CE is an employee of Pfizer Ltd. There are no patents, products in development or marketed products to declare. This does not alter our adherence to all the PLOS ONE policies on sharing data and materials.
Conceived and designed the experiments: AO JDR DCB JS. Performed the experiments: AO AA. Analyzed the data: AO JS AA JDR CE. Contributed reagents/materials/analysis tools: JDR. Wrote the paper: AO JS CE.
Elevated intraocular pressure (IOP) is a major risk factor for glaucoma. One consequence of raised IOP is that ocular tissues are subjected to increased hydrostatic pressure (HP). The effect of raised HP on stress pathway signaling and retinal ganglion cell (RGC) survival in the human retina was investigated.
A chamber was designed to expose cells to increased HP (constant and fluctuating). Accurate pressure control (10-100mmHg) was achieved using mass flow controllers. Human organotypic retinal cultures (HORCs) from donor eyes (<24h
Exposure of HORCs to constant (60mmHg) or fluctuating (10-100mmHg; 1 cycle/min) pressure for 24 or 48h caused no loss of structural integrity, LDH release, decrease in RGC marker expression (
Directly applied HP had no detectable impact on RGC survival and stress-signalling in HORCs. Simulated ischemia, however, activated stress pathways and caused RGC death. These results show that direct HP does not cause degeneration of RGCs in the
Glaucoma is a group of optic neuropathies leading to progressive loss of visual field due to the degeneration of retinal ganglion cells (RGCs) in the inner retina and loss of their axons in the optic nerve [
It has proven difficult to isolate the contribution of individual variables that are affected in the eye as a result of increased IOP, which may subsequently lead to RGC death. One direct component affected by raised IOP is an increase in hydrostatic pressure (HP): when IOP increases in the eye, the retina will experience an increase in HP, acting transversely across the retina.
Donor human eyes were obtained from the East Anglian Eye Bank with ethical approval (Ref 04/Q0102/57; NHS Research Ethics Committee), with written consent from the donors’ next-of-kin and in compliance with the tenets of the Declaration of Helsinki. Retinal dissection and HORC preparation was performed as described previously [
A custom-made chamber was constructed (UEA mechanical workshop, Norwich, UK) from Perspex to expose tissue explants to increased HP (
(A) Schematic diagram of the hydrostatic pressure system (not to scale). Examples of computer controlled protocols using the pressure system at (B) constant (60mmHg) pressure for 24h and (C) fluctuating (10–100mmHg; 1 cycle/min) pressure for 1h. MFC = mass flow controller.
The chamber used mass flow controllers (MFCs), positioned at the inlet and outlet ports, to simultaneously regulate the internal pressure and the rate of gas flow through the chamber. Pressurised gas (95% air/ 5% CO2) could be rapidly injected into the chamber using a 1000ml/min MFC and released via a solenoid exhaust valve. Custom written software regulated internal pressure based on levels measured by a digital pressure sensor (Omega Engineering Inc, Manchester, UK). The software was able to control gas flow via an analogue to digital interface which operated the MFC and exhaust valve (
No significant changes in pH or evaporation rate were detected between control and medium exposed to pressure for the experimental period (data not shown). pH was measured following removal of the medium from the chamber using a glass electrode (ThermoScientific, Loughborough, UK). Evaporation was assessed by weighing the medium before and after exposure to experimental conditions.
O2 concentration in distilled water or culture medium exposed to pressure was measured using a Hansatech DW1 Oxygen Electrode (Hansatech Instruments Ltd, Norfolk, UK). The system was calibrated before each use with air saturated water or medium and oxygen-free water or medium (bubbled with 95% N2, 5% CO2 for 10min). 35mm culture dishes containing 1.5ml solution were exposed to various pressures or control conditions for 30min. 1ml of treated solution was then placed in the oxygen electrode reaction vessel. Oxygen concentrations were measured every second for ~1min whilst constantly stirring at 450rpm. The mean values for each oxygen concentration measurement were recorded (nmol/ml). The effect of pressure on O2 concentration in our pressure system closely followed that predicted by Henry’s Law [
O2 concentration in water and medium is expressed as the percentage of the concentration recorded at atmospheric pressure (n = 4). The gas in the chamber was 95% air/ 5% CO2. The O2 concentration in pure water predicted by Henry’s Law is also shown.
HORCs were exposed to oxygen glucose deprivation (OGD) as described previously [
The level of cell death was determined by measuring the LDH activity in cell culture medium according to the manufacturer’s instructions (Roche Molecular Biochemicals, Burgess Hill, UK).
Total RNA was extracted from HORCs using the RNeasy Mini Kit (Qiagen, Crawley, UK) according to the manufacturer’s instructions. The concentration of total RNA was measured using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, USA). Total RNA was reverse transcribed to complementary DNA (cDNA) in a reaction mix of Superscript II reverse transcriptase (Invitrogen, Paisley, UK), dNTP mix (Bioline, London, UK) and random primers (Promega, Southampton, UK) according to manufacturer instructions.
TaqMan PCR was performed using 5ng of input cDNA and Taqman PCR mastermix (Applied Biosystems, Warrington, UK) and human
Immunohistochemistry and TUNEL-labelling were used to assess the number of surviving RGCs in HORCs as described previously [
The primary antibody used was mouse monoclonal NeuN (1:200) (Chemicon International, Millipore, Watford, UK) and the secondary antibody was goat anti-mouse AlexaFluor 488 or 555 (1:1000) (Invitrogen, Paisley, UK). For the TUNEL assay (Promega, Southampton, UK), retinal slices were washed and immersed in TUNEL equilibration buffer for 10min, 18h after primary antibody binding. Slices were incubated in TUNEL reaction mixture for 1h at 35°C before stopping the reaction by immersion in standard citrate solution (SCS). After further washing, nuclei were stained with DAPI (1:100; Sigma-Aldrich, Poole, UK).
18 × 200μm sections from each HORC were counted in a masked fashion. The number of NeuN-labelled cells co-localising with DAPI were used as a measure of RGC number. NeuN positive cells which also stained positive for TUNEL were identified as apoptotic RGCs. It is important to note that there is no major staining of NeuN in the inner nuclear layer suggesting that NeuN does not label amacrine cells [
Protein lysates were obtained from HORCs using Mammalian Protein Extract Reagent M-PER supplemented with Halt Phosphatase Inhibitor Cocktail, Protease Inhibitor Cocktail and 5mM EDTA (All from Thermo Scientific, Loughborough, UK) for 20min on ice followed by centrifugation at 13,000rpm for 5min. Protein concentration of each lysate was determined using a bicinchoninic acid (BCA) protein assay (Thermo Scientific, Loughborough, UK). Equal amounts of protein were loaded onto 10% SDS-PAGE gels and proteins separated by electrophoresis. Proteins were transferred to PVDF membrane (Perkin Elmer Life Sciences, Cambridge, UK) using a semi-dry transfer blotter (Bio-Rad Laboratories, Hemel Hempstead, UK). Membranes were blocked with PBS-T (0.1% Tween-20 in PBS, 5% fat-reduced milk), hybridized with primary antibody followed by incubation with secondary antibody (GE Healthcare, Buckinghamshire, UK). Bands were visualised using chemiluminescent ECL Plus Western Blot Detection reagent (GE Healthcare, Buckinghamshire, UK) and net band intensity determined (1D 3.5 software, Eastman Kodak, Rochester, NY). Primary antibodies (Cell Signaling Technology, Danvers, MA, USA) against phospho- and total p38, phospho- and total JNK were used at 1:250, 1:1000, 1:500 and 1:500 respectively.
Data shown is the mean ± standard error of the mean (S.E.M). Significance was determined using an unpaired Student’s t-test (GraphPad Prism version 6.0, San Diego, USA). Differences were considered significant at the p≤0.05 level. Groups were considered statistically similar if p≥0.2 (β=0.2) and p values are given throughout. Due to having only one chamber, pressure experiments were carried out independently using separate donors with appropriate same donor controls.
There was no significant increase in released LDH as a result of either constant or fluctuating pressure at 24h (HP(C) 60mmHg—n = 20, p = 0.564; HP(F) 10–100mmHg 1 cycle/min—n = 8, p = 0.794) or 48h (HP(C) 60mmHg—n = 20, p = 0.907; HP(F) 10–100mmHg—n = 8, p = 0.838) compared with controls (
(A) No increase in necrotic cell death, measured by released cytoplasmic LDH, was observed after constant (HP (C); 60mmHg) or fluctuating (HP (F); 10–100mmHg; 1cycle/min) pressure for 24 or 48h (HP(C) 60mmHg 24h—n = 20, p = 0.564; HP(C) 60mmHg 48h—n = 20, p = 0.907; HP(F) 10–100mmHg 24h—n = 8, p = 0.794; HP(F) 10–100mmHg 48h—n = 8; p = 0.838). A positive control of 3h OGD/21h control conditions led to a significant increase in released LDH compared to control conditions (n = 11; *p = 0.0001). (B-D) Representative immunofluorescence photomicrographs of HORCs; (B) 24h control (i) or pressure (ii, iii) exposure, (C) 48h control (i) or pressure (ii, iii) exposure and (D) 24h control (i) or 3h OGD/21h control conditions (ii). DAPI = blue, NeuN = green, GCL = ganglion cell layer, INL = inner nuclear layer, ONL = outer nuclear layer. Scale = 200μm.
Focussing more specifically on survival of RGCs in HORCs, NeuN labelling and
(A) Constant (HP(C); 60mmHg) or fluctuating (HP(F) 10–100mmHg; 1cycle/min) pressure did not decrease the number of NeuN-labelled RGCs at the 24 or 48h time-points (HP(C) 60mmHg 24h—n = 9, p = 0.947; HP(C) 60mmHg 48h—n = 9, p = 0.668; HP(F) 10–100mmHg 24h—n = 10, p = 0.955; (HP(F) 10–100mmHg 48h—n = 10; p = 0.733). A significant reduction in NeuN-labelled cells was observed following simulated ischemia (3h OGD/21h control conditions) (n = 9; *p = 0.002). (B) Elevated HP for 24 or 48h did not reduce
Since it might be expected that decline in RGC number could occur later than 48h, but that apoptosis may have been initiated during this period, the number of TUNEL-positive NeuN-labelled cells was also assessed (
Investigation of the stress pathways p38 and JNK showed no increased activation (phosphorylation) in HORCs following exposure to fluctuating pressure (10–100mmHg; 1 cycle/min) at 15 min (n = 3; p38 p = 0.769; JNK p = 0.354), 30 min (n = 3; p38 p = 0.696; JNK p = 0.667), 60 min (n = 3; p38 p = 0.232; JNK p = 0.891) and 90min (n = 3; p38 p = 0.0.273; JNK p = 0.833) (
Phosphorylation of (A) p38 and (B) JNK, relative to their total expression, did not significantly alter with fluctuating pressure in HORCs (n = 3; 15 min- p38 p = 0.769, JNK p = 0.354; 30 min—p38 p = 0.696, JNK p = 0.667; 60 min—p38 p = 0.232, JNK p = 0.891; 90min-p38 p = 0.273, JNK p = 0.833). Phosphorylation of (C) p38 and (D) JNK was observed immediately following 3h OGD (n = 3; 0 min—p38 p = 0.012, JNK p = 0.006), and in the during the following reperfusion period in control medium (n = 3; 60 min—p38 p = 0.019, JNK p = 0.039; 90 min—JNK p = 0.049). Results are expressed as a percentage of the untreated control. Representative blots are shown.
Although ocular hypertension has been identified as a major risk factor for glaucoma, precisely how raised IOP translates into loss of RGCs and consequent visual field deterioration is poorly understood. Several previous studies have suggested that increased HP can induce RGC death [
Since we were using a custom-made pressure chamber, it was important to validate the system and consider any potential confounding factors. By using MFCs it was shown that HP could be accurately increased within the chamber and also be tightly regulated. Pressure increased to the target pressure within 30sec and was maintained within ±1mmHg. Using this system, we could be confident that no uncontrolled initial pressure surges were experienced by the tissue, such as could occur if the chamber were connected directly to a gas cylinder. Also using this system we could be confident that there was no movement of the tissue, either via fluid turbulence or movement of the underlying substrate. We were, in turn, confident that the tissue was exposed purely to raised HP and that we had not inadvertently introduced any mechanical distortion. We measured evaporation of medium from dishes in the chamber and found no difference at raised HPs compared to control dishes outside of the chamber, such that one would not anticipate any exposure to differing osmotic conditions. In addition, in design of the system we enabled a constant gas flow through the chamber, independent of pressure regulation, in order to mitigate against changes in gas composition (albeit very small due to the large volume of this chamber) as a result of tissue respiration. It does, however, have to be addressed, that some changes could not be mitigated against when using this design of chamber. Specifically, in chambers that increase HP by raising the gas pressure at a gas-liquid interface, the concentration of dissolved gases in the medium must be considered. Increasing pressure in the gas phase increases the partial pressure of each gas within this phase; this leads to a proportional increase in the concentration of dissolved gases, including O2, in the liquid phase (ie. the medium) as described by Henry’s Law. An increase in O2 was measured in the medium within our chamber (
Exposing the retinal explants to increased HP for up to 48h did not cause a reduction in RGC survival or induction of apoptosis in response to constant (60mmHg) or fluctuating pressure (10–100mmHg; 1 cycle/min). In contrast, as a positive control, we exposed HORCs to simulated ischemia which did cause significant loss of RGCs. Increased p38 and JNK phosphorylation has previously been described in animal models of glaucoma [
To our knowledge, only one previous paper has investigated the effects of HP on retinal explants [
Other studies on the effects of raised HP have utilised isolated retinal cells, cultured on rigid, artificial substrates specifically glass and tissue culture plastic [
It should be remembered that HP only constitutes a small component of forces associated with elevated IOP, specifically, the transverse stress across the retina. In the eye
In our experiments, it was found that applying HP to retinal explants did not result in RGC death or influence pathways associated with changes in survival. We would therefore suggest that the component of raised IOP that is modelled by increasing HP, i.e. the transverse stress across the retina that increases as IOP is raised, is not a direct contributor to RGC death. Certainly our results are consistent with the compelling argument that application of HP alone is not a surrogate for IOP in glaucoma [
The authors would like to express their gratitude to Pamela Keeley, Mary Tottman and Samantha Major at the East Anglian Eye Bank for donor eye retrieval and EWS UEA for manufacturing the pressure chamber and control system.