Molecular Vision 2006; 12:1660-1673 <http://www.molvis.org/molvis/v12/a190/> Received 24 February 2006 | Accepted 27 November 2006 | Published 22 December 2006

Download Reprint

Gene expression changes in the retina following optic nerve transection

Natik Piri,^1,2 Jacky M.K. Kwong,¹ Min Song,¹ David Elashoff,³ Joseph Caprioli¹
 
 

¹Jules Stein Eye Institute, ²Brain Research Institute, and ³Department of Biostatistics, University of California Los Angeles School of Medicine, Los Angeles, CA

Correspondence to: Dr. Natik Piri, Jules Stein Eye Institute, UCLA School of Medicine, 100 Stein Plaza, Los Angeles, CA, 90095; Phone: (310) 825-9850; FAX: (310) 794-2144; email: piri@jsei.ucla.edu

Abstract

Purpose: To obtain and analyze the gene expression profiles of the retina following optic nerve transection (ONT).

Methods: An axotomy animal model was generated by taking a cross section of the optic nerve with care not to damage the adjacent blood supply. The extent of cell loss was evaluated by counting cells in the ganglion cell layer (GCL) of flat-mounted retinas stained with cresyl violet. Gene expression profiles of control and ONT retinas were analyzed with Agilent's rat oligo microarray slides. Differentially expressed genes were identified from three independent experiments and clustered based on their functions with expression analysis systematic explorer software. Real-time quantitative and semiquantitative RT-PCR were used to validate changes in gene expression levels.

Results: Gene expression changes in axotomized retinas were analyzed with rat oligo microarray slides containing 22,575 oligonucleotide probes that represent over 20,000 genes and expressed sequence tags (ESTs). The expression of 493 genes was increased more than 1.5 fold, including 85 that were upregulated more than 2 fold, and the expression of 113 genes was decreased 1.5 fold or more, including 21 that were downregulated more than 2 fold. Differentially expressed genes were clustered based on their functions. Several novel genes expressed in the GCL have been identified and their expression patterns in different tissues were analyzed. Among the genes differentially regulated in retinas with induced retinal ganglion cell (RGC) death, we have identified 13 genes that are mapped to known glaucoma loci and can be considered for mutation screening in patients with inherited forms of glaucoma.

Conclusions: The gene expression profiles of the ONT retinas can be used to identify RGC-expressed genes, which may be essential for the morphological and physiological characteristics of RGCs. The results of this study can also be used to evaluate the molecular changes taking place in the retina in response to RGC loss.

Introduction

An axotomy animal model is characterized by rapid and specific retinal ganglion cell (RGC) degeneration [1,2]. It has been shown that after intraorbital optic nerve transection, practically all RGCs survive during the first five days and then degenerate rapidly, reducing the population of RGCs by approximately 50% by day 7, and by more than 90% by two weeks, compared to the control. The axotomy model was used in this study to obtain gene expression profiles of the retina that sustained a significant loss of RGCs (GSE6303). One of the primary goals of this study was to identify RGC-expressed genes by comparing the gene profiles of RGC-depleted and control retinas. RGC-expressed genes were expected to be underrepresented in RGC-deficient retinas compared to the control. These genes required to be selected from the pool of differentially expressed genes that were modulated in response to the surgical procedure, or associated with the RGC degenerative process. Although RGCs have been studied extensively for over a hundred years, starting with the monumental work on ganglion cell classification by Cajal in 1892, current knowledge of RGC molecular biology is limited. Different types of RGCs have been identified based on their morphological characteristics, such as soma size, dendritic field size, and dendritic ramification. Most fundamental knowledge has been obtained from cat RGC studies [3-7], which established a correlation between the three main morphological classes of RGCs: the α (3% of all RGCs), β (45-50% of all RGCs), and non-α/non-β cells (NAB, 50-60% of all RGCs); and three physiological classes: Y, X, and W, classified based on characteristics such as receptive field properties, axonal conduction velocities, and the level of maintained activity [8-12]. The primate retina contains at least 18 different types of RGCs that were classified morphologically into Pα (parasol), Pβ (midget), and Pγ types [13,14], and physiologically into two major types: parasol or magnocellular (M), and midget or parvocellular (P) cells [15-18]. Midget cells are color-sensitive and distributed most densely in the foveal region [19]. They project exclusively to the parvocellular layers of the lateral geniculate nucleus (LGN) and play a key role in central acuity. Parasol ganglion cells are motion-sensitive and primarily project to the magnocellular layers of the LGN [16,20]. Other groups of primate RGCs include the following: (1) blue-yellow bistratified RGCs; (2) RGCs responsible for pupillary reaction; and (3) photosensitive melanopsin-containing RGCs involved in regulation of circadian rhythm. Undoubtedly, this variation in RGC morphology and function is supported at the molecular level, particularly at the level of gene expression.

The information obtained from this and similar studies can also be important for better understanding the pathology of various optic neuropathies associated with RGC degeneration, including glaucoma, which affects more than 70 million people worldwide. It is logical to assume that changes in the physiological levels of RGC-expressed genes in response to external or intracellular stimuli, or the presence of mutations affecting the structure or function of the encoded proteins could undermine RGC viability, and consequently, may be associated with glaucoma or other optic neuropathies. Moreover, the analysis of differentially expressed genes in the optic nerve transection (ONT) model could be relevant for understanding how different types of retinal cells respond to the RGC loss that takes place in optic neuropathies.

Methods

Animals

The use of animals for this study was approved by the Animal Research Committee of the University of California, Los Angeles, and was performed in compliance with the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research. Male Wistar rats weighing 250-300 g were housed with standard chow and water provided ad libitum. The animal room was lit with fluorescent lights (330 lux) automatically turned on at 3 AM and off at 3 PM, and the temperature was kept at 21 °C. The animals were maintained for at least one week prior to surgical procedures.

To generate the ONT model, the animals were anesthetized by inhalation of an isofluroane (1.5-3.5%) in oxygen (approximately 1 l per minute). The optic nerve was exposed through a lateral conjunctival incision, and the optic nerve sheath was incised with a needle knife 2 mm longitudinally, starting 3 mm behind the eye. A cross-section of the optic nerve was made with the needle knife through the opening of the optic nerve sheath, with care not to damage the adjacent blood supply. The conjunctival incision was sutured and Tobrex ophthalmic ointment (tobramycin; Alcon, Fort Worth, TX) was applied topically. The ONT procedure was performed on one eye of each rat, while the contralateral eye remained untreated. Animals were euthanized two weeks after the procedure.

Enucleated eyeballs were bisected and retinas were dissected from the eyeball of five rats. The temporal halves were used for cell counting in the ganglion cell layer (GCL), while nasal retinas were pooled for RNA isolation.

Cell counting

Cell loss in the experimental retinas was evaluated by counting cells in the GCL in cresyl-violet-stained retinas [21]. Cells with cytoplasm rich in Nissl substance and with irregular outlines were counted as neurons [22]. The number of neurons in the GCL was determined by counting cells with an eyepiece reticule of a microscope at 400x magnification in three adjacent fields at 1 mm, 2 mm, and 3 mm from the center of the optic nerve along the centerline of each retinal quadrant (superior, temporal, inferior, and nasal). The number of cells counted in 36 separate areas (0.23x0.23 mm) of each retina was averaged to arrive at one value (mean±standard deviation). Cell counting was performed by two investigators in a masked fashion.

RNA isolation and microarray experiments

Retinas of 18 rats were pooled with six retinas per tube (treated and contralateral control retinas were pooled in separate tubes). Total RNA was extracted with RNAzol B (Tel-Test, Friendswood, TX), and further purified with RNeasy MinElute Cleanup kit (Qiagen, Valencia, CA). RNA was quantified by spectrophotometry at 260 nm, and its integrity was analyzed by denaturing agarose gel electrophoresis (1% agarose, 2.2 M formaldehyde). Gene expression changes in axotomized retinas were analyzed with microarray technology using Rat Oligo Microarray slides (G4130A, Agilent Technologies, Palo Alto, CA) containing 22,575 oligonucleotide probes that represent over 20,000 genes, ESTs, and EST clusters. A complete probe list can be found at Agilent. Microarray slides were hybridized with fluorescent-labeled cRNA targets corresponding to the control and experimental retinal RNA.cRNA labeling was carried out with the Low Input RNA Fluorescent Linear Amplification kit (Agilent Technologies) that generates cyanine 3- or cyanine 5-labeled cRNA. The procedure consists of converting mRNA primed with an oligo (d)T-T7 primer into dsDNA, and amplifying the sample using T7 RNA polymerase in the presence of cyanine 3-CTP or cyanine 5-CTP. Briefly, 0.5 μg of the total RNA from treated retinas and 0.5 μg of the total RNA from control retinas were converted into dsDNA. The corresponding cDNA was independently labeled by incorporation of either cyanine 3-CTP or cyanine 5-CTP, respectively. cRNA products were purified and their concentrations were determined spectrophotometrically using NanoDrop ND-1000 Spectrophotometer (NanoDrop Technologies, Wilmington, DE). Dye-incorporation was measured using MicroArray Measurement Module of NanoDrop ND-1000. The control and experimental cRNA was combined and hybridized to the same microarray. For hybridization, the Agilent oligo microarray processing protocol was strictly followed. Briefly, 0.75 μg of labeled control cRNA and 0.75 μg of labeled treated cRNA were mixed and incubated with a microarray slide at 60 °C for 17 h using an Agilent in situ hybridization kit. Following hybridization, the slides were washed with Wash Solution 1 (6x SSC, 0.005% Triton X-102) at room temperature, and then with cold (4 °C) Wash Solution 2 (0.1xSSC, 0.005% Triton X-102). Slides were then immediately dried using an ultra pure filtered N₂ stream. After drying, the slides were scanned with Agilent's dual-laser Microarray Scanner, which has dynamic autofocusing for high resolution feature scanning and high-throughput analysis of multiple slides. Gene expression data were obtained with Agilent Feature Extraction software with default settings that select the locally weighted linear regression curve fit (lowess) normalization method after local background subtraction. It provides normalized signals, log ratios of gene expression ratio values, which are equal to the ratio of Cy5 processed signal to Cy3 processed signal, p-values, and pixel statistics. A detailed description of algorithms that were utilized can be found in "Robust local normalization of gene expression microarray data" at Scripts. After feature extraction the log ratios were converted to fold change. Output from the microarray analysis was merged with the Unigene or GenBank descriptor. Gene profilings were performed in triplicate i.e., three microarray slides were hybridized separately with cRNAs generated from six distinct experimental and six contralateral control retinas. Differentially expressed genes were identified by analyzing the data from three independent experiments using Student's t-test (p<0.05), and clustered based on their functions using the Expression Analysis Systematic Explorer software (EASE).

Semiquantitative polymerase chain reaction (sqPCR) and real-time quantitative PCR (qPCR). Total RNA (5 μg) was reverse transcribed to cDNA with SuperScript First-strand Synthesis System for RT-PCR (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. cDNA was amplified by PCR with primers specific to the target sequence. Amplification conditions were as follows: hot start of 2 min at 95 °C; 30 cycles of denaturing (95 °C for 30 s), annealing (60 °C for 15 s), and extension (72 °C for 30 s); and a final extension of 7 min at 72 °C. Dilutions of cDNA in the PCR were adjusted for each gene with the aim of staying within the linear range of amplification. The PCR products were separated by electrophoresis in a 2% agarose gel and visualized under ultraviolet light in the presence of ethidium bromide. Real-time PCR was performed with DyNAmo SYBR Green qPCR kit and Chromo4 detection system (MJ Research, Waltham, MA) according to the manufacturers' instructions. Cycling conditions were as follows: denaturation (95 °C for 15 min), amplification, and quantification (95 °C for 10 s, 60 °C for 10 s, and 72 °C for 5 s, with a single fluorescence measurement at the end of 72 °C for 5 s segment), repeated 40 times, a melting curve program (65 °C for 15 s and 95 °C with a heating rate of 0.1 °C/s and continuous fluorescence measurement), and a cooling step to 40 °C. β-actin mRNA was used as a standard to ensure that equivalent amounts of RNA were included in each assay. Primers used in qPCR and their sequences are shown in Table 1.

In situ hybridization

cDNA fragments corresponding to the genes of interest were obtained by RT-PCR with sequence-specific primers and first strand retinal cDNA. DNA fragments were purified and subcloned into pCRII-Topo vector (Invitrogen). Digoxigenin (DIG)-and fluorescein-labeled antisense and sense cRNA probes were synthesized by in vitro transcription with either T7 or Sp6 RNA polymerase according to the manufacturer's protocol (Roche Applied Science, Indianapolis, IN). Sense riboprobes were used in these experiments as the negative control. In situ hybridization was performed according to the published protocol [23] with minor modifications. Briefly, 10 μm-thick frozen sections, fixed in 4% paraformaldehyde/phosphate-buffered saline (PBS 1X PBS (137 mM NaCl, 2.7 mM KCl, 8 mM Na₂HPO₄, 2 mM KH₂PO₄, pH 7.4)), were washed with PBS for 30 min, and equilibrated for 15 min in 5X SSC (0.75 M NaCl, 0.075 M Na-Citrate). Prehybridization was carried out for 2 h in a solution containing 50% formamide, 5X SSC, and 40 μg/ml salmon sperm DNA. Sections were then hybridized for 12-24 h in a humid chamber at 58 °C in a prehybridization solution, with the addition of the DIG-labeled RNA probe at a concentration of 400 ng/ml. Next, the section were washed twice with 2X SSC, and twice with 0.1X SSC at 65 °C for 1 h, before being incubated with with alkaline phosphatase (AP) conjugated anti-fluorescein and anti-DIG antibodies. First, the sections were incubated with anti-fluorescein-AP antibody at 4 °C for 12-14 h. Color staining was obtained with HNPP/Fast Red TR (Roche). After deactivation of the anti-fluorescein-AP antibody's phosphatase activity at 65 °C for 20 min, sections were incubated with the anti Dig-AP antibody at 4 °C for 12-14 h. Color staining was developed by incubation sections with NBT/BCIP (Roche). Sections were stained with 4',6-diamino-2-phenylindole (DAPI) solution diluted 1:3000 (Sigma Aldrich, St. Louis, MO), mounted and viewed with a fluorescence microscope. Color staining was obtained by incubating sections with NBT/BCIP (Roche).

Statistical analysis

Quantitative results are expressed as the mean±standard deviation (SD). Significance was determined with Student's t-test, and p values <0.05 were considered statistically significant.

Results

Optic nerve transection animal model

The experimental strategy for this project was to produce rat models with severe degeneration of RGCs and to compare the gene expression profiles of retinas from these animals with control retinas. Axotomized animals were euthanized two weeks after the procedure, and the number of surviving cells in the GCL was estimated to evaluate the extent of RGC degeneration (Figure 1). An approximate 40% decrease in the number of cells in the GCL was observed after ONT. It has been shown that axotomy causes approximately 90% of RGC death by two weeks [1,2]. Kielczewski and coworkers [24] found no RGC bodies in the GCL two weeks after nerve transection. The relatively higher number of cells in the GCL that we observed in our experiments is due to the method used for cell counting, which stained both RGCs and displaced amacrine cells that were present in approximately equal proportion [24,25]. Although a significant loss of GABA and glycine labeling (markers for amacrine cells) was observed after ONT, it was interpreted to be not as a loss of amacrine cells, but as a change in expression of their neurotransmitters [24]. Based on these observations, the estimated number of surviving RGCs in the axotomy model used in this study would be approximately 5%.

Gene expression analysis of retinal ganglion cell-deficient retinas

Microarray analysis was carried out to obtain gene expression profiles from axotomized and control animal retinas. In ONT retinas, the expression of 493 genes was stimulated more than 1.5 fold, including 85 that were upregulated more than 2 fold. The expression of 113 genes was decreased 1.5 fold or more, including 21 that were downregulated more than 2 fold. Approximately one third of the differentially expressed genes were previously characterized. These genes were clustered based on their function and are presented in Table 2. Several genes involved in apoptotic cell death, including clusterin, lysozyme, Bax, Bad, apoptosis-inducing factor (AIF), Fas apoptotic inhibitory molecule (Faim), apolipoprotein E, and death-associated like kinase (Dapkl) were upregulated after axotomy, suggesting they play a role in the RGC degenerative process. Among those underrepresented in RGC-deficient retinas there were genes that are known to be expressed in the retina by RGCs, including genes encoding neurofilaments (light, medium, and heavy subunits), neurofilament-associated proteins (peripherin and internexin), neuritin, and adenylate cyclase activating protein. mRNA levels of the two other known RGC-specific genes, Thy-1 and Brn3b, which are not listed in Table 2, were analyzed in control and ONT retinas (Figure 2). According to microarray data the Thy-1 gene was downregulated in ONT retinas almost 1.5 fold. An approximate 2 fold decrease in the Thy-1 mRNA level was observed by sqPCR. The Brn3b gene was not represented in the microarray slides used in these studies. s-qRT-PCR showed weak expression of Brn3b in control retinas, whereas the PCR product corresponding to the mRNA for this gene was undetectable in ONT retinas.

Validation of microarray data

To verify the reliability of the microarray results, we arbitrarily chose several known and uncharacterized genes and analyzed their expression level with sqPCR and qPCR. Ten uncharacterized genes downregulated in axotomized retinas were evaluated by s-qPCR (Figure 3). The integrated density values (IDV) of RT-PCR products obtained with RNA isolated from axotomized and contralateral control retinas clearly indicated a decrease in expression level of the analyzed genes after ONT. Only one out of ten genes did not show significant change in expression level. Microarray data were also evaluated by qPCR, a more accurate technique to measure gene expression levels than sqPCR. The results of this experiment for six genes (Figure 4), including AI171466, AW529036, BQ781402, NM_053346 (neuritin, Nrn), NM_017009 (glial fibrillary acidic protein, Gfap), and NM_053955 (μ-crystallin, Crym), confirmed the changes in expression levels observed by the microarrays. AI171466, 290516_Rn, and Nrn were downregulated in the treated retinas although the treated/control ratio for the neuritin obtained by qPCR is about two times lower than that observed in microarray experiments. BQ781402, Gfap, and Crym showed higher levels of expression in response to ONT observed by microarrays as well as by qPCR.

Tissue distribution pattern and in situ hybridization of selected genes

Several previously uncharacterized genes with a reduced expression level in axotomized retinas were further analyzed with the aim of identifying their tissue distribution pattern and in situ localization in the retina. Relative expression level of these genes in the retina, cerebellum, cortex, heart, kidney, liver, and muscle are presented in Figure 5. The results were obtained with sqRT-PCR using gene sequence-specific primers (Table 2) and the first strand cDNA synthesized from the corresponding tissue. Nrn, Crym, and farnesyl diphosphate farnesyl transferase 1 (Fdft1) were used as controls since their tissue expression patterns are well known. All genes tested, except Crym, were downregulated in RGC-deficient retinas according to microarray data. The ethidium bromide staining intensity of the sqPCR products (Figure 5) in control and ONT retina lanes correlate with the microarray data. Genes AW529036, BF562863, AW522386, and BF404316 showed expression predominantly in the retina and brain (cerebellum and cortex). BM389543 showed strong expression in the retina, less in the kidney, and much fainter bands were present in all other tissues tested in this experiment.

In situ hybridization was performed with the aim of determining the retinal cell types expressing the genes of interest that were underrepresented after axotomy. The images presented in Figure 6 were obtained by the hybridization of control retinal sections with the DIG-labeled sense and anti-sense riboprobes corresponding to AI171466, AW529036, and BF404316. Cell staining with all three antisense riboprobes was mainly observed in large cell bodies in the GCL. No obvious staining was detected with the sense probes. Since the rodent retina contains approximately equal numbers of RGCs and non-RGCs, which are mostly displaced amacrine cells, double in situ hybridization with RNA probes recognizing the genes of interest (DIG-labeled) and Thy-1 (RGC marker in the retina, fluorescein-labeled) was performed to identity cells expressing these genes (Figure 7). The hybridization signals for all three genes and Thy-1 were colocalized to the same cells in the GCL, indicating that these genes expressed in the retina were expressed specifically by RGCs.

Sequence analysis of novel genes expressed in the GCL

Computational sequence analysis was performed in order to obtain information about the novel genes that were expressed in the GCL. The identification of the homologous genes with known functions, conserved domains, or motifs could provide us with essential ideas about their function. The results of the sequence analysis for AI171466, AW529036 and BF404316 are presented below.

AI171466 corresponds to the 1036 nt rat cDNA as predicted by automated computational analysis (NM_001009639). This record was derived from an annotated genomic sequence (NW_047535) using the gene prediction method, GNOMON. The rat sequence is 95% identical to mouse NM_026481, and about 90% identical to the human BC037798 clone, called the brain specific protein. According to UniGene's expression profiles, both mouse and human clones are expressed in a wide variety of tissues, which correlates with our data (Figure 4). The rat gene is mapped to chromosome 19q11. The protein encoded by this gene contains a p25-α conserved domain, which is characteristic to the family encoding a 25 kDa protein phosphorylated by a Ser/Thr-Pro kinase [26]. A brain-specific 25 kDa protein (p25), a substrate of tau protein kinase II, was found to be expressed in glial cells (oligodendrocytes), especially in the white matter, such as the corpus callosum, cingulum, external capsule, and internal capsule, of the adult rat brain. Expression of p25 was also observed in the neuropil in the first and second layers of the cerebral cortex and dentate and in fiber-like structures in the hippocampus [27].

We found AW529036 showed 93% similarity to mouse cDNA NM_177572, encoding a 380-amino acid hypothetical protein that contains an ATP binding domain of the carbamoyl-phosphate synthase (CPS) L chain. CPS initiates both the urea cycle and the biosynthesis of arginine as well as pyrimidines by catalyzing the ATP-dependent synthesis of carbamyl-phosphate from glutamine or ammonia and bicarbonate. The residues 87-298 aa of the predicted protein also share homology with the ATP binding domain of glutathione synthase. Alterations in glutathione level have been shown to change the neurotrophic effects of nitric oxide (NO) into neurotoxic effects and have triggered a programmed cell death of neurons in midbrain cultures [28]. NM_177572 was shown to be expressed in the brain, eye, pituitary gland, and kidney, and mapped to chromosome 4D2.1. We also detected strong expression in the retina, cerebellum, and cortex, as well as a low level of expression in the kidney.

We observed BF404316 to be 88% homologous to the mouse tripartite motif (TRIM) containing 36 (Trim36, NM_178872, located on chromosome 18), which was expressed in the brain, eye, kidney, pituitary gland, and in the testes. The TRIM proteins are also known as the 'RING-B box-coiled coil' (RBCC) subgroup of RING finger proteins. The TRIM motif includes three zinc-binding domains, a RING, a B-box type 1 and a B-box type 2, and a coiled-coil region. Although the exact function of TRIM36, is still unknown, it has been reported to be involved in acrosome reactions and fertilization [29]. Other members of the TRIM family have been shown to control cell proliferation (EFP and AFP), organ development (MID1 and TRIM37) and transcription (RFP, TIF1, and PML) [30,31]. TRIM36 does not share a high degree of homology with other known proteins. The best match was found between TRIM36 and midline 1 protein (MID1). These proteins also have a similar domain organization. MID1 is involved in the formation of multimeric protein structures acting as anchor points to microtubules during the cell cycle [32]. Mutations in MID1 have been associated with the X-linked form of Opitz syndrome, which is characterized by midline abnormalities such as cleft lip, laryngeal cleft, heart defects, hypospadias, and agenesis of the corpus callosum [33,34].

Discussion

The experimental strategy of this study was to compare the gene expression profiles between control retinas and retinas of the axotomy animal model that sustained significant degeneration of RGCs. The axotomy (or ONT) model is characterized by fast and specific degeneration of RGCs [1,2,24]. Initial damage to the optic nerve, which consists of RGC axons, leads to the death of RGC somatas. Depending on the severity of the damage, the rate of this process can vary. At first, only cells that are damaged undergo degeneration. Later, cells in close proximity to dying or dead cells can become involved in the degenerative process.

The differences in gene expression profiles between control and treated retinas were determined by the hybridization of labeled RNA targets corresponding to these tissues with microarray slides representing more than 22,000 genes and ESTs. Genes with average values of up- or downregulation of 1.5 fold or more were presented as differentially expressed. The genes that were underrepresented in RGC-deficient retinas were further analyzed with the aim of identifying novel RGC-expressed genes. Several characterized genes with known expression in RGCs were among those selected by microarrays, including Nrn [35,36] and neurofilaments [37]. Neuritin is expressed in postmitotic-differentiating neurons of the developing nervous system and neuronal structures associated with plasticity in the adult. It is induced by neuronal activity and by the activity-regulated neurotrophins, BDNF and NT-3 [35]. Neurofilaments are the major structural components of the neuron and are essential for the establishment and maintenance of axonal diameter, a property that determines the velocity of electrical signal conduction. Mature neurons contain three subunits: light, middle, and heavy chains (NF-L, NF-M, and NF-H). Neurofilaments may also contain the proteins α-internexin, peripherin, nestin, and vimentin. Neurofilament accumulations are seen in many neurological disorders, including Alzheimer's disease [38], amyotrophic lateral sclerosis (Lou Gehrig's disease) [39], giant axon neuropathies [40], and others. Two other RGC-specific genes, Thy-1 and Brn3b, were also downregulated in the retinas of ONT-treated animals. The downregulation of several RGC-expressed genes in ONT retinas, as well as the results of the in situ hybridization demonstrating the expression of three novel genes in the RGCs, indicate our strategy of identifying RGC-expressed genes was successful.

The expression profiles of the retinas following ONT also present an opportunity to identify genes that may be associated with RGC death during optic neuropathies such as glaucoma. The molecular mechanisms of the pathophysiology of many forms of glaucoma are unknown. There is evidence that more than 20% of glaucoma cases have a genetic basis [41]. Several loci have been identified [42,43] and two proteins, myocilin [44,45] and optineurin [46], have been associated with primary open angle glaucoma (POAG); one gene, cytochrome P450 [47], with primary congenital glaucoma (PCG); and three transcription factors, PITX2, FOXC1, and LMX1B [48-50], with forms of developmental glaucoma. Although the animal model used here cannot be considered ideal for glaucomatous neuropathy, the final common outcome is the death of RGCs. Several differentially expressed genes in the ONT model were mapped to the known loci associated with glaucomatous neuropathy and therefore, can be analyzed for mutations in the DNA of patients with familial glaucoma. In addition to myocilin (NM_030865), which is known to be associated with POAG, quiescin Q6 (NM_053431) was localized to human chromosome 1q24, the GLC1A locus for POAG; Fas apoptotic inhibitory molecule (NM_080895), ceruloplasmin (NM_012532), and transferrin (NM_001013110) were mapped to the 3q22.1, a GLC1C locus 3q21-24 for POAG. Secreted acidic cysteine rich glycoprotein (NM_012656) was localized to 5q31, a novel locus for JOAG at 5q22.1-q32 [43]. Endothelin converting enzyme 1 (NM_053596), guanine nucleotide-binding protein β-1 (NM_030987), 3-hydroxy-3-methylglutaryl CoA lyase (NM_024386), and enolase 1 α (NM_012554) were localized to human chromosome 1p36, the GLC3B locus for the PCG [51]. Vasopressive intestinal peptide receptor 2 was mapped to 7q36.3, the GLC1F (7q35-q38) and GPDS1 (7q35-36) loci for POAG and pigmentary glaucoma, respectively. Acetyl-coenzyme A acyltransferase 2 (NM_130433), retina and anterior neural fold homeobox (NM_053678), transthyretin (NM_012681), and parvalbumin (NM_022499) were mapped to GPDS2 (18q11-21) locus for pigmentary glaucoma. We believe that localization of the genes differentially regulated in the ONT model characterized by RGC degeneration to glaucoma loci justifies screening DNA of patients with inherited glaucoma for mutations in these genes.

Finally, we have compared the ONT array data with retinal gene expression profiles of two glaucoma models and ischemia-reperfusion model to identify genes that could be involved in the cell response mechanism to different stress conditions leading to RGC death. No similarities were found between array data obtained from ONT and glaucoma [52] or ischemic [53] models. Six genes were differentially regulated in retinas of ONT and glaucoma model generated in our laboratory (the array data for this glaucoma model are unpublished). These include eukaryotic translation elongation factor 2 (NM_017245), nuclear receptor subfamily 2, group F, member 2 (NM_080778), cyclin L (NM_053662), olfactory neuronal transcription factor 1 (NM_053820), ceruloplasmin (NM_012532), and crystallin β A4 (NM_031689). Only two of these genes, nuclear receptor subfamily 2, group F, member 2 (NR2F2) and ceruloplasmin, were similarly regulated in both models. NR2F (COUP-TFII) is an orphan member of the nuclear receptor superfamily of ligand-activated transcription factors [54,55]. It is known to be an important factor involved in regulation of many aspects of cell differentiation, metabolism and homeostasis. Genetic ablation of COUP-TFII results in early embryonic lethality [55].

Ceruloplasmin, the second gene that was upregulated in both ONT and glaucoma models, encodes a glycoprotein involved in copper and iron metabolism, found mainly in plasma where it binds 95% of serum copper [56]. In the retina, it is mainly expressed in the astrocytes in the GCL and in the Müller cells [57,58]. An elevated expression level of the ceruloplasmin was reported in human and monkey glaucomatous retinas following optic nerve crush, photic injury, and light damage, and in the retinas of age-related macular degeneration patients [58-63]. It has been suggested that increased ceruloplasmin expression may help protect the retinal cells from oxidative stress by reducing the amount of reactive oxygen species [61].

In summary, we obtained and analyzed the gene expression profiles of the axotomized retinas. Several novel genes expressed in the GCL were identified and their expression patterns in different tissues were analyzed. Among the differentially regulated genes in retinas with induced RGC death, we have identified 14 genes that were mapped to known glaucoma loci and can be considered for mutation screening in patients with inherited forms of glaucoma.

Acknowledgements

Supported by The Gerald Oppenheimer Family Foundation.

References

1. 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; 91:1632-6.

2. 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; 602:304-17.

3. Ammermuller J, Kolb H. The organization of the turtle inner retina. I. ON- and OFF-center pathways. J Comp Neurol 1995; 358:1-34.

4. Ammermuller J, Muller JF, Kolb H. The organization of the turtle inner retina. II. Analysis of color-coded and directionally selective cells. J Comp Neurol 1995; 358:35-62.

5. Thanos S, Vanselow J, Mey J. Ganglion cells in the juvenile chick retina and their ability to regenerate axons in vitro. Exp Eye Res 1992; 54:377-91.

6. Boycott BB, Dowling JE. Organization of the primate retina: light microscopy. Philos Trans R Soc Lond B Biol Sci 1969; 255:109-84.

7. Boycott BB, Wassle H. The morphological types of ganglion cells of the domestic cat's retina. J Physiol 1974; 240:397-419.

8. Enroth-Cugell C, Robson JG. The contrast sensitivity of retinal ganglion cells of the cat. J Physiol 1966; 187:517-52.

9. Cleland BG, Dubin MW, Levick WR. Sustained and transient neurones in the cat's retina and lateral geniculate nucleus. J Physiol 1971; 217:473-96.

10. Fukada Y. Receptive field organization of cat optic nerve fibers with special reference to conduction velocity. Vision Res 1971; 11:209-26.

11. Stone J, Hoffmann KP. Very slow-conducting ganglion cells in the cat's retina: a major, new functional type? Brain Res 1972; 43:610-6.

12. Stone J, Fukuda Y. Properties of cat retinal ganglion cells: a comparison of W-cells with X- and Y-cells. J Neurophysiol 1974; 37:722-48.

13. Perry VH, Oehler R, Cowey A. Retinal ganglion cells that project to the dorsal lateral geniculate nucleus in the macaque monkey. Neuroscience 1984; 12:1101-23.

14. Perry VH, Cowey A. Retinal ganglion cells that project to the superior colliculus and pretectum in the macaque monkey. Neuroscience 1984; 12:1125-37.

15. Dacey DM, Petersen MR. Dendritic field size and morphology of midget and parasol ganglion cells of the human retina. Proc Natl Acad Sci U S A 1992; 89:9666-70.

16. Rodieck RW, Watanabe M. Survey of the morphology of macaque retinal ganglion cells that project to the pretectum, superior colliculus, and parvicellular laminae of the lateral geniculate nucleus. J Comp Neurol 1993; 338:289-303.

17. Dacey DM. Primate retina: cell types, circuits and color opponency. Prog Retin Eye Res 1999; 18:737-63. Erratum in: Prog Retin Eye Res 2000; 19:following 646.

18. Yamada ES, Silveira LC, Perry VH, Franco EC. M and P retinal ganglion cells of the owl monkey: morphology, size and photoreceptor convergence. Vision Res 2001; 41:119-31. Erratum in: Vision Res 2001; 41:2027.

19. Dacey DM. The mosaic of midget ganglion cells in the human retina. J Neurosci 1993; 13:5334-55.

20. Schiller PH, Malpeli JG. Functional specificity of lateral geniculate nucleus laminae of the rhesus monkey. J Neurophysiol 1978; 41:788-97.

21. Ishii Y, Kwong JM, Caprioli J. Retinal ganglion cell protection with geranylgeranylacetone, a heat shock protein inducer, in a rat glaucoma model. Invest Ophthalmol Vis Sci 2003; 44:1982-92.

22. Gellrich MM, Gellrich NC. Quantitative relations in the retinal ganglion cell layer of the rat: neurons, glia and capillaries before and after optic nerve section. Graefes Arch Clin Exp Ophthalmol 1996; 234:315-23.

23. Braissant O, Wahli W. A simplified in situ hybridization protocol using non-radioactively labeled probes to detect abundant and rare mRNAs on tissue sections. Biochemica (Indianapolis, IN) 1998; 1:10-6.

24. Kielczewski JL, Pease ME, Quigley HA. The effect of experimental glaucoma and optic nerve transection on amacrine cells in the rat retina. Invest Ophthalmol Vis Sci 2005; 46:3188-96.

25. Jeon CJ, Strettoi E, Masland RH. The major cell populations of the mouse retina. J Neurosci 1998; 18:8936-46.

26. Takahashi M, Tomizawa K, Ishiguro K, Sato K, Omori A, Sato S, Shiratsuchi A, Uchida T, Imahori K. A novel brain-specific 25 kDa protein (p25) is phosphorylated by a Ser/Thr-Pro kinase (TPK II) from tau protein kinase fractions. FEBS Lett 1991; 289:37-43.

27. Takahashi M, Tomizawa K, Fujita SC, Sato K, Uchida T, Imahori K. A brain-specific protein p25 is localized and associated with oligodendrocytes, neuropil, and fiber-like structures of the CA3 hippocampal region in the rat brain. J Neurochem 1993; 60:228-35.

28. Canals S, Casarejos MJ, de Bernardo S, Rodriguez-Martin E, Mena MA. Glutathione depletion switches nitric oxide neurotrophic effects to cell death in midbrain cultures: implications for Parkinson's disease. J Neurochem 2001; 79:1183-95.

29. Kitamura K, Tanaka H, Nishimune Y. Haprin, a novel haploid germ cell-specific RING finger protein involved in the acrosome reaction. J Biol Chem 2003; 278:44417-23.

30. Balint I, Muller A, Nagy A, Kovacs G. Cloning and characterisation of the RBCC728/TRIM36 zinc-binding protein from the tumor suppressor gene region at chromosome 5q22.3. Gene 2004; 332:45-50.

31. Zhong S, Salomoni P, Pandolfi PP. The transcriptional role of PML and the nuclear body. Nat Cell Biol 2000; 2:E85-90.

32. Cainarca S, Messali S, Ballabio A, Meroni G. Functional characterization of the Opitz syndrome gene product (midin): evidence for homodimerization and association with microtubules throughout the cell cycle. Hum Mol Genet 1999; 8:1387-96.

33. Celton-Morizur S, Bordes N, Fraisier V, Tran PT, Paoletti A. C-terminal anchoring of mid1p to membranes stabilizes cytokinetic ring position in early mitosis in fission yeast. Mol Cell Biol 2004; 24:10621-35.

34. Mnayer L, Khuri S, Merheby HA, Meroni G, Elsas LJ. A structure-function study of MID1 mutations associated with a mild Opitz phenotype. Mol Genet Metab 2006; 87:198-203.

35. Naeve GS, Ramakrishnan M, Kramer R, Hevroni D, Citri Y, Theill LE. Neuritin: a gene induced by neural activity and neurotrophins that promotes neuritogenesis. Proc Natl Acad Sci U S A 1997; 94:2648-53.

36. Krishnamoorthy RR, Agarwal P, Prasanna G, Vopat K, Lambert W, Sheedlo HJ, Pang IH, Shade D, Wordinger RJ, Yorio T, Clark AF, Agarwal N. Characterization of a transformed rat retinal ganglion cell line. Brain Res Mol Brain Res 2001; 86:1-12.

37. Al-Chalabi A, Miller CC. Neurofilaments and neurological disease. Bioessays 2003; 25:346-55.

38. Nakamura Y, Hasimoto R, Kashiwagi Y, Miyamae Y, Shinosaki K, Nishikawa T, Hattori H, Kudo T, Takeda M. Abnormal distribution of neurofilament L in neurons with Alzheimer's disease. Neurosci Lett 1997; 225:201-4.

39. Bruijn LI, Miller TM, Cleveland DW. Unraveling the mechanisms involved in motor neuron degeneration in ALS. Annu Rev Neurosci 2004; 27:723-49.

40. Treiber-Held S, Budjarjo-Welim H, Reimann D, Richter J, Kretzschmar HA, Hanefeld F. Giant axonal neuropathy: a generalized disorder of intermediate filaments with longitudinal grooves in the hair. Neuropediatrics 1994; 25:89-93.

41. Wolfs RC, Klaver CC, Ramrattan RS, van Duijn CM, Hofman A, de Jong PT. Genetic risk of primary open-angle glaucoma. Population-based familial aggregation study. Arch Ophthalmol 1998; 116:1640-5.

42. Ray K, Mukhopadhyay A, Acharya M. Recent advances in molecular genetics of glaucoma. Mol Cell Biochem 2003; 253:223-31.

43. Pang CP, Fan BJ, Canlas O, Wang DY, Dubois S, Tam PO, Lam DS, Raymond V, Ritch R. A genome-wide scan maps a novel juvenile-onset primary open angle glaucoma locus to chromosome 5q. Mol Vis 2006; 12:85-92 <http://www.molvis.org/molvis/v12/a9/>.

44. Stone EM, Fingert JH, Alward WL, Nguyen TD, Polansky JR, Sunden SL, Nishimura D, Clark AF, Nystuen A, Nichols BE, Mackey DA, Ritch R, Kalenak JW, Craven ER, Sheffield VC. Identification of a gene that causes primary open angle glaucoma. Science 1997; 275:668-70.

45. Nguyen TD, Chen P, Huang WD, Chen H, Johnson D, Polansky JR. Gene structure and properties of TIGR, an olfactomedin-related glycoprotein cloned from glucocorticoid-induced trabecular meshwork cells. J Biol Chem 1998; 273:6341-50.

46. Rezaie T, Child A, Hitchings R, Brice G, Miller L, Coca-Prados M, Heon E, Krupin T, Ritch R, Kreutzer D, Crick RP, Sarfarazi M. Adult-onset primary open-angle glaucoma caused by mutations in optineurin. Science 2002; 295:1077-9.

47. Stoilov I, Akarsu AN, Sarfarazi M. Identification of three different truncating mutations in cytochrome P4501B1 (CYP1B1) as the principal cause of primary congenital glaucoma (Buphthalmos) in families linked to the GLC3A locus on chromosome 2p21. Hum Mol Genet 1997; 6:641-7.

48. Alward WL, Semina EV, Kalenak JW, Heon E, Sheth BP, Stone EM, Murray JC. Autosomal dominant iris hypoplasia is caused by a mutation in the Rieger syndrome (RIEG/PITX2) gene. Am J Ophthalmol 1998; 125:98-100.

49. Nishimura DY, Swiderski RE, Alward WL, Searby CC, Patil SR, Bennet SR, Kanis AB, Gastier JM, Stone EM, Sheffield VC. The forkhead transcription factor gene FKHL7 is responsible for glaucoma phenotypes which map to 6p25. Nat Genet 1998; 19:140-7.

50. Vollrath D, Jaramillo-Babb VL, Clough MV, McIntosh I, Scott KM, Lichter PR, Richards JE. Loss-of-function mutations in the LIM-homeodomain gene, LMX1B, in nail-patella syndrome. Hum Mol Genet 1998; 7:1091-8. Erratum in: Hum Mol Genet 1998; 7:1333.

51. Akarsu AN, Turacli ME, Aktan SG, Barsoum-Homsy M, Chevrette L, Sayli BS, Sarfarazi M. A second locus (GLC3B) for primary congenital glaucoma (Buphthalmos) maps to the 1p36 region. Hum Mol Genet 1996; 5:1199-203.

52. Ahmed F, Brown KM, Stephan DA, Morrison JC, Johnson EC, Tomarev SI. Microarray analysis of changes in mRNA levels in the rat retina after experimental elevation of intraocular pressure. Invest Ophthalmol Vis Sci 2004; 45:1247-58.

53. Yoshimura N, Kikuchi T, Kuroiwa S, Gaun S. Differential temporal and spatial expression of immediate early genes in retinal neurons after ischemia-reperfusion injury. Invest Ophthalmol Vis Sci 2003; 44:2211-20.

54. Chawla A, Repa JJ, Evans RM, Mangelsdorf DJ. Nuclear receptors and lipid physiology: opening the X-files. Science 2001; 294:1866-70.

55. Pereira FA, Qiu Y, Zhou G, Tsai MJ, Tsai SY. The orphan nuclear receptor COUP-TFII is required for angiogenesis and heart development. Genes Dev 1999; 13:1037-49.

56. Harris ZL, Gitlin JD. Genetic and molecular basis for copper toxicity. Am J Clin Nutr 1996; 63:836S-41S.

57. Klomp LW, Farhangrazi ZS, Dugan LL, Gitlin JD. Ceruloplasmin gene expression in the murine central nervous system. J Clin Invest 1996; 98:207-15.

58. Levin LA, Geszvain KM. Expression of ceruloplasmin in the retina: induction after optic nerve crush. Invest Ophthalmol Vis Sci 1998; 39:157-63.

59. Miyahara T, Kikuchi T, Akimoto M, Kurokawa T, Shibuki H, Yoshimura N. Gene microarray analysis of experimental glaucomatous retina from cynomologous monkey. Invest Ophthalmol Vis Sci 2003; 44:4347-56.

60. Farkas RH, Chowers I, Hackam AS, Kageyama M, Nickells RW, Otteson DC, Duh EJ, Wang C, Valenta DF, Gunatilaka TL, Pease ME, Quigley HA, Zack DJ. Increased expression of iron-regulating genes in monkey and human glaucoma. Invest Ophthalmol Vis Sci 2004; 45:1410-7.

61. Chen L, Dentchev T, Wong R, Hahn P, Wen R, Bennett J, Dunaief JL. Increased expression of ceruloplasmin in the retina following photic injury. Mol Vis 2003; 9:151-8 <http://www.molvis.org/molvis/v9/a22/>.

62. Chen L, Wu W, Dentchev T, Zeng Y, Wang J, Tsui I, Tobias JW, Bennett J, Baldwin D, Dunaief JL. Light damage induced changes in mouse retinal gene expression. Exp Eye Res 2004; 79:239-47.

63. Chowers I, Wong R, Dentchev T, Farkas RH, Iacovelli J, Gunatilaka TL, Medeiros NE, Presley JB, Campochiaro PA, Curcio CA, Dunaief JL, Zack DJ. The iron carrier transferrin is upregulated in retinas from patients with age-related macular degeneration. Invest Ophthalmol Vis Sci 2006; 47:2135-40.