resource.

Results

Identification and Quantitation of Differentially Expressed Proteins after RGC Axotomy

We used iTRAQ proteomics to identify and quantify proteins at 1, 3, 4, 7, 14, or 21 days after axotomy. These time points represent early signaling events that precede the initiation of RGC death (1 day), events immediately preceding cell death (3 days), the initiation phase of RGC degeneration (4 days), the peak of RGC DNA fragmentation (7 days), the end of the rapid phase of degeneration (14 days), and the protracted death phase (21 days). A total of 216 proteins were identified and quantified at an FDR of 5% (1% FDR = 204 proteins; 10% FDR = 226 proteins; see Supplementary Figs. S1–4). Seventy-one proteins showed expression changes (>1.3× or <0.7×) relative to the pooled sample of normal-unlesioned retinas. Of those proteins, 52 had expression peaks at minimum of one time point, whereas 19 showed a downregulation.

PANTHER was used to analyze axotomy-induced changes in the proportions of proteins based on cellular localization (Fig. 1). With respect to all identified proteins, 17 cellular locations were overrepresented relative to the full rat-proteome. These included tubulin complex, ATP synthase complex, and heterotrimeric G-protein complex (Fig. 1A). Extracellular matrix and extracellular region were among the underrepresented locations. When examining the proteins that were upregulated after axotomy, 19 cellular locations were overrepresented and 11 were underrepresented (Fig. 1B). Among the overrepresented cellular locations were tubulin complex, ATP synthase complex, and ribonucleoprotein complex, whereas heterotrimeric G-protein complex and vesicle coat were underrepresented. In contrast, 11 cellular locations were overrepresented in the downregulated group, including heterotrimeric G-protein complex, vesicle coat, and microtubule (Fig. 1B). Among the 19 categories of underrepresented proteins were tubulin complex, ATP synthase complex, and ribonucleoprotein complex.

Figure 1.

$Relative differences in protein distribution (by cellular location) after optic nerve transection. PANTHER analysis of iTRAQ data showing the relative difference (observed versus expected) in protein distribution by cell location, compared with the entire Rattus norvegicus proteome. Graphs indicate the log-fractional difference between the number of proteins that were identified by iTRAQ (observed) and the normal distribution (expected) for the rat proteome (calculated as: [number of observed proteins in a category – number of expected proteins]/number of expected proteins). The positive bars indicate categories that are overrepresented in each of the iTRAQ lists, whereas the negative bars indicate underrepresented categories. Three lists were analyzed: (A) the entire set of 216 proteins (5% FDR) that was identified after axotomy, (B) the proteins that were upregulated at a minimum of one time point.$

View Original Download Slide

Relative differences in protein distribution (by cellular location) after optic nerve transection. PANTHER analysis of iTRAQ data showing the relative difference (observed versus expected) in protein distribution by cell location, compared with the entire Rattus norvegicus proteome. Graphs indicate the log-fractional difference between the number of proteins that were identified by iTRAQ (observed) and the normal distribution (expected) for the rat proteome (calculated as: [number of observed proteins in a category – number of expected proteins]/number of expected proteins). The positive bars indicate categories that are overrepresented in each of the iTRAQ lists, whereas the negative bars indicate underrepresented categories. Three lists were analyzed: (A) the entire set of 216 proteins (5% FDR) that was identified after axotomy, (B) the proteins that were upregulated at a minimum of one time point.

Identified proteins were also categorized by molecular function (Fig. 2). In the axotomy group (Fig. 2A), transaldolase activity, transketolase activity, and amino acid kinase activity were some of the overrepresented functions, whereas G-protein coupled receptor activity, transcription factor activity, and transcription regulator activity were underrepresented. Analysis of the upregulated proteins revealed 27 overrepresented molecular functions, including transaldolase activity, transketolase activity, and antioxidant activity, and 15 underrepresented functions, such as carbohydrate kinase activity, transcription factor activity, and transcription regulator activity (Fig. 2B). With respect to the downregulated proteins (Fig. 2B), functions with the highest overrepresentation were phosphatase inhibitor activity, peroxidase activity, and phosphatase regulator activity. Transcription factor activity, transcription regulator activity, and receptor activity had the most pronounced underrepresentation.

Figure 2.

$Relative differences in protein distribution by molecular function, following optic nerve axotomy. PANTHER analysis of iTRAQ data showing the relative difference (observed versus expected) in protein distribution by molecular function, compared with the entire Rattus norvegicus proteome. Graphs show the log-fractional difference between the number of proteins that were identified by iTRAQ (observed) and the normal distribution (expected) for the rat proteome (calculated as: [number of observed proteins in a category – number of expected proteins]/number of expected proteins). Positive bars indicate categories that are overrepresented in each of the iTRAQ lists, whereas the negative bars indicate underrepresented categories. Three lists were analyzed: (A) the entire set of 216 proteins (5% FDR) that was identified after crush, (B) the proteins that were upregulated at a minimum of one time point (blue bars), and the proteins that were downregulated at a minimum of one time point (red bars).$

View Original Download Slide

Relative differences in protein distribution by molecular function, following optic nerve axotomy. PANTHER analysis of iTRAQ data showing the relative difference (observed versus expected) in protein distribution by molecular function, compared with the entire Rattus norvegicus proteome. Graphs show the log-fractional difference between the number of proteins that were identified by iTRAQ (observed) and the normal distribution (expected) for the rat proteome (calculated as: [number of observed proteins in a category – number of expected proteins]/number of expected proteins). Positive bars indicate categories that are overrepresented in each of the iTRAQ lists, whereas the negative bars indicate underrepresented categories. Three lists were analyzed: (A) the entire set of 216 proteins (5% FDR) that was identified after crush, (B) the proteins that were upregulated at a minimum of one time point (blue bars), and the proteins that were downregulated at a minimum of one time point (red bars).

Temporal Changes in the Retinal Proteome at Key Time Points during RGC Degeneration

When we examined each time point for peaks and depressions in protein levels, we uncovered many expression spikes over the course of RGC degeneration. These data are detailed in Table 1. At 1 day after axotomy, several proteins showed a marked increase in expression, including histone H4 (deacetylation plays a critical role in neuronal apoptosis)³⁷ (Table 1). Downregulated proteins included microtubule-associated protein RP/EB family member 1 and cytochrome c oxidase polypeptide 6A1 (suppresses Bax-induced cell death)³⁸ (Table 1).

On day 3, 10 kDa heat shock protein, peroxiredoxin-2, and cofilin-1 spiked upward (Fig. 3B). Interestingly, Purkinje cell protein 4, which regulates calmodulin signaling, was also upregulated. This protein shows reduced expression in Alzheimer's disease and Huntington's disease.³⁹ Conversely, proteins such as beta-crystallin S, which promotes axon regeneration,⁴⁰ and the small nuclear ribonucleoprotein-associated protein N showed marked depression (Table 1).

Figure 3.

View Original Download Slide

HDGF regulation and signal transduction after axotomy. (A) Protein expression profiles of HDGF and HDGF Related Protein-3 showed early increases after RGC axotomy, followed by sharp decreases at 4 days postaxotomy, when RGCs begin to die. Levels then increased at 7 days when RGC death peaks. These patterns are suggestive of an intrinsic protective response within the retina. (B) Western blots showing increases in HDGF levels at 7 days after axotomy, relative to normal retinas. On the right, densitometry was performed, showing a significant increase in HDGF levels at 7 days postaxotomy (**P < 0.01). Results were obtained by normalizing the HDGF band density against the corresponding GAPDH band density. The mean density for each group (axotomy versus normal) ± SEM was then calculated, and statistical significance was assessed by ANOVA followed by Tukey's post hoc comparisons. GAPDH loading controls are shown below each HDGF blot. (C) Western blots showing Akt and MAPK phosphorylation (activation) at 24 hours after intraocular injection of HDGF. HDGF increased Akt phosphorylation relative to controls, but did not have a noticeable effect on MAPK phosphorylation at 24 hours. Total Akt and total MAPK levels did not change significantly. On the right, densitometry was performed to compare means ± SEM between HDGF treated retinas and controls. **P < 0.01 relative to control.

Proteins with expression peaks at day 4 included hydroxyacylglutathione hydrolase, superoxide dismutase (Cu-Zn), and basigin (Table 1). In contrast, guanine nucleotide-binding protein beta-5 (regulates neuronal inhibitory signals),⁴¹ T-complex protein 1 subunit alpha, and hexokinase-2 were noticeably underexpressed (Table 1).

At 7 days, upward expression spikes in the levels of thioredoxin, dihydropyrimidinase-related protein 4, and eukaryotic translation initiation factor 5A-1 were observed (Table 1). Rab GDP dissociation inhibitor alpha (regulates proteins involved in neurotransmission and mutations have been associated with human X-linked mental retardation associated with epileptic seizures)^42,43 also peaked on day 7. In contrast, clathrin light chain B, nucleosome assembly protein 1-like 4, as well as beta-crystallin B2, which is involved in axon elongation during regeneration,⁴⁰ showed marked depressions at the 7-day time point (Table 1).

By 14 days postaxotomy, 90% of have degenerated after optic nerve transaction.^5,11–13 Upregulated proteins at this time point included interphotoreceptor matrix proteoglycan 1, proteasome subunit beta type-2, and heterogeneous nuclear ribonucleoproteins C and K. Microtubule-associated proteins 1A/1B light chain 3A, alpha-crystallin A chain, and secernin-1 were downregulated at 14 days postaxotomy (Table 1).

Several proteins were maximally expressed at 21 days postaxotomy; these included both histone H2A type 1-F and type 1-E, calnexin, s-arrestin, transketolase, and vimentin (Table 1).

Although many proteins showed expression changes after axotomy, several others, including the housekeeping gene GAPDH and gamma-enolase, remained constant (Table 1). Thus, axotomy induces salient changes in protein expression, with proteins often being drastically regulated at multiple time points along the axis of RGC death. These proteins represent potential targets for future therapeutic interventions to prevent RGC death.

Effect of HDGF on the Survival of Injured Adult RGCs

We examined the iTRAQ data for changes in the levels of soluble factors that could be delivered to the retina via intraocular injections, with the aim of RGC neuroprotection in mind. The expression of HDGF and HDGF-related protein 3 (HDGFRP3) showed regulation peaks after axotomy (Fig. 3A). Interestingly, the levels of HDGFRP3 peaked at 3 days postaxotomy, when RGCs first begin to die, whereas HDGF levels peaked at 7 days postaxotomy. Western blots also showed significantly higher levels of HDGF at 7 days (Fig. 3B), confirming the iTRAQ findings.

Previous studies have shown similar expression patterns for brain-derived neurotrophic factor and glial cell-line–derived neurotrophic factor after axotomy,^44,45 and that additional exogenous delivery of these trophic factors is neuroprotective.^7,46–49 As such, we evaluated whether retinal signal transduction was altered by exogenous delivery of HDGF. The PI3K and MAPK pathways are critical for RGC survival following axotomy.^34,50–54 PI3K activity can be assessed by looking at the PI3K-dependent phosphorylation of Akt (protein kinase B) at serine-473, and MAPK activation can be evaluated by examining MEK-dependent phosphorylation of MAPK at threonine 202/tyrosine 204 (Thr202/Tyr204). At 24 hours after intraocular injection of HDGF, we observed increased phosphorylation of Akt, but no difference in MAPK phosphorylation, whereas the total levels of Akt and MAPK remained constant (Fig. 3C). These findings show that HDGF modulates key survival pathways in the adult retina.

To examine the effect of HDGF on RGC degeneration, we administered an intraocular injection of HDGF at 3 days postaxotomy and RGC survival was quantified at 14 days. HDGF (Figs. 4B, 4D) significantly increased RGC survival by approximately 4-fold compared with control delivery of PBS. We then evaluated whether the effects of HDGF were dependent on activation of the PI3K or MAPK pathways, by codelivery of PI3K or MAPK inhibitors. When the PI3K inhibitor Wortmannin was coadministered with HDGF, RGC survival was significantly reduced in the middle and outer retina (Fig. 4D). The MAPK pathway inhibitor PD98059 induced a more pronounced depression in RGC rescue after HDGF delivery (Figs. 4C, 4D). Control treatment with PD98059 alone resulted in a small significant increase in RGC survival compared with DMSO (vehicle), whereas control treatment with Wortmannin alone had no effect (Fig. 4D). These findings show that HDGF-mediated neuroprotection in axotomized RGCs is dependent largely on MAPK function, with a lesser involvement of the PI3K-Akt signaling cascade.

Figure 4.

View Original Download Slide

Effect of HDGF on RGC survival after axotomy. (A–D) Epifluorescence micrographs of flat-mounted retinas, showing Fluorogold retrogradely labeled RGCs at 14 days after optic nerve transection and intraocular injection of DMSO (A), intraocular delivery of HDGF (B), or combined delivery of HDGF and the MAPK inhibitor PD98059 (C). Bar, 50 μm. (E) Quantification of the density of surviving retinal ganglion cells (mean ± SEM) at 14 days postaxotomy. Cell densities were quantified at three different retinal eccentricities (inner, mid-periphery, outer). Intraocular injection of HDGF significantly increased RGC survival after optic nerve transection (A = P < 0.01 relative to Control-PBS or DMSO). Codelivery of the PI3K inhibitor Wortmannin reduced the protective effects of HDGF in the middle and outer retina (B = P < 0.01 relative to HDGF alone), whereas there was no significant difference in the inner retina. The MAPK pathway inhibitor PD98059 significantly reduced the protective effects of HDGF throughout the retina (P < 0.01). PD98059 alone produced a minor increase in RGC survival relative to the DMSO control (C = P < 0.05 relative to DMSO). Wortmannin did not have a significant effect on RGC survival relative to DMSO.

Discussion

We performed a global study of protein expression in the retina because glial cells, surrounding neurons, and invading immune cells are known to participate in RGC degeneration after injury.^10,20,55,56 Our findings suggest that there are wide reaching changes in retinal physiology that extend beyond the ganglion cell layer following axotomy, and the proteins that we identified are potential targets for neuroprotection research.

The use of iTRAQ labeling reagents together with MS has several advantages over other proteome profiling techniques: (1) protein samples labeled with iTRAQ reagents are combined and concurrently analyzed by MS for peptide identification and quantitation,⁵⁷ allowing different protein samples to be processed in the same run, thereby negating intra-experimental variability⁵⁸; (2) iTRAQ provides a greater range of protein expression ratios, with more consistent quantitation, and internal controls in each run, unlike other protein-profiling techniques⁵⁸; (3) iTRAQ is also more sensitive than ICAT (isotope-coded affinity tags) and 2D-DIGE (two-dimensional difference in gel electrophoresis).⁵⁸ Accordingly, iTRAQ analysis is a powerful technique for studying changes in the retinal proteome after axotomy.

We previously used iTRAQ proteomics to study the expression of retinal proteins after optic nerve crush,³⁵ revealing several similarities with the present study. Both axotomy and crush increased the proportion of proteins in the tubulin complex, soluble N-ethylmaleimide sensitive factor attachment protein receptor (SNARE) complex, and proton-transporting ATP synthase families relative to the normal rat proteome. Additionally, extracellular matrix and the extracellular region families were reduced in proportion, in both injury models. Surprisingly, there were notable discrepancies between proteome profiles obtained after optic nerve axotomy and crush.³⁵ For instance, axotomy and crush-regulated actin cytoskeletal components in an inverse fashion: in axotomized retinas, the upregulated proteins showed an overrepresentation of actin cytoskeleton, whereas upregulated proteins in crush samples had a reduced proportion of actin cytoskeleton. This effect was reversed for the downregulated proteins in these two models. Among the downregulated proteins, axotomized retinas contained fewer mitochondrial proteins than the normal rat proteome, whereas crush retinas contained more. There were also salient differences in molecular functions, such as SNAP receptor activity, anion channel activity, antioxidant activity, isomerase activity, and peroxidase activity, between the two models. In addition, several molecular functions had altered protein proportions after axotomy, but remained normal after crush. These included actin binding, cytoskeletal protein binding, enzyme inhibitor activity, growth factor activity, microtubule binding, and nucleic acid binding. Thus, the retinal proteome profiles of optic nerve axotomy versus crush differ substantially from one another.

Discrepancies in protein expression between axotomy and crush have also been observed after microarray analysis. A previously published study revealed that transcriptome regulation after axotomy was 36.2% greater than crush.⁵⁹ More specifically, 1117 genes were uniquely regulated after axotomy, 293 genes were uniquely regulated after optic nerve crush, and 661 genes were commonly regulated after both injuries.⁵⁹ Furthermore, gene regulation was more transient after crush, whereas axotomy tended to trigger regulation at multiple time points for any particular gene,⁵⁹ similar to our current findings.

There are several possible reasons for these discrepancies in protein expression between axotomy and crush, despite the fact that these forms of injury appear to be relatively similar. First, the extent of RGC death differs between these two models. By 14 days postinjury, axotomy causes the death of approximately 90% of the RGC population,^5,11–13 whereas only approximately 70% of RGCs die after crush.⁶⁰ Second, axotomy involves the complete transection of all RGC axons, whereas optic nerve crush induces varying degrees of axon damage throughout the RGC population. This graded axon injury contributes to the secondary degeneration of incompletely injured RGCs via the release of toxic compounds from the degenerating fibers in the crushed nerve.⁶¹

Microarray analysis of retinas in a rat glaucoma model⁶² also showed similar patterns to our axotomy data. Guo et al.⁶² compared gene microarrays of whole retinas to that of the retinal ganglion cell layer (RGCL) isolated by laser capture microdissection. In the whole retina, they found a high probability of upregulated proteins in the cytoskeleton, protein complex, receptor activity, and transporter activity, whereas downregulated proteins belonged to the cytoskeletal and the cytoplasmic classes,⁶² similar to our axotomy results. When comparing their RGCL results to our axotomy analysis, there were also similarities in the increase of upregulated proteins in the ribonucleoprotein complex, and cytoskeleton families. Furthermore, increased proportions of downregulated genes having oxidoreductase activity, hydrogen ion transporter activity, and cation transporter activity in the RGCL were similar to the present study. The previous findings⁶² and ours suggest that gene/protein regulation after axotomy is relatively similar to elevated IOP in rats, and substantially different from optic nerve crush. Another recent proteomic study in primates⁶³ reported that axotomy-induced changes differ from that of elevated IOP; however, this discrepancy may be due to differences in experimental design. Stowell et al.⁶³ sampled retinas at 21 days after axotomy in primates, whereas our analysis used multiple samplings at 1, 3, 4, 7, 14, and 21 days postaxotomy in rats. This is important considering that protein levels were drastically regulated after axotomy, and often cycled from up- to downregulated within a single day. Additionally, in the rat, apoptosis begins at approximately 4 days postaxotomy and 90% of RGCs die by 14 days; in the primate, RGC death begins between 2 and 4 weeks after axotomy, and continues for approximately 16 weeks until most cells have died.^64,65 Furthermore, the kinetics of RGC death and the sensitivity of RGCs to elevated IOP differ between primates and rats.⁶⁶ Together, these studies are helping to build our understanding of global processes that are involved in RGC degeneration; nevertheless, further work is required to understand the complexities that account for the observed differences. Future research directed at uncovering the pro- or antiapoptotic roles of potential targets identified in the present study and previous articles will aid in this regard.

HDGF belongs to a family of proteins containing a well-conserved N-terminal amino acid sequence,^67,68 which includes the HDGF-related proteins (HDGFRP-1 to -4) and lens epithelial factor.⁶⁹ Although HDGF, HDGFRP-2, and HDGFRP-3 are expressed in developing neurons, their biological functions in the adult central nervous system are not well defined.⁶⁹ It has been reported that HDGF is localized to the nuclei of neurons, astrocytes, and oligodendrocytes in adulthood.⁷⁰ HDGFRP-3 has also been found to promote neurite outgrowth in mouse cortical neurons through a direct interaction with tubulin.⁶⁹ Outside the nervous system, HDGF knockdown induces Bad- and Fas-mediated apoptosis in human cancer cells,²⁸ and has growth-stimulating activity in fibroblasts, hepatoma cells, vascular smooth muscle cells, and endothelial cells.^67,71–73 It has been shown that the mitogenic activity of HDGF is dependent on nuclear localization or translocation.^74,75 Other identified roles for HDGF include nephrogenesis, vascular development, and tumorigenesis.^72,74,76 HDGF, like other trophic factors, has been postulated to act via receptor binding at the cell surface; however, the HDGF receptor(s) has yet to be identified. Despite this, it was recently demonstrated that residues 81 to 100 are critical for binding to the cell surface and stimulating proliferation in 3T3 fibroblasts.⁷⁷ Furthermore, heparin sulfate on the cell surface appears to be necessary for HDGF to exert biological effects that are partially dependent on activation of the MAPK pathway.⁷⁸ Our iTRAQ proteomics show that HDGF is expressed in the adult rat retina, in addition to previous data showing that HDGF is expressed in porcine fetal retinal pigment epithelium.⁷⁹ Furthermore, the present findings also show that HDGF has biological effects in the adult retina, as a potent neuroprotective factor.

In the present study, intraocular delivery of HDGF significantly increased RGC survival after axotomy. Although HDGF induces activation of the PI3K-Akt pathway in the retina, this cascade contributes only a small component of the neuroprotective response induced by HDGF. Specifically, only cell survival in the middle and outer retina was dependent on PI3K activity, suggesting that HDGF may have differing effects on specific subclasses of RGCs. HDGF has previously been reported to increase MAPK phosphorylation in cancer cells^78,80,81; however, we did not observe this effect in our study. Nevertheless, MAPK activity appears to be critical for HDGF-mediated rescue of RGCs, as PD98059 noticeably antagonized the protective effects of HDGF. It is possible that MAPK activation by HDGF is transient, or that an accessory function of MAPK that is independent of phosphorylation at Thr202/Tyr204 is necessary for HDGF signal transduction. Overall, HDGF shows robust survival-promoting effects on injured adult RGCs.

Many of the proteins revealed in our study are novel potential targets for therapeutic interventions directed at preventing RGC degeneration and visual impairment. For example, we found that cytochrome c oxidase subunit 6A1 (Cox6a1) levels were reduced at several time points after axotomy. Cox6a1 has previously been shown to abrogate Bax-induced cell death in yeast,⁸² and Bax is a key player in RGC death.⁸³ Thus, overexpression of Cox6a1 in axotomized RGCs may hold therapeutic potential if introduced shortly after optic nerve insult. Other proteins that spiked at key time points after axotomy represent potential leads in the search for therapeutic strategies to treat RGC loss in visual diseases.

Supplementary Materials

pdf S1

Acknowledgments

The authors thank Jian Chen and Peihong Zhu at Ontario Cancer Biomarker Network for their help in the iTRAQ analysis.

References

Cheung N Mitchell P Wong TY . Diabetic retinopathy. Lancet . 2010;376:124–136. [CrossRef] [PubMed]

Milea D Amati-Bonneau P Reynier P Bonneau D . Genetically determined optic neuropathies. Curr Opin Neurol . 2010;23:24–28. [CrossRef] [PubMed]

Qu J Wang D Grosskreutz CL . Mechanisms of retinal ganglion cell injury and defense in glaucoma. Exp Eye Res . 2010;91:48–53. [CrossRef] [PubMed]

Warner N Eggenberger E . Traumatic optic neuropathy: a review of the current literature. Curr Opin Ophthalmol . 2010;21:459–462. [CrossRef] [PubMed]

Berkelaar M Clarke DB Wang YC Bray GM Aguayo AJ . Axotomy results in delayed death and apoptosis of retinal ganglion cells in adult rats. J Neurosci . 1994;14:4368–4374. [PubMed]

Quigley HA Nickells RW Kerrigan LA Pease ME Thibault DJ Zack DJ . Retinal ganglion cell death in experimental glaucoma and after axotomy occurs by apoptosis. Invest Ophthalmol Vis Sci . 1995;36:774–786. [PubMed]

Koeberle PD Ball AK . Effects of GDNF on retinal ganglion cell survival following axotomy. Vision Res . 1998;38:1505–1515. [CrossRef] [PubMed]

Kermer P Klocker N Bahr M . Neuronal death after brain injury. Models, mechanisms, and therapeutic strategies in vivo. Cell Tissue Res . 1999;298:383–395. [CrossRef] [PubMed]

Bahr M . Live or let die—retinal ganglion cell death and survival during development and in the lesioned adult CNS. Trends Neurosci . 2000;23:483–490. [CrossRef] [PubMed]

10.

Koeberle PD Bahr M . Growth and guidance cues for regenerating axons: where have they gone? J Neurobiol . 2004;59:162–180. [CrossRef] [PubMed]

11.

Villegas-Perez MP Vidal-Sanz M Bray GM Aguayo AJ . Influences of peripheral nerve grafts on the survival and regrowth of axotomized retinal ganglion cells in adult rats. J Neurosci . 1988;8:265–280. [PubMed]

12.

Villegas-Perez MP Vidal-Sanz M Rasminsky M Bray GM Aguayo AJ . Rapid and protracted phases of retinal ganglion cell loss follow axotomy in the optic nerve of adult rats. J Neurobiol . 1993;24:23–36. [CrossRef] [PubMed]

13.

Peinado-Ramon P Salvador M Villegas-Perez MP Vidal-Sanz M . Effects of axotomy and intraocular administration of NT-4, NT-3, and brain-derived neurotrophic factor on the survival of adult rat retinal ganglion cells. A quantitative in vivo study. Invest Ophthalmol Vis Sci . 1996;37:489–500. [PubMed]

14.

Isenmann S Wahl C Krajewski S Reed JC Bähr M . Up-regulation of Bax protein in degenerating retinal ganglion cells precedes apoptotic cell death after optic nerve lesion in the rat. Eur J Neurosci . 1997;9:1763–1772. [CrossRef] [PubMed]

15.

Kermer P Ankerhold R Klöcker N Krajewski S Reed JC Bähr M . Caspase-9: involvement in secondary death of axotomized rat retinal ganglion cells in vivo. Brain Res Mol Brain Res . 2000;85:144–150. [CrossRef] [PubMed]

16.

Kermer P Klöcker N Labes M Bähr M . Inhibition of CPP32-like proteases rescues axotomized retinal ganglion cells from secondary cell death in vivo. J Neurosci . 1998;18:4656–4662. [PubMed]

17.

Kikuchi M Tenneti L Lipton SA . Role of p38 mitogen-activated protein kinase in axotomy-induced apoptosis of rat retinal ganglion cells. J Neurosci . 2000;20:5037–5044. [PubMed]

18.

Kuehn MH Fingert JH Kwon YH . Retinal ganglion cell death in glaucoma: mechanisms and neuroprotective strategies. Ophthalmol Clin North Am . 2005;18:383–395, vi. [CrossRef] [PubMed]

19.

Seki M Lipton SA . Targeting excitotoxic/free radical signaling pathways for therapeutic intervention in glaucoma. Prog Brain Res . 2008;173:495–510. [PubMed]

20.

Koeberle PD Ball AK . Nitric oxide synthase inhibition delays axonal degeneration and promotes the survival of axotomized retinal ganglion cells. Exp Neurol . 1999;158:366–381. [CrossRef] [PubMed]

21.

Koeberle PD Gauldie J Ball AK . Effects of adenoviral-mediated gene transfer of interleukin-10, interleukin-4, and transforming growth factor-beta on the survival of axotomized retinal ganglion cells. Neuroscience . 2004;125:903–920. [CrossRef] [PubMed]

22.

Koeberle PD Wang Y Schlichter LC . Kv1.1 and Kv1.3 channels contribute to the degeneration of retinal ganglion cells after optic nerve transection in vivo. Cell Death Differ . 2010;17:134–144. [CrossRef] [PubMed]

23.

Kipnis J Yoles E Schori H Hauben E Shaked I Schwartz M . Neuronal survival after CNS insult is determined by a genetically encoded autoimmune response. J Neurosci . 2001;21:4564–4571. [PubMed]

24.

Schori H Yoles E Schwartz M . T-cell-based immunity counteracts the potential toxicity of glutamate in the central nervous system. J Neuroimmunol . 2001;119:199–204. [CrossRef] [PubMed]

25.

Schori H Yoles E Wheeler LA Raveh T Kimchi A Schwartz M . Immune-related mechanisms participating in resistance and susceptibility to glutamate toxicity. Eur J Neurosci . 2002;16:557–564. [CrossRef] [PubMed]

26.

Koeberle PD Schlichter LC . Targeting K(V) channels rescues retinal ganglion cells in vivo directly and by reducing inflammation. Channels (Austin) . 2010;4:337–346. [CrossRef] [PubMed]

27.

Koeberle PD Tura A Tassew NG Schlichter LC Monnier PP . The repulsive guidance molecule, RGMa, promotes retinal ganglion cell survival in vitro and in vivo. Neuroscience . 2010;169:495–504. [CrossRef] [PubMed]

28.

Tsang TY Tang WY Tsang WP Co NN Kong SK Kwok TT . Mechanistic study on growth suppression and apoptosis induction by targeting hepatoma-derived growth factor in human hepatocellular carcinoma HepG2 cells. Cell Physiol Biochem . 2009;24:253–262. [CrossRef] [PubMed]

29.

Magharious MM D'Onofrio PM Koeberle PD . Methods for experimental manipulations after optic nerve transection in the mammalian CNS. J Vis Exp . 2011;(51):2261.

30.

Magharious MM D'Onofrio PM Koeberle PD . Optic nerve transection: a model of adult neuron apoptosis in the central nervous system. J Vis Exp . 2011;(51):2241.

31.

Arcaro A Wymann MP . Wortmannin is a potent phosphatidylinositol 3-kinase inhibitor: the role of phosphatidylinositol 3,4,5-trisphosphate in neutrophil responses. Biochem J . 1993;296:297–301. [PubMed]

32.

Powis G Bonjouklian R Berggren MM Wortmannin, a potent and selective inhibitor of phosphatidylinositol-3-kinase. Cancer Res . 1994;54:2419–2423. [PubMed]

33.

Alessi DR Cuenda A Cohen P Dudley DT Saltiel AR. PD . 098059 is a specific inhibitor of the activation of mitogen-activated protein kinase kinase in vitro and in vivo. J Biol Chem . 1995;270:27489–27494. [CrossRef] [PubMed]

34.

Koeberle PD Bahr M . The upregulation of GLAST-1 is an indirect antiapoptotic mechanism of GDNF and neurturin in the adult CNS. Cell Death Differ . 2008;15:471–483. [CrossRef] [PubMed]

35.

Magharious M D'Onofrio PM Hollander A Zhu P Chen J Koeberle PD . Quantitative iTRAQ analysis of retinal ganglion cell degeneration after optic nerve crush. J Proteome Res . 2011;10:3344–3362. [CrossRef] [PubMed]

36.

Tang WH Shilov IV Seymour SL . Nonlinear fitting method for determining local false discovery rates from decoy database searches. J Proteome Res . 2008;7:3661–3667. [CrossRef] [PubMed]

37.

Norris EH Giasson BI Lee VM . Alpha-synuclein: normal function and role in neurodegenerative diseases. Curr Top Dev Biol . 2004;60:17–54. [PubMed]

38.

Keppler D Sordat B Sierra F . T-kininogen present in the liver of old rats is biologically active and readily forms complexes with endogenous cysteine proteinases. Mech Ageing Dev . 1997;98:151–165. [CrossRef] [PubMed]

39.

Harashima S Wang Y Horiuchi T Seino Y Inagaki N . Purkinje cell protein 4 positively regulates neurite outgrowth and neurotransmitter release. J Neurosci Res . 2011;89:1519–1530. [CrossRef] [PubMed]

40.

Liedtke T Schwamborn JC Schröer U Thanos S . Elongation of axons during regeneration involves retinal crystallin beta b2 (crybb2). Mol Cell Proteomics . 2007;6:895–907. [CrossRef] [PubMed]

41.

Shin OH Xu J Rizo J Südhof TC . Differential but convergent functions of Ca2+ binding to synaptotagmin-1 C2 domains mediate neurotransmitter release. Proc Natl Acad Sci U S A . 2009;106:16469–16474. [CrossRef] [PubMed]

42.

Ishizaki H Miyoshi J Kamiya H Role of rab GDP dissociation inhibitor alpha in regulating plasticity of hippocampal neurotransmission. Proc Natl Acad Sci U S A . 2000;97:11587–11592. [CrossRef] [PubMed]

43.

D'Adamo P Menegon A Lo Nigro C Mutations in GDI1 are responsible for X-linked non-specific mental retardation. Nat Genet . 1998;19:134–139. [CrossRef] [PubMed]

44.

Hirsch S Labes M Bahr M . Changes in BDNF and neurotrophin receptor expression in degenerating and regenerating rat retinal ganglion cells. Restor Neurol Neurosci . 2000;17:125–134.

45.

Lindqvist N Peinado-Ramónn P Vidal-Sanz M Hallböök F. GDNF . Ret, GFRalpha1 and 2 in the adult rat retino-tectal system after optic nerve transection. Exp Neurol . 2004;187:487–499. [CrossRef] [PubMed]

46.

Klocker N Cellerino A Bahr M . Free radical scavenging and inhibition of nitric oxide synthase potentiates the neurotrophic effects of brain-derived neurotrophic factor on axotomized retinal ganglion cells in vivo. J Neurosci . 1998;18:1038–1046. [PubMed]

47.

Koeberle PD Ball AK . Neurturin enhances the survival of axotomized retinal ganglion cells in vivo: combined effects with glial cell line-derived neurotrophic factor and brain-derived neurotrophic factor. Neuroscience . 2002;110:555–567. [CrossRef] [PubMed]

48.

Mansour-Robaey S Clarke DB Wang YC Bray GM Aguayo AJ . Effects of ocular injury and administration of brain-derived neurotrophic factor on survival and regrowth of axotomized retinal ganglion cells. Proc Natl Acad Sci U S A . 1994;91:1632–1636. [CrossRef] [PubMed]

49.

Mey J Thanos S . Intravitreal injections of neurotrophic factors support the survival of axotomized retinal ganglion cells in adult rats in vivo. Brain Res . 1993;602:304–317. [CrossRef] [PubMed]

50.

Nakazawa T Tamai M Mori N . Brain-derived neurotrophic factor prevents axotomized retinal ganglion cell death through MAPK and PI3K signaling pathways. Invest Ophthalmol Vis Sci . 2002;43:3319–3326. [PubMed]

51.

Park K Luo JM Hisheh S Harvey AR Cui Q . Cellular mechanisms associated with spontaneous and ciliary neurotrophic factor-cAMP-induced survival and axonal regeneration of adult retinal ganglion cells. J Neurosci . 2004;24:10806–10815. [CrossRef] [PubMed]

52.

Pernet V Hauswirth WW Di Polo A . Extracellular signal-regulated kinase 1/2 mediates survival, but not axon regeneration, of adult injured central nervous system neurons in vivo. J Neurochem . 2005;93:72–83. [CrossRef] [PubMed]

53.

Sapieha PS Hauswirth WW Di Polo A . Extracellular signal-regulated kinases 1/2 are required for adult retinal ganglion cell axon regeneration induced by fibroblast growth factor-2. J Neurosci Res . 2006;83:985–995. [CrossRef] [PubMed]

54.

Zhou Y Pernet V Hauswirth WW Activation of the extracellular signal-regulated kinase 1/2 pathway by AAV gene transfer protects retinal ganglion cells in glaucoma. Mol Ther . 2005;12:402–412. [CrossRef] [PubMed]

55.

Fitzgerald M Bartlett CA Harvey AR Dunlop SA . Early events of secondary degeneration after partial optic nerve transection: an immunohistochemical study. J Neurotrauma . 2010;27:439–452. [CrossRef] [PubMed]

56.

Garcia-Valenzuela E Sharma SC Pina AL . Multilayered retinal microglial response to optic nerve transection in rats. Mol Vis . 2005;11:225–231. [PubMed]

57.

Gafken PR Lampe PD . Methodologies for characterizing phosphoproteins by mass spectrometry. Cell Commun Adhes . 2006;13:249–262. [CrossRef] [PubMed]

58.

Tannu NS Hemby SE . Methods for proteomics in neuroscience. Prog Brain Res . 2006;158:41–82. [PubMed]

59.

Agudo M Pérez-Marín MC Lönngren U Time course profiling of the retinal transcriptome after optic nerve transection and optic nerve crush. Mol Vis . 2008;14:1050–1063. [PubMed]

60.

Selles-Navarro I Ellezam B Fajardo R Latour M McKerracher L . Retinal ganglion cell and nonneuronal cell responses to a microcrush lesion of adult rat optic nerve. Exp Neurol . 2001;167:282–289. [CrossRef] [PubMed]

61.

Schwartz M . Optic nerve crush: protection and regeneration. Brain Res Bull . 2004;62:467–471. [CrossRef] [PubMed]

62.

Guo Y Cepurna WO Dyck JA Doser TA Johnson EC Morrison JC . Retinal cell responses to elevated intraocular pressure: a gene array comparison between the whole retina and retinal ganglion cell layer. Invest Ophthalmol Vis Sci . 2010;51:3003–3018. [CrossRef] [PubMed]

63.

Stowell C Arbogast B Cioffi G Burgoyne C Zhou A . Retinal proteomic changes following unilateral optic nerve transection and early experimental glaucoma in non-human primate eyes. Exp Eye Res . 2011;93:13–28. [CrossRef] [PubMed]

64.

Quigley HA Davis EB Anderson DR . Descending optic nerve degeneration in primates. Invest Ophthalmol Vis Sci . 1977;16:841–849. [PubMed]

65.

66.

Morrison JC . Elevated intraocular pressure and optic nerve injury models in the rat. J Glaucoma . 2005;14:315–317. [CrossRef] [PubMed]

67.

Nakamura H Izumoto Y Kambe H Molecular cloning of complementary DNA for a novel human hepatoma-derived growth factor. Its homology with high mobility group-1 protein. J Biol Chem . 1994;269:25143–25149. [PubMed]

68.

Izumoto Y Kuroda T Harada H Kishimoto T Nakamura H . Hepatoma-derived growth factor belongs to a gene family in mice showing significant homology in the amino terminus. Biochem Biophys Res Commun . 1997;238:26–32. [CrossRef] [PubMed]

69.

El-Tahir HM Dietz F Dringen R Expression of hepatoma-derived growth factor family members in the adult central nervous system. BMC Neurosci . 2006;7:6. [CrossRef] [PubMed]

70.

Zhou Z Yamamoto Y Sugai F Hepatoma-derived growth factor is a neurotrophic factor harbored in the nucleus. J Biol Chem . 2004;279:27320–27326. [CrossRef] [PubMed]

71.

Nakamura H Kambe H Egawa T Partial purification and characterization of human hepatoma-derived growth factor. Clin Chim Acta . 1989;183:273–284. [CrossRef] [PubMed]

72.

Oliver JA Al-Awqati Q . An endothelial growth factor involved in rat renal development. J Clin Invest . 1998;102:1208–1219. [CrossRef] [PubMed]

73.

Everett AD Lobe DR Matsumura ME Nakamura H McNamara CA . Hepatoma-derived growth factor stimulates smooth muscle cell growth and is expressed in vascular development. J Clin Invest . 2000;105:567–575. [CrossRef] [PubMed]

74.

Everett AD Stoops T McNamara CA . Nuclear targeting is required for hepatoma-derived growth factor-stimulated mitogenesis in vascular smooth muscle cells. J Biol Chem . 2001;276:37564–37568. [CrossRef] [PubMed]

75.

Kishima Y Yamamoto H Izumoto Y Hepatoma-derived growth factor stimulates cell growth after translocation to the nucleus by nuclear localization signals. J Biol Chem . 2002;277:10315–10322. [CrossRef] [PubMed]

76.

Matsuyama A Inoue H Shibuta K Hepatoma-derived growth factor is associated with reduced sensitivity to irradiation in esophageal cancer. Cancer Res . 2001;61:5714–5717. [PubMed]

77.

Abouzied MM El-Tahir HM Prenner L Häberlein H Gieselmann V Franken S . Hepatoma-derived growth factor. Significance of amino acid residues 81-100 in cell surface interaction and proliferative activity. J Biol Chem . 2005;280:10945–10954. [CrossRef] [PubMed]

78.

Wang CH Davamani F Sue SC Cell surface heparan sulfates mediate internalization of the PWWP/HATH domain of HDGF via macropinocytosis to fine-tune cell signalling processes involved in fibroblast cell migration. Biochem J . 2011;433:127–138. [CrossRef] [PubMed]

79.

Kolomeyer AM Sugino IK Zarbin MA . Characterization of conditioned media collected from cultured adult versus fetal retinal pigment epithelial cells. Invest Ophthalmol Vis Sci . 2011;52:5973–5986. [CrossRef] [PubMed]

80.

Everett AD Narron JV Stoops T Nakamura H Tucker A . Hepatoma-derived growth factor is a pulmonary endothelial cell-expressed angiogenic factor. Am J Physiol Lung Cell Mol Physiol . 2004;286:L1194–1201. [CrossRef] [PubMed]

81.

Mao J Xu Z Fang Y Hepatoma-derived growth factor involved in the carcinogenesis of gastric epithelial cells through promotion of cell proliferation by Erk1/2 activation. Cancer Sci . 2008;99:2120–2127. [CrossRef] [PubMed]

82.

Eun SY Woo IS Jang HS . Identification of cytochrome c oxidase subunit 6A1 as a suppressor of Bax-induced cell death by yeast-based functional screening. Biochem Biophys Res Commun . 2008;373:58–63. [CrossRef] [PubMed]

83.

Nickells RW Semaan SJ Schlamp CL . Involvement of the Bcl2 gene family in the signaling and control of retinal ganglion cell death. Prog Brain Res . 2008;173:423–435. [PubMed]

Footnotes

Supported by an operating grant to PDK from the Canadian Institutes of Health Research (MOP 86523).

Footnotes

³ These authors contributed equally to the work presented here and should therefore be considered equivalent authors.

Footnotes

Disclosure: A. Hollander, None; P.M. D'Onofrio, None; M.M. Magharious, None; M.D. Lysko, None; P.D. Koeberle, None