SEARCH

SEARCH BY CITATION

Keywords:

  • bipolar cell;
  • cell swelling;
  • glutamate;
  • Müller cell;
  • osmotic stress;
  • retina

Abstract

  1. Top of page
  2. Abstract
  3. Experimental procedures
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Regulation of cellular volume is of great importance to avoid changes in neuronal excitability resulting from a decrease in the extracellular space volume. We compared the volume regulation of retinal glial (Müller) and neuronal (bipolar) cells under hypoosmotic and glutamate-stimulated conditions. Freshly isolated slices of the rat retina were superfused with a hypoosmotic solution (60% osmolarity; 4 min) or with a glutamate (1 mM)-containing isoosmotic solution (15 min), and the size changes of Müller and bipolar cell somata were recorded. Bipolar cell somata, but not Müller cell somata, swelled under hypoosmotic conditions and in the presence of glutamate. The hypoosmotic swelling of bipolar cell somata might be mediated by sodium flux into the cells, because it was not observed under extracellular sodium-free conditions, and was induced by activation of metabotropic glutamate receptors and sodium-dependent glutamate transporters. The glutamate-induced swelling of bipolar cell somata was mediated by sodium chloride flux into the cells induced by activation of NMDA- and non-NMDA glutamate receptors, glutamate transporters, and voltage-gated sodium channels. The glutamate-induced swelling of bipolar cell somata was abrogated by adenosine and γ-aminobutyric acid, but not by vascular endothelial growth factor and ATP. The data may suggest that Müller cells, in contrast to bipolar cells, possess endogenous mechanisms which tightly regulate the cellular volume in response to hypoosmolarity and prolonged glutamate exposure. Inhibitory retinal transmission may regulate the volume of bipolar cells, likely by inhibition of the excitatory action of glutamate.

Abbreviations used
AMPA

α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid

ATP

adenosine 5′-triphosphate

BAPTA-AM

bis-(o-aminophenoxy)ethane-N,N,N″,N″-tetra-acetic acid acetoxymethyl ester

CNQX

cyanonitroquinoxalinedione

EAAT

excitatory amino acid transporter

GABA

γ-aminobutyric acid

Kir

inwardly rectifying potassium

l3HA

l-(−)-threo-3-hydroxyaspartic acid

l-AP3

l-2-amino-3-phophonopropanoicacid

l-NAME

Nω-nitro-l-arginine methyl ester hydrochloride

lβBA

l-β-threo-benzyl-aspartate

LY341495

(2S)-2-amino-2-[(1S,2S)-2-carboxycycloprop-1-yl]-3-(xanth-9-yl) propanoic acid

MK-801

dizocilpine

MPDC

l-anti-endo-3,4-methanopyrrolidinedicarboxylate

MPEP

2-methyl-6-(phenylethynyl)-pyridine

NMDA

N-methyl-d-aspartate

NPPB

5-nitro-2-(3-phenylpropylamino)benzoic acid

TBOA

d,l-threo-ß-benzyloxyaspartate

VEGF

vascular endothelial growth factor

Intense activation of retinal neurons is associated with a swelling of neuronal somata, processes, and synapses (Uckermann et al. 2004). Glutamate-induced swelling of retinal ganglion cells of the guinea pig is caused by ion flux through ionotropic receptors which is associated with a water flux (Uckermann et al. 2004). Cellular swelling results in a decrease in the extracellular space volume (Uckermann et al. 2004) which, if uncompensated, will cause neuronal hyperexcitation (Dudek et al. 1990; Chebabo et al. 1995). Intense neuronal activity is also associated with a decreased osmolarity of the extracellular fluid because the decrease in extracellular calcium, sodium, and chloride is approximately twice as large as the increase in potassium (Dmitriev et al. 1999).

It has been shown that Müller glial cells of the rodent retina do not swell for at least 10 min under hypoosmotic conditions (Hirrlinger et al. 2008). However, a blockade of inwardly rectifying potassium (Kir) channels results in immediate swelling of Müller cells under hypoosmotic conditions, suggesting that potassium flux through Kir channels is one mode of Müller cell volume regulation (Pannicke et al. 2004, 2006). In addition, Müller cells of the rodent retina possess a volume-regulatory glutamatergic-purinergic signal transduction cascade which, upon activation by, for example, vascular endothelial growth factor (VEGF), inhibits cellular swelling under hypoosmotic conditions (Wurm et al. 2008, 2011; Krügel et al. 2010). This signaling cascade consists of a consecutive release of glutamate, adenosine 5′-triphosphate (ATP), and adenosine from Müller cells which activate metabotropic glutamate receptors, P2Y1, and adenosine A1 receptors, respectively (Uckermann et al. 2006; Wurm et al. 2008, 2010; Krügel et al. 2010). The final step of the signaling cascade is the autocrine activation of adenosine A1 receptors which results in opening of potassium and chloride channels; the ion efflux compensates the osmotic gradient across the Müller cell membrane and thus prevents cellular swelling (Uckermann et al. 2006; Wurm et al. 2009).

The receptor-mediated inhibition of hypoosmotic swelling suggests that Müller cells regulate the extracellular space volume in the retina (Bringmann et al. 2006; Wurm et al. 2011). However, it is unclear whether retinal neurons contribute to the regulation of the extracellular space volume under hypoosmotic and glutamate-stimulated conditions. Therefore, we compared the osmotic swelling characteristics of Müller cell and bipolar cell somata in the inner nuclear layer of freshly isolated slices of the rat retina. We show that bipolar cell somata, but not Müller cell somata, swell under hypoosmotic conditions and in the presence of glutamate. The size of bipolar cell somata was recorded under iso- and hypoosmotic conditions, in the absence and presence of receptor agonists which have been shown to inhibit the barium-induced hypoosmotic swelling of Müller cells, that is, VEGF, glutamate, ATP, and adenosine (Wurm et al. 2008, 2011; Krügel et al. 2010). None of these receptor ligands inhibited the hypoosmotic swelling of bipolar cell somata, while adenosine and γ-aminobutyric acid (GABA) prevented the swelling of bipolar cell somata induced by glutamate. Using pharmacological blockers, we determined the cellular mechanisms of hypoosmotic and glutamate-induced swelling of bipolar cells.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Experimental procedures
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Materials

Chloromethyltetramethylrosamine (Mitotracker Orange) was from Molecular Probes (Invitrogen, Eugene, OR, USA). Human recombinant VEGF-A165 was from Chemicon (Temecula, CA, USA). Nω-nitro-l-arginine methyl ester hydrochloride (l-NAME) was from Alexis Biochemicals (San Diego, CA, USA). AMPA, BAPTA-AM, dihydrokainic acid, l-(−)-threo-3-hydroxyaspartic acid (l3HA), LY341495, l-anti-endo-3,4-methanopyrrolidinedicarboxylate (MPDC), 2-methyl-6-(phenylethynyl)-pyridine (MPEP), d,l-threo-ß-benzyloxyaspartate (TBOA), and tertiapin-Q were from Tocris Cookson (Ellisville, MO, USA). Cyanonitroquinoxalinedione (CNQX), dithiothreitol, kainate, l-ß-threo-benzyl-aspartate (lßBA), l-2-amino-3-phophonopropanoicacid (l-AP3), MK-801, NMDA, 5-nitro-2-(3-phenylpropylamino) benzoic acid (NPPB), tetrodotoxin, and all other agents used were purchased from Sigma-Aldrich (Taufkirchen, Germany), unless stated otherwise. The following antibodies were used: rabbit anti-protein kinase-α (1 : 200; Santa Cruz Biotechnology, Santa Cruz, CA, USA), mouse anti-glutamine synthetase (1 : 1000; Millipore, Schwalbach, Germany), Cy2-coupled goat anti-rabbit (1 : 400; Jackson Immuno Research, Newmarket, UK), and Cy3-coupled goat anti-mouse (1 : 400; Jackson).

Animals

All experiments were done in accordance with the European Communities Council Directive 86/609/EEC, and were approved by the local authorities (Medical Faculty of the University of Leipzig and Landesdirektion Leipzig). Adult pigmented Long-Evans rats (250–350 g; both sexes) were used. Animals were bred in the Medical-Experimental Center of the Faculty of Medicine, University of Leipzig, and were maintained with free access to water and food in an air-conditioned room on a 12-h light–dark cycle. Animals were killed with carbon dioxide, and the eyes were removed.

Retinal ischemia-reperfusion

In three animals, retinal ischemia-reperfusion was induced in one eye. The animals were anesthesized with intramuscular ketamine (50 mg/kg body weight) and xylazine (3 mg/kg). The anterior chamber of the treated eye was cannulated from the pars plana with a 30-gauge infusion needle, connected to a bag containing normal saline. The intraocular pressure was increased to 160 mmHg for 60 min by elevating the saline bag. After 3 days, the animals were killed with carbon dioxide, and the eyes were removed.

Preparation of retinal slices

To prepare retinal slices, freshly isolated retinas were placed with the photoreceptor side onto membrane filters (mixed cellulose ester, 0.45 μm pore size, 50 mm diameter; Schleicher & Schuell MicroScience, Dassel, Germany). Retinal slices (thickness, 1 mm) were cut from these tissues adhered to the membrane filters.

Cell soma swelling

All experiments were performed at 20–23°C. To determine the volume changes of Müller and bipolar cell somata induced by hypoosmotic challenge, the cell bodies in the inner nuclear layer of retinal slices were recorded. Müller cell bodies are localized to the central part of the inner nuclear layer, while bipolar cell bodies are localized to the outer part of this layer (Fig. 1a and b). In addition, the outer processes of both cell types display different morphology, that is, bipolar cells have a short and thick dendritic tree which contacts the synapses in the outer plexiform layer (Fig. 1a and b), whereas Müller cells have one relatively thin outer stem process which draws to the outer nuclear layer (Fig. 1b).

image

Figure 1. Localization of Müller and bipolar cell somata in slices of the rat retina. (a) Retinal slices were immunostained against the glial cell marker glutamine synthetase (red) and the marker of bipolar cells and distinct amacrines, protein kinase-α (green). Cell nuclei were labeled with Hoechst 33258 (blue). Note that bipolar cell somata (arrowheads) are localized at the outer border of the inner nuclear layer (INL), whereas Müller cell somata (arrows) are localized to the mid of the INL. (b) Examples of acutely isolated retinal slices stained with Mitotracker Orange. The focus is at the INL which contains Müller cell somata (arrows) and bipolar cell somata (arrowheads). Note that, in the case of Müller cells, two thin processes arise from the soma in opposite direction whereas bipolar cells have one thin process which draws to the inner plexiform layer (IPL) and short thick dendrites which have contact to the ribbon synapses in the outer plexiform layer (OPL). GCL, ganglion cell layer; ONL, outer nuclear layer. Bars, 5 μm.

Download figure to PowerPoint

The filter stripes with the retinal slices were transferred to a custom made perfusion chamber and kept submerged in extracellular solution. The chambers were mounted on the stage of an upright confocal laser scanning microscope (LSM 510 Meta; Zeiss, Oberkochen, Germany). To identify cell somata, retinal slices were loaded with the vital dye Mitotracker Orange (1 μM) for 3 min. The stock solution of the dye was prepared in dimethylsulfoxide and resolved in saline. After dye loading, the slices were continuously superfused with extracellular solution at a flow rate of 2 mL/min. Recordings were made with an Achroplan 63x/0.9 water immersion objective (Zeiss). The pinhole was set at 151 μm; the thickness of the optical section was adjusted to 1 μm. Mitotracker Orange was excited at 543 nm with a HeNe laser, and emission was recorded with a 585 nm long-pass filter. Images were obtained with an x-y frame size of 256 × 256 pixel (73.1 × 73.1 μm). The somata of dye-filled glial and bipolar cells were focused at the plane of their maximal extension. Cell somata were recorded at depths between 15 and 45 μm from the slice surface. To assure that the maximum soma areas were precisely recorded, the focal plane was continuously adjusted during the course of the experiments.

Solutions

A gravity-fed system with multiple reservoirs was used to perfuse the recording chamber continuously with extracellular solution; test substances were applied by rapid change in the perfusate. The bathing solution in the recording chamber was totally changed within 1 min. The extracellular solution consisted of (in mM) 136 NaCl, 3 KCl, 2 CaCl2, 1 MgCl2, 10 HEPES, and 11 glucose, adjusted to pH 7.4 with Tris. The hypoosmotic solution (60% of control osmolarity) was made up by adding distilled water or by reducing the NaCl concentration to 60%. The hyperosmotic solution was made up by addition of sucrose (100 mM) to the extracellular solution. Barium chloride (1 mM) was pre-incubated for 10 min in extracellular solution before it was applied within the hypoosmotic solution. Receptor agonists were applied simultaneously with the hypoosmotic solution. Blocking agents were pre-incubated for 15–45 min before experimental treatments. The nominally calcium-free extracellular solution contained 0.1 mM CaCl2 and 1 mM ethyleneglycolbis(aminoethyl)-(ether)tetra-acetate. Sodium-free solution was prepared by replacing NaCl by choline chloride. For preparing the low-chloride solution, NaCl was exchanged by sodium gluconate. The tissues were superfused for 15 min with calcium- or sodium-free solution before the addition of test substances.

Immunohistochemistry

Isolated retinas were fixed in 4% paraformaldehyde for 45 min. After several washing steps in buffered saline, the tissues were embedded in saline containing 3% agarose (w/v), and 60-μm thick slices were cut with a vibratome. The slices were incubated in 5% normal goat serum plus 0.3% Triton X-100 and 1% dimethylsulfoxide in saline for 1 h at 37°C and, subsequently, in primary antibodies overnight at 4°C. After several washing steps with saline, the secondary antibodies were applied for 2 h at 20–23°C. No specific staining was found in negative control slices which were stained without primary antibodies (not shown). Images were taken with the laser scanning microscope.

Data analysis

To determine the extent of cell soma swelling, the cross-sectional area of the cell bodies was measured off-line using the image analysis software of the laser scanning microscope (Zeiss LSM Image Examiner version 3.2.0.70.). For each test, 5–16 cells from two to three animals were measured. Values are given in mean ± SEM. Statistical analysis was made with the Prism program (Graphpad Software, San Diego, CA, USA); significance was determined by the non-parametric Mann–Whitney U-test, and was accepted at < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Experimental procedures
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Hypoosmotic swelling of Müller cell somata

The hypoosmolarity-induced volume changes of Müller and bipolar cell somata were recorded in acutely isolated retinal slices which were stained with the vital dye Mitotracker Orange (Fig. 1b). As previously described (Pannicke et al. 2004; Uckermann et al. 2006; Hirrlinger et al. 2008), superfusion of rat retinal slices with iso- or hypoosmotic extracellular solution up to 7 min did not induce swelling of Müller cell somata (Figs 2a, b and 3). The absence of Müller cell soma swelling was observed in hypoosmotic solutions which were prepared by dilution of the extracellular solution to 60% of control osmolarity (Figs 2a and 3) and by reducing only the NaCl concentration to 60% (Fig. 2b), respectively. However, Müller cell somata swelled immediately when the slices were superfused with hypoosmotic solution containing barium chloride (Figs 2b and 3). Barium ions block Kir channels of Müller cells (Pannicke et al. 2004, 2006). It has been suggested that the efflux of potassium through Kir4.1 channels provides rapid compensation of the osmotic gradient across the plasma membrane and, thus, prevents cellular swelling under hypoosmotic conditions (Pannicke et al. 2004; Wurm et al. 2006a, b). Inactivation of Kir channels by barium ions blocks this pathway of rapid compensation of osmotic gradients across the Müller cell membrane.

image

Figure 2. Superfusion of freshly isolated slices of the rat retina with a hypoosmotic solution results in a swelling of bipolar cell somata but not Müller cell somata. The hypoosmotic solution was prepared by dilution of the extracellular solution to 60% of control osmolarity (a) and by reducing the NaCl concentration to 60% (b), respectively. (a) The diagram displays the time-dependent alterations in the mean (± SEM) cross-sectional areas of cell somata (n = 5–6 each) during the change from isoosmotic to hypoosmotic extracellular solution. The slices were superfused with the hypoosmotic solution at 2 min and thereafter. In addition, the mean area of bipolar cell somata (n = 6), which were superfused with isoosmotic extracellular solution for 9 min, is shown. Data are expressed in percent of the soma size recorded at the beginning of the experiment (100%). The images display original records of bipolar cell somata (above and middle) and a Müller cell soma (below) obtained before (left) and during (right) superfusion of retinal slices with hypoosmotic solution. Bars, 5 μm. *< 0.05 compared with data of Müller cell somata. (b) Mean cross-sectional areas of cell somata which were measured after a 4-min superfusion of retinal slices with a hypoosmotic solution, and are expressed in percent of the soma size recorded before beginning of superfusion (100%). The hypoosmotic solution was tested in the absence (control) and presence of barium chloride (1 mM). Each bar represents values obtained in 7 to 11 cells. *< 0.05 compared with data of Müller cell control.

Download figure to PowerPoint

image

Figure 3. Effects of vascular endothelial growth factor (VEGF), glutamate (Glu), ATP, adenosine (Ade), and GABA on the mean (± SEM) cross-sectional area of Müller cell (above) and bipolar cell somata (below) under isoosmotic and hypoosmotic conditions. Slices of control (left) and 3-days post-ischemic retinas (right) were superfused with iso- and hypoosmotic extracellular solution, respectively. The hypoosmotic solution (60% osmolarity) was tested in the absence and presence of barium chloride (1 mM). The receptor agonists were co-administered with the iso- and hypoosmotic solution, respectively, and were tested at the following concentrations: vascular endothelial growth factor (VEGF), 10 ng/mL; glutamate, 1 mM; ATP, 10 μM; adenosine, 10 μM; GABA, 1 mM. The data were measured after a 4-min superfusion of retinal slices with iso- or hypoosmotic solution, and are expressed in percent of the soma size recorded before beginning of superfusion (100%). The images display original records of swelling Müller cell (above) and bipolar cell somata (below) obtained before (left) and during (right) superfusion of the retinal slice with the hypoosmotic solution with (above) and without (below) barium chloride. Bars, 5 μm. Each bar represents values obtained in 5 to 12 cells. *< 0.05 compared with isoosmolarity control. < 0.05 compared with hypoosmolarity plus barium control. < 0.05 compared with hypoosmolarity control.

Download figure to PowerPoint

Hypoosmotic swelling of bipolar cell somata

Superfusion of retinal slices with isoosmotic extracellular solution up to 9 min did not alter the size of bipolar cell somata (Figs 2a and 3). However, superfusion of slices with hypoosmotic solution resulted in immediate time-dependent (Fig. 2a) swelling of bipolar cell bodies (Figs 2b and 3). Swelling of bipolar cell somata was also observed during superfusion of the slices with hypoosmotic solution containing barium chloride (Figs 2b and 3).

Hyperosmotic decrease in soma size

Under pathological conditions such as the diabetic state of hyperglycemia, the blood, ocular humors, and retinal extracellular fluid become hyperosmotic. Thus, we determined the effect of hyperosmolarity on the size of Müller cell and bipolar cell somata. The hyperosmotic extracellular solution was made up by addition of sucrose (100 mM) to the extracellular solution. As shown in Fig. 4, hyperosmotic challenge caused decreases in the size of both Müller cell and bipolar cell somata. The extent of the decrease in soma size was similar in Müller and bipolar cells, and in the absence and presence of barium chloride (Fig. 4). The lack of effect of barium ions may suggest that the hyperosmotic decrease in soma size was not mediated by potassium currents through Kir channels. The data suggest that Müller cells possess endogenous mechanisms that regulate cellular volume under hypoosmotic, but not under hyperosmotic conditions.

image

Figure 4. Comparison of the effects of hypo- and hyperosmotic extracellular solutions on the size of Müller cell and bipolar cell somata. The effects were recorded in the absence and presence of barium chloride (Ba; 1 mM). The hypoosmotic solution contained 60% of control osmolarity. The hyperosmotic solution was made up by addition of sucrose (100 mM) to the extracellular solution. The data were measured after a 4-min superfusion of retinal slices with hypo- or hyperosmotic solution, and are expressed in percent of the soma size recorded before beginning of superfusion (100%). Each bar represents values obtained in 5 to 13 cells. *< 0.05 compared with Müller cell hypoosmolarity control.

Download figure to PowerPoint

Receptor-mediated regulation of cell soma size

It has been shown that activation of various receptor subtypes, including VEGF receptor-2, metabotropic glutamate receptors, P2Y1, and adenosine A1 receptors, blocks the osmotic swelling of Müller cell somata under hypoosmotic conditions (Uckermann et al. 2005, 2006; Wurm et al. 2008, 2009, 2010). We found that agonists of these receptors, that is, VEGF, glutamate, ATP, and adenosine, inhibited the swelling of Müller cell somata observed during superfusion of retinal slices with hypoosmotic solution containing barium chloride (Fig. 3). On the other hand, GABA did not inhibit the barium-induced hypoosmotic swelling of Müller cell somata (Fig. 3). The swelling of bipolar cell somata observed under hypoosmotic conditions was not prevented by each of the receptor agonists tested (Fig. 3). The data may suggest that Müller cells, but apparently not bipolar cells, possess receptor-mediated mechanisms which prevent cellular swelling under hypoosmotic conditions.

Cell soma swelling in the post-ischemic retina

It has been shown that transient retinal ischemia-reperfusion induces alterations in the swelling characteristics of Müller cells, that is, superfusion of post-ischemic retinal slices with hypoosmotic solution (without barium) results in swelling of Müller cell somata which is not observed in control retinal slices (Fig. 3) (Pannicke et al. 2004). As previously described (Uckermann et al. 2006; Wurm et al. 2008), VEGF, glutamate, ATP, and adenosine blocked the hypoosmotic Müller cell swelling in slices of post-ischemic retinas (Fig. 3). Hypoosmotic exposure of post-ischemic retinal slices also induced swelling of bipolar cell somata (Fig. 3). However, the receptor agonists tested did not prevent the swelling of bipolar cell somata under these conditions (Fig. 3).

Glutamate-induced swelling of bipolar cell somata

It has been shown that prolonged exposure of tissue preparations of the guinea pig retina to glutamate results in swelling of retinal ganglion cells (Uckermann et al. 2004). To determine whether prolonged exposure to glutamate causes alterations in the size of rat Müller and bipolar cell somata, we superfused slices of the rat retina with isoosmotic solution containing glutamate for 5, 10, and 15 min, respectively. As shown in Fig. 5a, glutamate induced a swelling of bipolar cell somata (but not Müller cell somata) after 10 and 15 min of exposure. The glutamate-induced swelling was prevented by the inhibitory transmitters adenosine (Housley et al. 2009) and GABA (Fig. 5b), but not by VEGF and ATP, respectively (Fig. 5c).

image

Figure 5. Glutamate induces swelling of bipolar cell somata but not Müller cell somata. (a) Rat retinal slices were superfused with isoosmotic solution containing glutamate (1 mM) for 5, 10, and 15 min, respectively. For comparison, the effect of hypoosmolarity was tested after 4 min. The cross-sectional area of cell somata is expressed in percent of the soma size recorded before beginning of superfusion (100%). (b) The glutamate (1 mM)-induced swelling of bipolar cell somata was abrogated in the presence of adenosine (Ade; 10 μM) and GABA (1 mM), respectively. (c) The glutamate (1 mM)-induced swelling of bipolar cell somata was not prevented by vascular endothelial growth factor (VEGF) (10 ng/mL) and ATP (10 μM), respectively. In (b) and (c), the data were obtained after superfusion of retinal slices for 15 min with isoosmotic solution. Receptor agonists were pre-incubated for 15 min and simultaneously applied with the test solution containing glutamate. Each bar represents values obtained in 6 to 11 cells. *< 0.05 compared with Müller cell hypoosmolarity control. < 0.05 compared with isoosmolarity control. < 0.05 compared with glutamate control.

Download figure to PowerPoint

Mechanisms of glutamate-induced swelling of bipolar cell somata

The data indicate that both hypoosmolarity and glutamate may induce a swelling of bipolar cell somata. To determine the mechanisms of glutamate-induced swelling, different conditions and blocking agents were tested. As shown in Fig. 6, calcium-free extracellular conditions did not inhibit the glutamate-induced swelling of bipolar cell somata. However, the glutamate-induced swelling was fully abrogated under sodium-free and low-chloride extracellular conditions (Fig. 6). Superfusion of retinal slices for 15 min with calcium- or sodium-free extracellular solution, or with low-chloride solution (in the absence of glutamate), did not alter the size of bipolar cell somata (not shown). The data suggest that the glutamate-induced swelling of bipolar cell somata is mediated by influx of sodium chloride from the extracellular space; the ion influx is accompanied by a water influx resulting in cellular swelling.

image

Figure 6. Mechanisms of glutamate-induced swelling of bipolar cell somata. The cross-sectional soma area was measured after superfusion of retinal slices with isoosmotic solution for 15 min, and is expressed in percent of the soma size recorded before beginning of superfusion (100%). Superfusion of the slices with glutamate (1 mM) induced a swelling of bipolar cell somata. Glutamate was tested in the absence and presence of the following agents and solutions, respectively: a nominally calcium-free extracellular solution, sodium-free extracellular solution, low-chloride extracellular solution, the selective NMDA receptor antagonist, MK-801 (10 μM), the antagonist of non-NMDA glutamate receptors, cyanonitroquinoxalinedione (CNQX) (50 μM), the antagonist of group II metabotropic glutamate receptors, LY341495 (100 μM), the inhibitor of sodium-dependent high-affinity glutamate transporters, l-anti-endo-3,4-methanopyrrolidinedicarboxylate (MPDC) (100 μM), the blocker of excitatory amino acid transporters, d,l-threo-ß-benzyloxyaspartate (TBOA) (100 μM), the inhibitor of voltage-gated sodium channels, tetrodotoxin (TTX; 1 μM), plus the inhibitor of voltage-gated calcium channels, cadmium chloride (Cd; 100 μM), the inhibitors of potassium channels tetraethylammonium chloride (TEA; 10 mM) and barium chloride (1 mM), and the chloride channel blocker 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB; 100 μM), respectively. Each bar represents values obtained in 6–16 cells. *< 0.05 compared with control. < 0.05 compared with glutamate control.

Download figure to PowerPoint

Glutamate-induced sodium influx into bipolar cells may be mediated by ionotropic glutamate receptors, sodium-dependent glutamate transporters, and voltage-gated sodium channels. As shown in Fig. 6, the glutamate-induced swelling of bipolar cell somata was partly inhibited by the antagonist of non-NMDA glutamate receptors, CNQX, and fully prevented by the selective NMDA receptor antagonist, MK-801. In contrast, the antagonist of group II metabotropic glutamate receptors, LY341495, was without effect (Fig. 6). Superfusion of retinal slices with isoosmotic solution containing kainate or NMDA for 15 min induced a swelling of bipolar cell somata; this was not the case when AMPA was applied (Fig. 7). The data suggest that sodium flux through ionotropic glutamate receptors is one mechanism of glutamate-induced swelling of bipolar cell somata.

image

Figure 7. Activation of ionotropic glutamate receptors induces swelling of bipolar cell somata. The cross-sectional soma area was measured after superfusion of retinal slices with isoosmotic solution for 15 min, and is expressed in percent of the soma size recorded before beginning of superfusion (100%). Superfusion of the slices with glutamate (1 mM), kainate (100 μM), and NMDA (100 μM), respectively, induced swelling, while AMPA (100 μM) had no effect. Each bar represents values obtained in 7–14 cells. *< 0.05 compared with control. < 0.05 compared with glutamate control.

Download figure to PowerPoint

To determine whether glutamate transporter-mediated sodium flux contributes to the glutamate-induced swelling of bipolar cell somata, we tested the inhibitor of sodium-dependent high-affinity glutamate transporters, MPDC, and the blocker of excitatory amino acid transporters, TBOA. As shown in Fig. 6, both blockers decreased significantly (< 0.05) the glutamate-induced swelling, suggesting that sodium influx mediated by glutamate transporters contributes to the glutamate-induced swelling. Inhibition of neuronal activation by administration of the blocker of voltage-gated sodium channels, tetrodotoxin, plus the inhibitor of voltage-gated calcium channels, cadmium chloride, nearly fully inhibited the glutamate-induced swelling of bipolar cell somata (Fig. 6). The data suggest that sodium flux through voltage-gated sodium channels may play a role in glutamate-induced swelling of bipolar cell somata.

The involvement of potassium channels in the glutamate-induced swelling of bipolar cell somata was investigated by administration of the blocker of outwardly rectifying potassium channels, tetraethylammonium chloride, and the Kir channel blocker barium chloride. Whereas barium had no effect, tetraethylammonium prevented in part the glutamate-induced swelling of bipolar cell somata (Fig. 6). In addition, the chloride channel blocker NPPB did not inhibit the glutamate-induced swelling of bipolar cell somata (Fig. 6).

Mechanisms of hypoosmolarity-induced swelling of bipolar cell somata

In addition to glutamate, hypoosmolarity induced a swelling of bipolar cell somata (Figs 2a, b and 3). It has been shown that oxidative stress and influx of sodium ions are causative factors of hypoosmotic swelling of Müller cell somata in the presence of barium ions and under pathological conditions (Uckermann et al. 2005, 2006; Wurm et al. 2006b). To determine whether these factors are also involved in mediating the hypoosmotic swelling of bipolar cell somata, we tested various blocking agents. As shown in Fig. 8a, oxidative stress is unlikely to be a causative factor of the hypoosmotic swelling of bipolar cell somata because the swelling was not prevented by the sulfhydryl reducing reagent dithiothreitol, the nitric oxide synthase inhibitor l-NAME, and the peroxynitrite scavenger uric acid. These inhibitors were previously described to inhibit hypoosmotic swelling of Müller cell somata (Karl et al. 2011; Krügel et al. 2011).

image

Figure 8. Mechanisms of hypoosmolarity-induced swelling of bipolar cell somata. The cross-sectional soma area was measured after superfusion of retinal slices with hypoosmotic solution (60% osmolarity) for 4 min, and is expressed in percent of the soma size recorded before beginning of superfusion (100%). (a) The soma area was recorded in calcium-free, low-chloride, and sodium-free hypoosmotic extracellular solutions. The soma area was also recorded in the absence (control) and presence of the following agents: the reducing agent dithiothreitol (DTT; 3 mM), the nitric oxide synthase inhibitor Nω-nitro-l-arginine methyl ester hydrochloride (l-NAME) (250 μM), the peroxynitrite scavenger uric acid (1 mM), the cell-permeable calcium chelator, BAPTA-AM (100 μM), the NMDA receptor antagonist, MK-801 (10 μM), the antagonist of non-NMDA glutamate receptors, cyanonitroquinoxalinedione (CNQX) (50 μM), the antagonist of group I metabotropic glutamate receptors, l-2-amino-3-phophonopropanoicacid (l-AP3) (100 μM), the antagonist of group II metabotropic glutamate receptors, LY341495 (100 μM), and the antagonist of subtype 5 of group I metabotropic glutamate receptors, 2-methyl-6-(phenylethynyl)-pyridine (MPEP) (10 μM). (b) The soma area was recorded in the absence (control) and presence of the following agents: the blocker of excitatory amino acid transporters (EAATs) 1-5, d,l-threo-ß-benzyloxyaspartate (TBOA) (100 μM), the transportable inhibitor of EAAT1-4, l-(−)-threo-3-hydroxyaspartic acid (l3HA) (200 μM), the non-transportable inhibitor of EAAT2, dihydrokainic acid (DHK; 200 μM), and the non-transportable inhibitor of EAAT3 (and to a lesser extent, EAAT1), l-ß-threo-benzyl-aspartate (lßBA) (100 μM). (c) The soma area was recorded in the absence (control) and presence of the following agents: the inhibitor of voltage-gated sodium channels, tetrodotoxin (TTX; 1 μM), plus the inhibitor of voltage-gated calcium channels, cadmium chloride (Cd; 100 μM), the chloride channel blocker 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB) (100 μM), the inhibitor of outwardly rectifying potassium channels, tetraethylammonium chloride (TEA; 10 mM), and the blocker of G protein-gated Kir channels, tertiapin-Q (200 nM). Each bar represents values obtained in 5 to 11 cells. < 0.05 compared with control.

Download figure to PowerPoint

Superfusion of retinal slices with a calcium-free extracellular solution or a solution that contained the cell-permeable calcium chelator BAPTA-AM did not prevent the hypoosmotic swelling of bipolar cell somata (Fig. 8a). The swelling was also not abrogated during superfusion of the slices with a low-chloride extracellular solution (Fig. 8a). However, superfusion of the slices with a sodium-free extracellular solution fully prevented the swelling of bipolar cell somata (Fig. 8a). The data may suggest that hypoosmolarity might cause a sodium influx into the cells.

Sodium influx may be mediated by voltage-gated and receptor channels, and transporter molecules. The antagonists of NMDA and non-NMDA glutamate receptors, MK-801 and CNQX, did not prevent the hypoosmotic swelling of bipolar cell somata (Fig. 8a). However, the antagonists of group I and II metabotropic glutamate receptors, l-AP3 and LY341495, respectively, fully prevented the swelling, whereas the antagonist of the subtype 5 of group I metabotropic glutamate receptors, MPEP, was without effect (Fig. 8a). The hypoosmotic swelling of bipolar cell somata was also abrogated by the non-transportable inhibitor of excitatory amino acid transporters (EAATs) 1-5, TBOA, and the transportable inhibitor of EAAT1-4, l3HA (Fig. 8b). On the other hand, the non-transportable inhibitor of EAAT2 and EAAT3, dihydrokainic acid, and the inhibitor of EAAT1-3, lßBA, were without effects (Fig. 8b). The data suggest that activation of metabotropic glutamate receptors and glutamate transporters by endogenously released glutamate is involved in the induction of hypoosmotic swelling of bipolar cell somata.

Inhibition of voltage-gated sodium and calcium channels by tetrodotoxin and cadmium chloride did not affect the hypoosmotic swelling of bipolar cell somata (Fig. 8c). Similarly, the chloride channel blocker NPPB did not prevent the hypoosmotic swelling (Fig. 8c). However, the blocker of outwardly rectifying potassium channels, tetraethylammonium, prevented the swelling, whereas the blocker of G protein-gated Kir channels, tertiapin-Q (Fig. 8c), as well as the Kir channel blocker barium chloride (Fig. 3) were without effects.

Discussion

  1. Top of page
  2. Abstract
  3. Experimental procedures
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Intense neuronal activity in the retina is associated with a swelling of neuronal cell structures, a decrease in the extracellular space volume, and a decrease in the osmolarity of the extracellular fluid (Dmitriev et al. 1999; Uckermann et al. 2004). It has been shown that Müller glial cells possess endogenous receptor-mediated mechanisms of cell volume regulation which inhibit cellular swelling under hypoosmotic conditions (Uckermann et al. 2006; Wurm et al. 2008, 2010; Krügel et al. 2010). However, it is unclear whether retinal neurons also possess endogenous cell volume-regulatory mechanisms. In this study, we compared the volume regulation of Müller and bipolar cells in response to hypoosmotic and glutamate stimulation. We found that bipolar cell somata, but not Müller cell somata, swell under hypoosmotic conditions (Figs 2a, b and 3) and in the presence of glutamate (Fig. 5a). The data suggest that Müller cells, but not bipolar cells, possess endogenous mechanisms which regulate the cellular volume in response to hypoosmolarity and prolonged glutamate exposure. Under hyperosmotic conditions, both Müller and bipolar cells displayed a decrease in the size of their somata (Fig. 4), suggesting that Müller cells may regulate their volume under hypoosmotic, but not under hyperosmotic conditions. However, we found that the glutamate-induced swelling of bipolar cell somata is inhibited by exogenous adenosine and GABA (Fig. 5b). Thus, inhibitory retinal transmission may regulate the volume of bipolar cells, likely by inhibition of the glutamate action in the cells. It cannot be ruled out that an activation of other receptor types, not investigated here, may also inhibit the hypoosmotic swelling of bipolar cells. However, as the bipolar cell somata clearly swell under hypoosmotic conditions (Figs 2a, b and 3), such putative receptors are probably not endogenously activated under hypoosmotic conditions.

In a second part of the study, we determined the mechanisms of hypoosmolarity- and glutamate-induced swelling of bipolar cell somata. We found that hypoosmotic swelling might be mediated by sodium flux into the cells (Fig. 8a), while the glutamate-induced swelling is mediated by sodium chloride influx (Fig. 6). The hypoosmotic swelling is apparently induced by activation of metabotropic glutamate receptors and sodium-dependent glutamate transporters (Fig. 8a and b), while the glutamate-induced sodium influx is mediated by NMDA- and non-NMDA glutamate receptors, glutamate transporters, and voltage-gated sodium channels (Fig. 6). However, the relative contribution of the different modes of transmembrane ion transport to the hypoosmotic and glutamate-induced swelling remains to be determined in future experiments. In addition, it remains to be determined whether the glutamate-induced swelling of bipolar cell somata is mediated, at least in part, by an indirect mode, for example, by activation of NMDA receptors on amacrine cells.

We found that the potassium channel blocker tetraethylammonium decreased the glutamate-induced swelling of bipolar cell somata (Fig. 6) and abrogated the hypoosmotic swelling of bipolar cell somata (Fig. 8c). Though it cannot be ruled out that an influx of potassium through outwardly rectifying potassium channels might be implicated in mediating the hypoosmotic swelling of bipolar cell somata, it is more likely that inhibition of potassium channels by tetraethylammonium causes a depolarization of the cells which decreases the electrochemical driving force for the sodium influx through electrogenic glutamate transporters, for example.

The hypoosmotic swelling of bipolar cells is mediated by activation of glutamate transporters and metabotropic glutamate receptors (Fig. 8a and b). This suggests that hypoosmolarity induces a release of endogenous glutamate in the retinal tissue. It has been shown that glutamate released from Müller cells induces a swelling of retinal ganglion cell somata (Wurm et al. 2008; Slezak et al. 2012). Because inhibition of the neuronal activity by cadmium and tetrodotoxin did not inhibit the hypoosmotic swelling (Fig. 8c), the release of glutamate is likely mediated by non-vesicular mechanisms. In addition to its function as excitatory neurotransmitter, glutamate plays a role as an osmolyte within the central nervous system. Hypoosmolarity is known to cause an increase in the extracellular glutamate level in the neural tissue (Haskew-Layton et al. 2008). In response to reduced osmolarity, neural cells extrude (in addition to ions) also organic osmolytes such as glutamate through volume-regulated anion channels to obtain cell volume decrease (Pasantes-Morales and Tuz 2006; Verbalis 2010). In addition, a reduction in the glial and neuronal glutamate uptake may contribute to the hypoosmotic increase in extracellular glutamate. A reduction in the glutamate uptake under hypoosmotic conditions was recently described in neuroblastoma cells (Foster et al. 2010).

Regulation of the cellular volume is of great importance to avoid changes in neuronal excitability resulting from a decrease in the extracellular space volume. The fact that hypoosmotic conditions induce a swelling of bipolar but not Müller cell somata (Figs 2a, b and 3) supports the assumption that Müller cells but not bipolar cells possess endogenous cell volume-regulatory mechanisms which are involved in the activity-dependent regulation of the extracellular space volume. In addition to the Kir channel-mediated cell volume regulation, Müller cells possess receptor-mediated mechanisms which prevent hypoosmotic swelling (Uckermann et al. 2006; Wurm et al. 2008, 2010; Krügel et al. 2010). We found that various receptor agonists inhibited the barium-induced hypoosmotic swelling of Müller cell somata but not the hypoosmotic swelling of bipolar cell somata (Fig. 3). The reason for this difference is unclear but may be explained with the assumption that (in contrast to the receptors expressed by Müller cell somata) the receptors expressed by bipolar cell somata are not coupled to cell volume-regulatory intracellular signal transduction pathways. The inhibitory effects of adenosine and GABA on the glutamate-induced swelling of bipolar cell somata (Fig. 5b) may be explained with inhibition of the excitatory action of glutamate (Housley et al. 2009). The lack of osmotic Müller cell volume regulation in the post-ischemic retina (Fig. 3) may enhance the likelihood of detrimental activity-dependent decreases in the extracellular space volume, resulting in neuronal hyperexcitation (Dudek et al. 1990; Chebabo et al. 1995), glutamate toxicity, and neuronal cell swelling which together are major factors contributing to neuronal degeneration in ischemic and inflammatory retinal diseases (Osborne et al. 2004; Bringmann et al. 2005).

Acknowledgements

  1. Top of page
  2. Abstract
  3. Experimental procedures
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This study was supported by grants from the Deutsche Forschungsgemeinschaft (GRK 1097/1; RE 849/10-2; RE 849/12-2). All authors disclose any actual or potential conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Experimental procedures
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Bringmann A., Uckermann O., Pannicke T., Iandiev I., Reichenbach A. and Wiedemann P. (2005) Neuronal versus glial cell swelling in the ischaemic retina. Acta Ophthalmol. Scand. 83, 528538.
  • Bringmann A., Pannicke T., Grosche J., Francke M., Wiedemann P., Skatchkov S. N., Osborne N. N. and Reichenbach A. (2006) Müller cells in the healthy and diseased retina. Prog. Retin. Eye Res. 25, 397424.
  • Chebabo S. R., Hester M. A., Aitken P. G. and Somjen G. G. (1995) Hypotonic exposure enhances synaptic transmission and triggers spreading depression in rat hippocampal tissue slices. Brain Res. 695, 203216.
  • Dmitriev A. V., Govardovskii V. I., Schwahn H. N. and Steinberg R. H. (1999) Light-induced changes of extracellular ions and volume in the isolated chick retina-pigment epithelium preparation. Vis. Neurosci. 16, 11571167.
  • Dudek F. E., Obenhaus A. and Tasker J. G. (1990) Osmolality-induced changes in extracellular volume alter epileptiform bursts independent of chemical synapses in the rat: importance of non-synaptic mechanisms in hippocampal epileptogenesis. Neurosci. Lett. 120, 267270.
  • Foster D. J., Heacock A. M. and Fisher S. K. (2010) Muscarinic receptor stimulation of D-aspartate uptake into human SH-SY5Y neuroblastoma cells is attenuated by hypoosmolarity. J. Pharmacol. Exp. Ther. 333, 297309.
  • Haskew-Layton R. E., Rudkouskaya A., Jin Y., Feustel P. J., Kimelberg H. K. and Mongin A. A. (2008) Two distinct modes of hypoosmotic medium-induced release of excitatory amino acids and taurine in the rat brain in vivo. PLoS ONE 3, e3543.
  • Hirrlinger P. G., Wurm A., Hirrlinger J., Bringmann A. and Reichenbach A. (2008) Osmotic swelling characteristics of glial cells in the murine hippocampus, cerebellum, and retina in situ. J. Neurochem. 105, 14051417.
  • Housley G. D., Bringmann A. and Reichenbach A. (2009) Purinergic signaling in special senses. Trends Neurosci. 32, 128141.
  • Karl A., Wurm A., Pannicke T., Krügel K., Obara-Michlewska M., Wiedemann P., Reichenbach A., Albrecht J. and Bringmann A. (2011) Synergistic action of hypoosmolarity and glutamine in inducing acute swelling of retinal glial (Müller) cells. Glia 59, 256266.
  • Krügel K., Wurm A., Linnertz R., Pannicke T., Wiedemann P., Reichenbach A. and Bringmann A. (2010) Erythropoietin inhibits osmotic swelling of retinal glial cells by Janus kinase and extracellular signal-regulated kinases1/2-mediated release of vascular endothelial growth factor. Neuroscience 165, 11471158.
  • Krügel K., Wurm A., Pannicke P., Hollborn M., Karl A., Wiedemann P., Reichenbach A., Kohen L. and Bringmann A. (2011) Involvement of oxidative stress and mitochondrial dysfunction in the osmotic swelling of retinal glial cells from diabetic rats. Exp. Eye Res. 92, 8793.
  • Osborne N. N., Casson R. J., Wood J. P., Chidlow G., Graham M. and Melena J. (2004) Retinal ischemia: mechanisms of damage and potential therapeutic strategies. Prog. Retin. Eye Res. 23, 91147.
  • Pannicke T., Iandiev I., Uckermann O., Biedermann B., Kutzera F., Wiedemann P., Wolburg H., Reichenbach A. and Bringmann A. (2004) A potassium channel-linked mechanism of glial cell swelling in the postischemic retina. Mol. Cell. Neurosci. 26, 493502.
  • Pannicke T., Iandiev I., Wurm A., Uckermann O. and vom Hagen F., Reichenbach A., Wiedemann P., Hammes H.-P. and Bringmann A. (2006) Diabetes alters osmotic swelling characteristics and membrane conductance of glial cells in rat retina. Diabetes 55, 633639.
  • Pasantes-Morales H. and Tuz K. (2006) Volume changes in neurons: hyperexcitability and neuronal death. Contrib. Nephrol. 152, 221240.
  • Slezak M., Grosche A., Niemiec A. et al. (2012) Relevance of exocytotic glutamate release from retinal glia. Neuron 74, 504516.
  • Uckermann O., Vargová L., Ulbricht E., Klaus C., Weick M., Rillich K., Wiedemann P., Reichenbach A., Syková E. and Bringmann A. (2004) Glutamate-evoked alterations of glial and neuronal cell morphology in the guinea-pig retina. J. Neurosci. 24, 1014910158.
  • Uckermann O., Kutzera F., Wolf A., Pannicke T., Reichenbach A., Wiedemann P., Wolf S. and Bringmann A. (2005) The glucocorticoid triamcinolone acetonide inhibits osmotic swelling of retinal glial cells via stimulation of endogenous adenosine signaling. J. Pharmacol. Exp. Ther. 315, 10361045.
  • Uckermann O., Wolf A., Kutzera F., Kalisch F., Beck-Sickinger A., Wiedemann P., Reichenbach A. and Bringmann A. (2006) Glutamate release by neurons evokes a purinergic inhibitory mechanism of osmotic glial cell swelling in the rat retina: activation by neuropeptide Y. J. Neurosci. Res. 83, 538550.
  • Verbalis J. G. (2010) Brain volume regulation in response to changes in osmolality. Neuroscience 168, 862870.
  • Wurm A., Pannicke T., Iandiev I., Wiedemann P., Reichenbach A. and Bringmann A. (2006a) The developmental expression of K+ channels in retinal glial cells is associated with a decrease of osmotic cell swelling. Glia 54, 411423.
  • Wurm A., Pannicke T., Iandiev I., Bühner E., Pietsch U. C., Reichenbach A., Wiedemann P., Uhlmann S. and Bringmann A. (2006b) Changes in membrane conductance play a pathogenic role in osmotic glial cell swelling in detached retinas. Am. J. Pathol. 169, 19901998.
  • Wurm A., Pannicke T., Wiedemann P., Reichenbach A. and Bringmann A. (2008) Glial cell-derived glutamate mediates autocrine cell volume regulation in the retina: activation by VEGF. J. Neurochem. 104, 386399.
  • Wurm A., Lipp S., Pannicke T., Linnertz R., Färber K., Wiedemann P., Reichenbach A. and Bringmann A. (2009) Involvement of A1 adenosine receptors in osmotic volume regulation of retinal glial cells in mice. Mol. Vis. 15, 18581867.
  • Wurm A., Lipp S., Pannicke T. et al. (2010) Endogenous purinergic signaling is required for osmotic volume regulation of retinal glial cells. J. Neurochem. 112, 12611272.
  • Wurm A., Pannicke T., Iandiev I., Francke M., Hollborn M., Wiedemann P., Reichenbach A., Osborne N. N. and Bringmann A. (2011) Purinergic signaling involved in Müller cell function in the mammalian retina. Prog. Retin. Eye Res. 30, 324342.