Immunofluorescent double-labeling for nitric oxide synthase-2 (red), glial fibrillary acidic protein (green), and the nucleus (blue-purple). Labeling for nitric oxide synthase-2 is not present in control astrocytes (A) but is intensely present in the rough endoplasmic reticulum–Golgi region and cytoplasm of astrocytes grown under elevated hydrostatic pressure for 48 hours (B).
Immunoblot for nitric oxide synthase-2(NOS-2) protein. A, Intense bands of protein for NOS-2 are detected in astrocytes grown under elevated hydrostatic pressure for 24, 48, and 72 hours but not in controls (C). B, Semiquantitation of the optical densitometry values (±SD) for the immunoblot bands of the pressure-exposed cultures compared with controls(n = 3).
Reverse transcription polymerase chain reaction for nitric oxide synthase-2 (NOS-2) messenger RNA. A, Intense bands for NOS-2 messenger RNA are detected in astrocytes grown under elevated pressure for 12, 24, and 48 hours but not in controls (C) (bp indicates base pairs). B, Band intensity for β-actin is similar under all conditions. C, Semiquantitation of the optical densitometry values (± SD) for the NOS-2 messenger RNA bands of the pressure-exposed cultures compared with controls(n = 3).
Liu B, Neufeld AH. Nitric Oxide Synthase-2 in Human Optic Nerve Head Astrocytes Induced by Elevated Pressure In Vitro. Arch Ophthalmol. 2001;119(2):240-245. doi:10-1001/pubs.Ophthalmol.-ISSN-0003-9950-119-2-els00027
To determine whether astrocytes of the human optic nerve head can induce nitric oxide synthase-2 (NOS-2) in response to elevated hydrostatic pressure as a mechanism for directly damaging the axons of the retinal ganglion cells in glaucoma.
Primary cultures of astrocytes from human optic nerve heads were placed in chambers, either pressurized at elevated hydrostatic pressure (60 mm Hg) or maintained at ambient pressure. The induction of NOS-2 was studied by immunocytochemistry, immunoblot, and semiquantitative reverse transcription polymerase chain reaction.
In astrocyte cultures under ambient pressure, NOS-2 was almost undetectable. In astrocyte cultures under elevated hydrostatic pressure for 24, 48, and 72 hours, intensive labeling of NOS-2 in the Golgi body and the cytoplasm was observed by immunocytochemistry and intense bands of NOS-2 were detected by immunoblotting. As detected by semiquantitative reverse transcription polymerase chain reaction, the messenger RNA level of NOS-2 increased significantly in the astrocytes under elevated hydrostatic pressure within 12 hours, peaking earlier than the protein level of NOS-2.
Elevated hydrostatic pressure induces the astrocytes of the human optic nerve head to express NOS-2.
In glaucoma, the appearance of the neurodestructive NOS-2 in astrocytes of the optic nerve head may be a primary response to elevated intraocular pressure, in vivo, and therefore damaging to the axons of the retinal ganglion cells.
GLAUCOMA, a nerve degeneration that causes loss of retinal ganglion cells and a characteristic visual field defect, is in many patients associated with elevated intraocular pressure. In response to elevated intraocular pressure, the optic disc is compressed and the cribriform plates of the lamina cribrosa are stretched and become disorganized.1 The initial site of neuronal degeneration in glaucomatous optic neuropathy is believed to be the axons of the retinal ganglion cells at the level of the lamina cribrosa of the optic nerve head.1,2 Within this region, the connective tissue undergoes extensive remodeling of the extracellular matrix,3 and there are marked changes in astrocytes4 and microglia5 in both morphologic characteristics and distribution. These changes may result primarily in response to elevated intraocular pressure or secondarily after tissue damage. Local cellular responses may alter the microenviroment of the axons of the retinal ganglion cells and contribute to axonal damage as the chronic glaucomatous process proceeds. From our past work, we have demonstrated that one cellular response pathway that contributes to local neurotoxic effects to cause degeneration of the axons of the retinal ganglion cells in the glaucomatous optic nerve head is mediated by nitric oxide (NO).6
Recently, NO has received much attention because of its wide range of biological effects.6- 8 Physiologically, NO serves both intracellularly, as a second messenger that responds to activation of plasma membrane receptors, and extracellularly, as a paracrine factor that carries information between cells.7,8 Pathologically, NO is cytodestructive and, in particular, can cause neuronal degeneration.9- 11 Nitric oxide is synthesized from the guanidino-nitrogen of L-arginine and molecular oxygen by nitric oxide synthase (NOS). Three isoforms of NOS have been cloned and demonstrated in many tissues. Constitutive NOS isoforms (NOS-1 and NOS-3) are activated by biological signals that transiently increase intracellular Ca2+ and are identifiable in a variety of cells in normal tissues. The inducible isoform (NOS-2) is usually not present under normal conditions, is Ca2+ independent, and is induced by cytokines. The expression of NOS-2 results in the sustained and unregulated release of excessive amounts of NO that is cytotoxic to neighboring cells.12,13
Our laboratory has reported that NOS-2 is present in the optic nerve heads of patients with primary open-angle glaucoma and that there is positive staining for nitrotyrosine in the tissue, suggesting that NO through peroxynitrite may contribute to the local damage of the axons of the retinal ganglion cells.6,14 Support for the hypothesis of NO neurotoxic effects in glaucoma is provided by pharmacologic experiments by Neufeld et al15 that demonstrated that an NOS-2 inhibitor can significantly protect against the loss of retinal ganglion cells in an animal model of glaucoma. To understand the regulation of NOS-2 expression in glaucomatous optic neuropathy, we demonstrated that the major cell type that expresses NOS-2 in the glaucomatous optic nerve head is the reactive astrocyte.16
In this study, we asked whether the expression of NOS-2 in reactive astrocytes of the glaucomatous optic nerve head is a direct response to elevated intraocular pressure or a secondary response to tissue destruction. In the work presented herein, we have used cultured human optic nerve head astrocytes to demonstrate that a mechanical stress in vitro, elevated hydrostatic pressure, induces transcription of the NOS-2 gene and synthesis of the NOS-2 protein. Our results suggest that, in glaucoma, elevated intraocular pressure may directly induce astrocytes of the human optic nerve head to express NOS-2, thus playing a primary role in neurotoxic effects by producing excessive NO that can damage the axons of the retinal ganglion cells.
Twelve human eyes from donors (aged 22-65 years) with no history of eye disease were obtained within 24 hours after death from eye banks throughout the United States. The eyes were stored at 4°C and processed within 8 hours of enucleation. Primary lamina cribrosa astrocyte cultures were derived as described by Hernandez et al.17 The posterior pole of the eyes was dissected and the optic nerve head was freed from sclera and other neighboring tissues. The optic nerve head was sliced sagittally and under a dissecting microscope the lamina cribrosa was identified. With the use of a sharp blade, the lamina cribrosa was dissected from prelaminar and postlaminar regions. Each half disk of tissue was then cut into 2 or 3 explants that were placed into culture. From every eye, 4 or 5 explants of the lamina cribrosa were obtained.
The dissected samples were placed in 25-cm2 plastic tissue culture flasks, which had been conditioned with Dulbecco modified Eagle medium supplemented with 10% fetal bovine serum. The explants were incubated in 1.5 mL of medium at 37°C in a 5% oxygen humidified atmosphere; the medium was changed twice a week. Two weeks after initial outgrowth, the cells were treated with 0.05% trypsin-EDTA and split at a ratio of 1:2. The first-passage cells were grown to confluency on glass coverslips for characterization by immunofluorescent staining of glial fibrillary acidic protein (GFAP; 1:400)(Sigma-Aldrich Corp, St Louis, Mo), HLA-DR (1:50; Accurate Chemical & Scientific Corp, New York, NY), von Willebrand factor (1:300; Sigma-Aldrich Corp), and α–smooth muscle actin (1A4; 1:100; Sigma-Aldrich Corp). The primary cell cultures were purified for astrocytes by growing the cells for 1 week in modified, astrocyte-defined, serum-free medium (Dulbecco modified Eagle medium, 0.2% ITS+Premix [combination of insulin, transferrin, selenium, linoleic acid, and bovine serum albumin]; Collaborative Biomedical Products, Bedford, Mass; and 0.1% bovine serum albumin; Sigma-Aldrich Corp) containing forskolin, which suppressed the growth of fibroblasts. The second-passage cell cultures, which had more than 95% cells positive for GFAP, were grown to preconfluency in serum-free medium and used for the following experiments.
At the preconfluent stage, 7 days after growing in serum-free medium, cells, either in culture dishes or on coverslips, were placed in a closed chamber equipped with a manometer that was designed and used previously by Yang et al.18 Pressure was elevated in the chamber to 60 mm Hg with the use of 92% air and 8% carbon dioxide. This pressure was chosen because, at 60 mm Hg, marked alterations in collagen type I synthesis,18 increased synthesis of cyclic adenosine monophosphate,19 and selective expression of a specific isoform of neural cell adhesion molecule synthesis20 have been demonstrated in astrocytes of the human optic nerve head. As controls, the same numbers of cells in culture dishes or on coverslips were placed in a similar chamber at ambient atmospheric pressure in 95% air and 5% carbon dioxide. The variation in carbon dioxide was necessary to maintain the pH in both chambers at 7.4. Calculations with the Henry law showed that the amounts of oxygen dissolved in the media and thus available to the cells did not differ significantly under the pressurized and control conditions.18 Both pressurized and control chambers were placed in a tissue culture incubator at 37°C and maintained for 12, 24, 48, and 72 hours. The pH of the culture media of cells in both chambers was monitored daily and confirmed to remain constant at 7.4. The numbers of cells were monitored by counting in an electronic cell counter (Coulter Electronic Ltd, Luton, England).
Preconfluent cells grown on glass coverslips were fixed in 4% paraformaldehyde at 4°C for 30 minutes, washed in phosphate-buffered saline (PBS), and treated with 0.5% fetal bovine serum–0.2% Triton X-100–0.5% glycine in PBS for 20 minutes. The coverslips were incubated with 50 µL of monoclonal primary antibody against GFAP (Sigma-Aldrich Corp; working dilution, 1:50) for 30 minutes. After washing several times with PBS, the coverslips were incubated with goat anti–mouse Oregon green-X conjugated secondary antibody(Molecular Probes Inc, Eugene, Ore; working dilution, 1:500). The coverslips were then washed with PBS, incubated with a second, polyclonal primary antibody against NOS-2 (Santa Cruz Biotechnology Inc, Santa Cruz, Calif; working dilution, 1:50), washed as above, and incubated with goat anti–rabbit rhodamine red-X conjugated secondary antibody (Molecular Probes Inc; working dilution, 1:1000). After washing several times with PBS, the coverslips were mounted in Vectashield with 4′, 6-diamidino-2-phenylindole (Vector Laboratories Inc, Burlingame, Calif) and imaged by fluorescence microscopy (Olympus AX70; Tokyo, Japan).
Immunoblot for NOS-2 was performed as described previously.16 Briefly, astrocyte monolayers were washed with PBS and lysed in 8-mol/L urea solution containing protease inhibitor (cocktail tablet from Boehringer Mannheim, Penzberg, Germany). Lysates were homogenized and protein concentration was determined with the Bradford colorimetric assay. Thirty micrograms of protein lysates for determination of NOS-2 was loaded in each lane in a sample buffer (2% sodium dodecyl sulfate, 10% glycerol, 0.001% bromophenol blue, 1% dithiothreitol, and 0.05-mmol/L Tris hydrochloride, pH 6.8), separated on 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis, and transferred to a nitrocellulose filter. The blots were blocked with 5% nonfat milk in PBS, then incubated with anti–NOS-2 polyclonal antibody (Santa Cruz Biotechnology Inc; working dilution, 1:100), followed by peroxidase-conjugated goat anti–rabbit IgG2a, and imaged by means of the enhanced chemiluminescence detection system (Amersham Life Science Inc, Arlington Heights, Ill). Negative control was run in parallel by preincubation at 4°C overnight with a specific blocking peptide (Santa Cruz Biotechnology Inc; working dilution, 1:5) to neutralize the anti–NOS-2 antibody. Quantitation was performed by scanning the Hyperfilm radiographs and measuring the band intensity by optical densitometry (OD) with an image densitometer (Model GS670; Bio-Rad Laboratories, Hercules, Calif) and inage quantitating software (Molecular Analysis 1.5; Bio-Rad Laboratories). The comparative increase of the protein level from control to pressure-exposed groups was calculated by comparing the OD values from 3 independent experiments with cells from different donors.
Semiquantitative reverse transcription polymerase chain reaction (RT-PCR) was performed as published previously.21 Briefly, 106 cells were homogenized in 1 mL of Trizol (Life Technologies, Rockville, Md), and total RNA was extracted according to the manufacturer's recommendations and quantified by measuring the absorbance at 260 nm of an aliquot. The ratio of 260/280 nm was greater than 1.8. Complementary DNA (cDNA) was synthesized by using 1 µg of RNA that was reverse transcribed by reverse transcriptase (SuperScript II RT; Life Technologies) with oligo (dT) primers according to the manufacturer's instructions (Life Technologies). The primers used for NOS-2 detection were 112S (5′-CCAGTGACACAGGATGACCTTCAG-3′), complementary to bases 112 through 135, and 715A (5′-TGCCATTGTTGGTGGAGTAACG-3′), complementary to bases 715 through 694 of the human NOS-2 cDNA sequence. The primers used for β-actin detection were 792S (5′-CTCTCTTCCAACCTTCCTTCCTG-3′), complementary to bases 792 through 814, and 1110A (5′-CCAGACTCGTCATACTCCTGCTTG-3′), complementary to bases 1110 through 1087 of human β-actin cDNA sequence. Polymerase chain reaction conditions were first optimized to ensure that the PCR was in the linear range. As a control for calibrating an equivalent amount of input cDNA, the messenger RNA (mRNA) level of constitutively expressed β-actin was determined in parallel aliquots of cDNA to control for differences in cDNA synthesis efficiency. Polymerase chain reaction conditions were 25 cycles of denaturation at 94°C for 1 minute, annealing at 60°C for 1 minute, and extension at 72°C for 30 seconds for NOS-2 and 15 cycles for β-actin. After amplification, amplicons were electrophoresed on 2% agarose gel and were blotted onto a positively charged nylon membrane (Hybond-N+;Amersham Life Science Inc). Hybridization was performed at 60°C with specific internal DNA probes corresponding to bases 142 to 552 of NOS-2 cDNA sequence for the PCR production of NOS-2, or corresponding to bases 871 to 998 of β-actin cDNA sequence for PCR production of β-actin. Samples were washed 2 times at 60°C under increasingly stringent salt conditions. Chemiluminescent detection of hybridized nucleotides was performed with enhanced chemiluminescent(ECL) detection system (Amersham Life Science Inc) and autoradiography for approximately 60 minutes on film (Hyperfilm-ECL; Amersham Life Science). Quantitation was performed by scanning the Hyperfilm radiographs and measuring the band intensity by OD with the use of an image densitometer (Model GS670; Bio-Rad Laboratories) and ImageQuant software. The OD value of the band intensity for NOS-2 was calibrated by dividing by that for β-actin from the parallel sample. The comparative increase of the RT-PCR production of NOS-2 mRNA from control to pressure-exposed groups was calculated by comparing the OD values from 3 independent experiments with cells from different donors.
Cultures of lamina cribrosa astrocytes from normal human optic nerve heads have positive staining for intracellular filamentous GFAP, confirming a distinguishing characteristic of astrocytes. When grown in a chamber at ambient atmospheric pressure for 48 hours, the astrocytes have flat, star-shaped cell bodies with a few short, thick processes. Immunocytochemical colocalization for NOS-2 demonstrated that most of the GFAP-positive astrocytes grown under ambient atmospheric pressure are negative for NOS-2, but a few have very faint positive labeling for NOS-2 in the paranuclear area (Figure 1A). When exposed to elevated hydrostatic pressure (60 mm Hg) for up to 48 hours, the shape and size of the astrocytes changed. These cells have much larger cell bodies with extensive, thinner, and longer processes. Immunocytochemical colocalization for NOS-2 demonstrated that most of the astrocytes exposed to elevated hydrostatic pressure for 48 hours are intensely positive for NOS-2. The presence of NOS-2 protein was apparent in the paranuclear region, most likely in the rough endoplasmic reticulum–Golgi region, as well as diffusely in the cell body and processes (Figure 1B).
The synthesis of NOS-2 protein in astrocytes under ambient pressure and elevated hydrostatic pressure was also demonstrated by immunoblot. Using a polyclonal antibody against NOS-2, a specific band, approximately 120 kd, was detected by immunoblotting of the cell lysates. By pretreating the antibody against NOS-2 with a specific blocking peptide, the band at 120 kd was not apparent, confirming that this band is specific for NOS-2 (data not shown). The cell lysates from astrocyte cultures exposed to elevated hydrostatic pressure for 24, 48, and 72 hours contain much more intense bands for NOS-2 than the control culture at ambient pressure (Figure 2A). Similar results were obtained from astrocyte cultures from different human donors. Evaluation of the band intensity of NOS-2 at 24, 48, and 72 hours of elevated hydrostatic pressure indicated that the level of NOS-2 protein peaks at 48 hours of elevated hydrostatic pressure (Figure 2B).
To investigate the effect of hydrostatic pressure on the gene transcription of NOS-2, semiquantitative RT-PCR was performed on the NOS-2 mRNA of astrocytes exposed to elevated hydrostatic pressure for 12, 24, and 48 hours and compared with control cultures exposed to atmospheric pressure. Our results demonstrated that the mRNA level for NOS-2 was significantly enhanced in the astrocytes under elevated hydrostatic pressure. By RT-PCR, very intense bands indicating mRNA for NOS-2 were detected by hybridizing with the specific NOS-2 DNA probe from the samples of astrocytes exposed to elevated hydrostatic pressure for 12, 24, and 48 hours (Figure 3A), but not from samples from the control culture. The RT-PCR for mRNA for β-actin from the same samples showed no significant difference (Figure 3B). The constancy of the β-actin results confirms that the amount of mRNA from the individual samples for the determination of NOS-2 was relatively similar and that the induction of gene transcription for NOS-2 was not a general phenomenon that up-regulates the transcription of all genes. Similar results were obtained from astrocyte cultures from different human donors. Evaluation of the band intensity for mRNA for NOS-2 at 12, 24, and 48 hours of elevated hydrostatic pressure indicated that the mRNA level for NOS-2 in the astrocytes peaked markedly at 12 hours of exposure to elevated hydrostatic pressure and remained at a high level through 24 hours. After 48 hours of exposure to elevated hydrostatic pressure, the mRNA level for NOS-2 was still elevated compared with control but appeared to decrease from the peak (Figure 3C).
In this in vitro study, we investigated the effects of biomechanical stress on the induction of NOS-2 in astrocytes of the human optic nerve head by detecting the protein and mRNA levels for NOS-2 in cell cultures exposed to elevated hydrostatic pressure. Our results show that, under elevated hydrostatic pressure, the astrocytes of the human optic nerve head exhibit significantly increased gene transcription and synthesis of NOS-2 in vitro. Although we performed these initial experiments at 60 mm Hg, well above the intraocular pressure normally observed in patients with glaucoma, we believe that further investigations will demonstrate that the pressure-sensitive induction of NOS-2 will occur at lower hydrostatic pressures. The appearance of NOS-2 in the optic nerve head in glaucoma could be a primary event in the disease or secondary to cellular and tissue changes and/or the response to cytokines released in the damaged tissue. Our studies suggest that, in glaucomatous optic neuropathy, the astrocytes of the human optic nerve head may directly respond to elevated intraocular pressure by expressing NOS-2 and thereby damage the axons of the retinal ganglion cells by producing excessive NO.
Immunocytochemical analysis demonstrates an apparent increase in the amount of NOS-2 protein in the astrocytes exposed to elevated hydrostatic pressure when compared with controls. The immunolabeling of NOS-2 in the astrocytes exposed to elevated hydrostatic pressure is most intense in the rough endoplasmic reticulum–Golgi region, the site of new protein synthesis, as well as in the cytoplasm. The appearance of the immunolabeling of NOS-2 in the astrocytes induced by elevated hydrostatic pressure is the same as our previous observations with astrocytes that were stimulated by cytokines. The astrocytes that are exposed to elevated hydrostatic pressure exhibit larger cell bodies with extensive processes, perhaps indicating that elevated hydrostatic pressure stimulates the cells to become reactive astrocytes.
By immunoblot and semiquantitative RT-PCR, the protein and mRNA levels for NOS-2 in the astrocytes were demonstrated and compared semiquantitatively at sequential time points of exposure to elevated hydrostatic pressure and to ambient pressure. The synthesis of NOS-2 protein induced by elevated hydrostatic pressure reaches the peak level at 48 hours of exposure; however, the transcription of mRNA for NOS-2 reaches the peak level earlier, at 12 hours of exposure. The difference in the peak time between the mRNA level and the protein level of NOS-2 is consistent with the process of protein synthesis.
In optic nerve head astrocytes in vitro, the mRNA for NOS-2 is relatively short-lived. After stimulation with cytokines, the mRNA of NOS-2 is even more transiently expressed, with a 6-hour half-life.22 At the 3′ untranslated region of NOS-2 mRNA, there is a conserved adenine and uracile–rich octanucleotide sequence that mediates mRNA instability. Thus, the mRNA level of NOS-2 does not remain elevated for sustained periods, but reaches a maximum, after which induction is terminated and the mRNA is destroyed. It is therefore unlikely that any one astrocyte maintains induced NOS-2 throughout the chronic glaucomatous process over many years. In vivo, as glaucomatous optic neuropathy proceeds, individual astrocytes in local areas of the optic nerve head may be serially induced to express NOS-2, leading to focal areas of damage within the optic nerve head.
Induction of NOS-2 in astrocytes in response to a variety of stimuli has been reported,23- 25 but, to our knowledge, this is the first report that a mechanical stimulus, elevated hydrostatic pressure, can induce astrocytes to express NOS-2. The subcellular mechanisms of the responses of astrocytes to elevated hydrostatic pressure are unknown. With the use of patch clamp techniques, membrane ion channels have been shown to be pressure sensitive. In rat neonatal astrocytes, one class of ion channels, called curvature-sensitive channels, is activated when the cell membrane curves toward the soma under pressure, and another class of ion channels, called stretch-activated channels, is activated by suction.26,27 Using astrocytes from the human optic nerve head in experiments similar to ours in vitro, Ricard et al20 reported that hydrostatic pressure causes changes in the cytoskeleton and an increase in expression and synthesis of a specific isoform of neural cell adhesion molecule, which changes cell adhesion properties. How an extracellular mechanical stress signal is transmitted intracellularly and which signal cascades promote new gene expression in optic nerve head astrocytes remain to be determined.
A variety of cell types, such as vascular endothelial cells,28,29 neuronal cells,30 bone cells,31 marrow cells,32 articular chondrocytes,33 vascular smooth muscle cells,34 and periodontal ligament cells,35 have been reported to respond to hydrostatic pressure or other biomechanical stresses. Biomechanical stress causes cellular responses, including synthesis of cellular mediators such as growth factors and cytokines, opening and closing of ion channels, and the synthesis and degradation of extracellular matrix macromolecules. Agar et al30 reported that elevated hydrostatic pressure (100 mm Hg) causes 2 neuronal cell line cultures to undergo apoptosis. In vascular endothelium, the gene of another isoform of nitric oxide synthase, the endothelial nitric oxide synthase (NOS-3), can be potently and rapidly up-regulated by certain kinds of shear stress, particularly steady laminar shear stress, but not by turbulent shear stress.36 The shear stress–mediated stimulation of NO production through up-regulation of gene expression may involve the activation of "shear stress receptors" that are present in the endothelial cells, and the subsequent rapid production of an intracellular second messenger.37 A variety of other proteins are rapidly induced and activated by shear stress in endothelial cells, including certain cell surface potassium channels, members of the mitogen-activated and stress-activated protein kinase cascades, certain transcription factors such as nuclear factor-κB, and subsets of receptor-associated G proteins.38 Thus, for certain cell types, the biomechanical microenvironment of the cells might be as important as the biochemical one.
Our studies in vitro indicate that the in vivo expression of NOS-2 in the astrocytes of the human glaucomatous optic nerve head, which we have demonstrated previously,6 may be a direct response to the elevated intraocular pressure that is characteristic of glaucoma. The optic nerve head fills the scleral canal and is, therefore, between 2 pressure compartments, the intraocular compartment (intraocular pressure) and the central nervous system (retrolaminar tissue pressure). The lamina cribrosa of the optic nerve head is a connective tissue suspended perpendicularly to the pressure gradient between the intraocular compartment and the central nervous system. In eyes with normal intraocular pressure, this pressure gradient varies because of the ocular pulse and the diurnal changes in intraocular pressure.39,40 In eyes with glaucoma, there are elevated intraocular pressure, spikes of increased intraocular pressure, and diurnal changes of intraocular pressure. Thus, significant deformation of the lamina cribrosa occurs in glaucoma, which may generate biomechanical stress on the astrocytes and other cell types that are embedded in the connective tissue and attached to the extracellular matrix macromolecules.41 The heightened and changing hydrostatic pressure gradient within the optic nerve head may directly induce astrocytes to express NOS-2 in vivo, as we find in vitro. Further studies on the molecular regulation of intraocular pressure–induced NOS-2 expression in astrocytes of the human optic nerve head may lead to a new appreciation of the role of cellular responses in glaucomatous optic neuropathy and new therapeutic approaches for accomplishing neuroprotection in patients with glaucoma.
Accepted for publication August 9, 2000.
This work was supported in part by grant EY12017 and core grant EY02687 from the National Eye Institute, Bethesda, Md, and by the Glaucoma Foundation, New York, NY.
We acknowledge M. Rosario Hernandez, DDS, for developing the pressure chamber used in these experiments and for assistance with the chamber and the cell culture.
Corresponding author and reprints: Arthur H. Neufeld, PhD, Department of Ophthalmology and Visual Sciences, Box 8096, Washington University School of Medicine, 660 S Euclid Ave, St Louis, MO 63110 (e-mail: firstname.lastname@example.org).