Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. 2009 Feb 3; 106(5): 1602–1607.
Published online 2009 Jan 21. doi:  10.1073/pnas.0812708106
PMCID: PMC2629445

Erythropoietin modulation of astrocyte water permeability as a component of neuroprotection


Disturbed brain water homeostasis with swelling of astroglial cells is a common complication in stroke, trauma, and meningitis and is considered to be a major cause of permanent brain damage. Astroglial cells possess the water channel aquaporin 4 (AQP4). Recent studies from our laboratory have shown that glutamate, acting on group I metabotropic glutamate receptors (mGluRs), increases the permeability of astrocyte AQP4, which, in situations of hypoxia-ischemia, will increase astrocyte water uptake. Here we report that erythropoietin (EPO), which in recent years has emerged as a potent neuro-protective agent, antagonizes the effect of a group I mGluR agonist on astrocyte water permeability. Activation of group I mGluRs triggers fast and highly regular intracellular calcium oscillations and we show that EPO interferes with this signaling event by altering the frequency of the oscillations. These effects of EPO are immediate, in contrast to the neuroprotective effects of EPO that are known to depend upon gene activation. Our findings indicate that EPO may directly reduce the risk of astrocyte swelling in stroke and other brain insults. In support of this conclusion we found that EPO reduced the neurological symptoms in a mouse model of primary brain edema known to depend upon AQP4 water transport.

Keywords: aquaporin 4, brain edema, glutamate

Erythropoietin, the major haemopoietic growth factor, has in recent years emerged as one of the most efficient neuro-protective agents (13). It is well documented from both experimental and clinical studies that erythropoietin (EPO) attenuates the degree of brain damage after stroke (46). Stroke is associated with a massive release of glutamate and experimental studies have indicated that EPO also protects from glutamate-triggered neurotoxicity (7). EPO acts by preventing the destruction of viable tissue surrounding the site of an injury, that is, the penumbra that develops during the first 24–48 h after a brain insult (1). Cell swelling is one of the characteristic features of the penumbra (8, 9).

Mounting evidence suggests that water uptake in astrocytes via the water channel aquaporin 4 (AQP4) is an important factor contributing to post-ischemic cell swelling (10, 11). Immunohistochemical studies have indicated a colocalization between EPO receptors and AQP4 in the brain (12, 13), particularly in glial cells at the interface between the brain parenchyma and the periphery. This raises the question whether EPO may modulate astrocyte water permeability and whether this effect may attenuate astrocyte water uptake after a brain insult. We have in a recent study demonstrated that glutamate increases astrocyte water permeability via activation of AQP4 (14). In the present study we have tested the possibility that EPO may counteract the effect of glutamate on astrocyte water permeability.

To document the tissue protective role of EPO in brain edema, we first tested whether EPO reduces symptoms in a mouse model of water intoxication known to depend upon glial AQP4 (11). Observing a robust effect, we then asked whether EPO and glutamate interact with respect to astrocyte water permeability. These studies were performed on astrocytes in primary culture or an astrocyte cell line transfected with AQP4. We showed in a previous study that glutamate modulates astrocyte AQP4 water permeability via activation of group I metabotropic glutamate receptors (mGluRs). This leads to intracellular calcium oscillations, activation of a number of signaling molecules, resulting in phosphorylation of AQP4 and increased permeability of the water channel. Here we have tested the interaction between EPO and glutamate at various steps in this signaling cascade. These experiments show that EPO can completely antagonize glutamate-mediated water flux and indicates a promising role of EPO or EPO analogues in the treatment of conditions associated with disturbances in brain water balance.


EPO Protects Against Neurological Symptoms Caused by Brain Edema.

Female C3H/HEN mice pretreated with drug or saline 24 h before the experiment received an interperitoneal (i.p.) injection of distilled water corresponding to 20% of body weight along with desmopressin (DDAVP; 400 ng/kg). In this model, cellular (cytotoxic) brain edema is produced as a result of the rapid i.p. water infusion in conjunction with the antidiuresis caused by DDAVP, which causes profound hyponatremia. The resulting hypoosmotic state creates an osmotic gradient that favors water entry into the brain, probably through the astrocytic endfeet in contact with blood vessels, without disruption of the bloodbrain-barrier (11, 15). During the acute phase of this brain edema, experimental animals exhibit signs of neurological dysfunction secondary to brain swelling. In both the saline and the EPO groups, neurological symptoms appeared within 45 min after water load and peaked by 75 min in the saline group and by 60 min in the EPO group (Fig. 1A). The total neurological score in the group receiving EPO was significantly less severe when compared with the saline group (Fig. 1B). In contrast, EPO administered 30 min before water loading was not effective (data not shown). This is likely due to the substantial delay of several hours in the transport of peripherally-administered EPO into the brain (5, 16).

Fig. 1.
EPO protects against neurological symptoms caused by brain edema. (A) Female C3H/HEN mice received an i.p. injection of distilled water corresponding to 20% of body weight along with DDAVP (400 ng/kg). Animals were pretreated with vehicle (saline) or ...

EPO Effect on Water Permeability in Astrocytes.

We first tested whether EPO given alone had any effect on astrocyte water permeability. An astrocyte cell line transfected with GFP-tagged AQP4 was used for these studies. The use of this cell line, which does not express endogenous AQP4, allows us to discriminate between AQP4 dependent and independent effects on water permeability.

Transfection efficacy, judged from the number of GFP-positive cells, was in the range of 10%. Recordings were made on GFP-positive cells (AQP4-positive cells) and on neighboring GFP-negative cells (AQP4-negative cells).

The cells were loaded with the inert fluorescent dye calcein and incubated with vehicle or EPO in an isoosmotic medium before exposure to hypoosmotic solution. Water permeability was measured as the time constant of the exponential decay of the calcein signal as described in Methods. Under control conditions astrocytes that expressed AQP4 had 2- to 4- fold higher water permeability (Pf) than AQP4-negative cells.

Exposure to EPO 10−7 g/ml caused an almost immediate, albeit transient increase of water permeability in AQP4-positive cells. This effect had vanished after 5 min (Fig. 2A). A similar transient increase in water permeability was observed after incubation with EPO 10−8 and 10−9 g/ml (data not shown). In no case did EPO have any effect on water permeability when measured at 5, 7, 10, 15, and 30 min after incubation. At the time points that the effect of EPO on the mGluR-induced increase in AQP4 water permeability were measured in cell cultures and in acute slices (indicated by arrows in Fig. 2A), there was no effect of EPO itself. The mechanism for the transient effect of EPO on water permeability remains unclear. EPO had no effect on water permeability of AQP4-negative cells.

Fig. 2.
EPO abolishes group I mGluR-induced increase in water permeability in AQP4-expressing astrocytes. (A) Water permeability (Pf) in AQP4-positive cells after exposure to EPO (10−7 g/ml) for 2, 5, and 15 min, respectively. Exposure to EPO caused a ...

EPO Abolishes Group I mGluR-Induced Increase in Water Permeability.

We recently reported that glutamate increases AQP4 water permeability in astrocytes via activation of group I metabotropic glutamate receptors (14). To address the question whether EPO may affect the mGluR-triggered increase in astrocyte water permeability, cells were preincubated with EPO 10−8 g/ml or vehicle for 10 min before they were exposed to DHPG.

In cells preincubated with vehicle, DHPG (20 μM) caused a significant (P < 0.001) increase in water permeability in AQP4-positive cells (Pf 8.9 ± 0.4 μm/s in vehicle-exposed cells; n = 56 and 12.0 ± 0.6 μm/s in cells exposed to DHPG; n = 28) (Fig. 2B). Pretreatment of cells with EPO abolished (P < 0.001) the DHPG-induced increase in water permeability in the AQP4-positive cells (Pf 9.2 ± 0.3 μm/s in cells pretreated with EPO before DHPG; n = 46, and 12.0 ± 0.6 μm/s in cells exposed to DHPG alone; n = 28) (Fig. 2B).

Neither DHPG alone nor pretreatment with EPO before DHPG had any effect on water permeability in AQP4-negative cells (Pf 4.3 ± 0.2 μm/s, 4.1 ± 0.2 μm/s and 3.9 ± 0.2; n = 23–37, in vehicle-, DHPG- and EPO/DHPG-exposed cells, respectively) (Fig. 2C).

The GFP signal was evenly distributed in the plasma membrane in the AQP4-positive cells. The signal from the cytoplasm was generally very low. Exposure to DHPG alone or to EPO and DHPG did not have any visible effect on the subcellular distribution of AQP4.

Studies were also performed on astroglial in cells in primary culture. These cells express endogenous AQP4 (14). DHPG (100 μM) significantly (P < 0.001) increased water permeability in the astrocytes in primary culture (water permeability (a.u.) 0.030 ± 0.0007 in vehicle-treated cells; n = 85, and 0.037 ± 0.0009 in cells exposed to DHPG; n = 53) (Fig. 3A). When the cells were preincubated with EPO (10−7 g/ml) for 10 minutes before exposure to DHPG, the DHPG-induced increase in water permeability was abolished (P = 0.001) (water permeability; a.u, 0.032 ± 0.0009 in cells pretreated with EPO before DHPG; n = 73, compared with 0.037 ± 0.0009 in DHPG-treated cells; n = 53) (Fig. 3A).

Fig. 3.
EPO abolishes group I mGluR-induced increase in water permeability in astroglial cells in primary culture and inhibits group I mGluR-induced increase in hypotonic swelling rate in acute hippocampal slices. (A) Rat astroglial cells in primary culture, ...

EPO Inhibits Group I mGluR-Induced Increase in Hypotonic Swelling Rate in Acute Hippocampal Slices.

To achieve a more physiological model, studies were also performed on rat brain slices. We used a technique developed in our laboratory, where the rate of slice swelling is recorded after exposure to a hypotonic solution. This challenge will primarily increase the water uptake in astrocytes, whereas the increase of water uptake in neurons and microglia cells is much less pronounced (14). Here, the studies were performed on acute rat hippocampal slices. The slices were incubated with vehicle, the group I mGluR agonist DHPG (100 μM), or DHPG together with EPO (10 ng/ml) for 7 min. Exposure to DHPG significantly increased the rate of swelling of the slices in response to the hypotonicity (Fig. 3B). EPO abolished the DHPG-induced increase of the swelling rate in the slices (Fig. 3B) (P < 0.05).

EPO Modify mGluR Calcium Signaling.

The group I metabotropic glutamate receptors mGluR1 and mGluR5 are coupled to phospholipase C and will, upon activation, trigger the release of calcium from intracellular stores resulting in intracellular calcium oscillations (17, 18). Here we show that preincubation with EPO perturbs the DHPG-triggered calcium signal. We recorded intracellular calcium levels ([Ca2+]i) in astroglial cells in primary culture exposed to DHPG and to cells preincubated with EPO before addition of DHPG.

In the astroglial cells DHPG induced a transient peak in [Ca2+]i followed by periodic oscillations with a periodicity of approximately 15 seconds (representative recordings shown in Fig. 4A). This response was observed in approximately 31% of all cells (n = 199) and corresponds to a frequency of 65 mHz. Two percent of the cells did not respond to DHPG. The oscillatory response to DHPG resulted in more than 3 [Ca2+]i peaks in 40% of the cells. The mean number of peaks was 8. The oscillatory response was also observed when a calcium-free buffer was used, indicating that the rise in [Ca2+]i was a result of release of Ca2+ from intracellular stores. Pretreatment of the cells with EPO (10−7 g/ml) for 10 min before exposure to DHPG resulted in an altered oscillatory response in [Ca2+]i to DHPG (representative recordings shown in Fig. 4B). Approximately 18% of the cells still responded with periodic changes in [Ca2+]i, but the oscillations had a less regular pattern and the synchronicity of the response disappeared. After pretreatment with EPO 10% of the cells (n = 217) did not respond to DHPG. Approximately 30% of the cells responded with 3 or more [Ca2+]i peaks. The mean number of peaks was 6. The difference in number of [Ca2+]i -peaks between the DHPG and EPO/DHPG groups was significant (P < 0.01). EPO alone did not induce any calcium response in the astroglial cells in primary culture (as seen in Fig. 4B before addition of DHPG).

Fig. 4.
EPO reduces mGluR-induced calcium oscillations. (A) Representative recording of intracellular Ca2+ ([Ca2+]i) after DHPG in astroglial cells in primary culture loaded with Fura-2/AM. Large arrow indicates addition of DHPG. Small arrows indicate periodic ...

We performed a frequency analysis of the oscillatory calcium signals and calculated a power spectrum from a typical cell exposed to DHPG only and in a cell after pretreatment with EPO, supplied as see SI.


EPO administration provides neural protection in animal models of brain ischemia and trauma (4, 5, 19), conditions that are associated with brain edema. EPO has been shown to protect from apoptosis (1, 20, 21), but, aside from this, the mechanisms by which EPO protects from brain damage are largely unknown. The results from this study show that EPO is involved in the regulation of brain water homeostasis and suggest that EPO can act to decrease the susceptibility to brain edema.

Brain edema can be vasogenic and/or cellular in origin. The edema that develops after ischemia and trauma is primarily of cellular origin (2224). Our studies on intact animals showed that EPO has a neuroprotective effect in an acute model of water intoxication, that is, a primary cause of cellular brain edema. In cellular edema the astrocytes take up inappropriate amounts of extracellular fluid. Several lines of evidence suggest that water influx via AQP4 contributes to this inappropriate water uptake (10, 11). For example, mice with AQP4 knocked out are resistant to water intoxication similar to our observations.

Glutamate, which is released in excess in most types of brain insults, increases AQP4-dependent water permeability in astrocytes (14). The glutamate effect is mediated by activation of group I metabotropic glutamate receptors and intracellular calcium release. Here we found that preexposure to EPO abolished the effect of activation of group I metabotropic glutamate receptors on AQP4 water permeability in cultured astrocytes. Preexposure to EPO also abolished the increased rate of hypo-osmotic brain tissue swelling triggered by activation of group I metabotropic glutamate receptors. Although EPO has been reported to directly inhibit glutamate release in cultured neurons (25) and could potentially do so in the hippocampal slice model, the cultured glial cell experiments show that EPO directly inhibits glutamate-mediated water permeability.

The effects of activation of group I metabotropic glutamate receptors by DHPG were abolished if cells and tissue had been preexposed to EPO. This antagonizing effect was associated with perturbation of the oscillatory calcium signal triggered by mGluR5 activation. EPO exerts its biological functions through a cell surface receptor, which has been known to be expressed in the cerebral cortex, midbrain, and hippocampus of the CNS (1, 3, 26). EPO is expressed both in neurons and astrocytes and recent studies demonstrated that EPO receptor (EPOR) is expressed in cultured astrocytes (27). The EPO receptor belongs to a subfamily of the type I cytokine receptor superfamily, which also includes the β common receptor (βcR). EPOR and βcR can form a low-affinity, heteromeric EPOR−βcR receptor, and it is this heteromer that is responsible for the neuroprotective effects of EPO (1, 26). The binding of EPO to this receptor has been reported to activate several signaling pathways, including the Jak2/Stat5 and NFκ-B pathways (28), resulting in transcription of new proteins. Short-term effects of EPO activation are less well known, but in heart tissue EPO has been shown to activate protein kinase C epsilon (PKCε) within minutes (29), which contributes one component of the tissue protective effect of EPO in heart tissue. A number of studies have also provided evidence that PKC affects intracellular calcium levels in a number of cell types and does so via activity of specific calcium channels. A role for PKCε and its effects on calcium signaling in brain tissue remains to be elucidated. Future studies may reveal whether the interaction between EPO and mGluRs also can explain the molecular mechanisms of the long term effects of EPO.

It is well documented that EPO can protect from brain damage caused by ischemia and hypoxia (1, 4, 6, 30). Neurons directly exposed to hypoxia-ischemia are rapidly committed to programmed cell death. This immediate effect is followed by the development of a slowly increasing zone of swollen cells, the penumbra (9, 31). Astroglial cells appear to be more susceptible to swelling than neurons and microglial cells (23, 32). The swelling of the astroglial cells will, because of the restricted space within the skull, have an adverse effect on surrounding neurons, and many of the neurons within the penumbra zone will, without intervention, undergo apoptosis or necrotic cell death. EPO acts to protect from these effects via its anti-apoptotic effect (21, 30, 33) and, as indicated from the results in the current study, by preventing cell swelling. Because neuronal death is preceded by an outflow of glutamate, mGluR5, the predominant glutamate receptor in astrocytes, will be activated. This in turn will lead to increased water permeability of astrocytes. This effect can, as shown in the present study, be antagonized by EPO. The notion that EPO also exert its neuroprotective effect by reducing cell swelling is supported by the observation that EPO does attenuate the spread of the penumbra. The effect on cell swelling occurred within minutes and should be considered as a short term effect of EPO, as opposed to long term effects that require gene transcription and expression changes.

A prime goal of neuroprotection in stroke/ischemic lesion has been salvage of the ischemic penumbra. The cells in this region are at risk due to several features of the penumbra such as cell swelling, effects of increased glutamate levels and elevations of K+ (8, 9, 23, 34). We propose that EPO may inhibit astrocyte swelling in the penumbra through an effect on AQP4 water permeability, thereby preserving astrocyte function and increase the potential of tissue salvation. In conclusion, the results of this study indicate that EPO or EPO derivatives may have an important therapeutic effect during the development of the ischemic penumbra, and that EPO may have a therapeutic potential in preventing cellular brain edema after other brain insults.



In the present study, we used recombinant human erythropoietin, referred to as EPO. EPO was purchased from Sigma. (S)-3,5 -Dihydroxyphenylglycine (DHPG) is an agonist for group I metabotropic glutamate receptors (mGluR1 and mGluR5) (Sigma-Aldrich).

Water Intoxication.

Female C3H/HEN mice (n = 10 in each group) received an i.p. (i.p.) injection of distilled water corresponding to 20% of body weight along with 1-Deamino (8-d-arginine) vasopressin (DDAVP) (400 ng/kg). Animals were pretreated with vehicle (saline) or EPO (50 μg/kg BW i.p) 24 h before the water load. Neurological symptoms were evaluated by a scale modified from Manley et al. (11) for every 15 min up to 120 min, and at 150 and 180 min, after water load. The assessment variables included the following parameters: exploring of cage, visually tracking of objects, whisker movement, leg-tail movements, pain withdrawal, coordination of movement, and stop at edge of table. A score of zero was normal and the maximum score of neurological symptoms was 8. Scores above 1 were considered to be pathological. The water intoxication experiments were approved by the local Institutional Animal Use and Care Committee and followed the Guide for the Care and Use of Laboratory Animals, U.S. National Research Council.

Cell Cultures.

To obtain astroglial cells in primary mixed culture, rat hippocampi were dissected from pups at embryonic age of 18 days, placed in 0.25% trypsin (Gibco Invitrogen), and incubated for 7 min. at 37 °C. The cells were subsequently triturated with 1% DNase and seeded onto 40 mm coverslips precoated with polyD-lysine (1 mg/100 mL; Sigma). The cells were grown in 70% DMEM/30% Neurobasal medium supplemented with 10% FBS (Invitrogen) and 1% penicillin/streptomycin/amphotericin (Sigma). After 24 h, the medium was replaced to 70% DMEM/30% neurobasal medium with 5% FBS and B27 and plated at a concentration of 1–1.5 million cells per coverslip. Cultures were maintained in an incubator with 5% CO2 at 37 °C and used after 8–10 days, at the time of which the cells were confluent. It was shown with RT-PCR that these cells express endogenous AQP4 mRNA when cultured for 4–12 days (data not shown).

A rat astrocyte cell line (CTX TNA2, European Collection of Cell Cultures, Centre for Applied Microbiology and Research, Salisbury, Wiltshire, U.K.) was used in subpassages 3–7. It was shown with RT-PCR that the astrocyte cell line does not express endogenous AQP4 (data not shown). The astrocytes were grown on 40-mm coverslips (Bioptechs) in Dulbecco's Modified Eagle's medium (DMEM, Sigma Aldrich) containing 0.5 units/ml penicillin and 50 μg/ml streptomycin supplemented with 10% FBS (FBS), 0.11 mg/ml sodium pyruvate and 2 mM l-glutamine. The cells were transfected on the second day of culture with cDNA constructs encoding mouse AQP4 tagged with green fluorescent protein (GFP) on the NH2 terminus (35, 36) using TransFast Transfection reagent (Promega) according to the manufacturer's protocol. The experiments were performed on the 4th day of culture when the cells were subconfluent.

DNA Constructs.

We used constructs encoding mouse AQP4.M23 tagged with green fluorescent protein (GFP) at the NH2 terminus (pGFP-AQP4.M23), described previously (35, 36).

Water Permeability Measurements.

Water permeability (Pf) was measured using a method that we described in detail previously (35, 37). By use of this method Pf in individual cells within cell monolayers can be determined and compared between cells that do or do not express GFP-labeled proteins. Briefly, coverslips with the transfected astrocytes or primary cultured astroglial cells were mounted in a closed perfusion chamber (Focht Live Cell Chamber System) on the stage of a Zeiss 410 inverted laser scanning microscope. At the beginning of each study, an image showing the distribution of GFP-tagged proteins in the transfected cell line was recorded. Cells were loaded with the fluorescent dye calcein AM (Molecular Probes), which is inert to changes in intracellular pH and calcium. Calcein loaded cells were identified as GFP-AQP4-positive (AQP4-positive cells) or GFP-AQP4-negative (AQP4-negative cells) by superimposing the two images. The loading solution was changed to 300 mOsm PBS containing EPO or vehicle and the cells were incubated for 10 min followed by 2-min exposure to vehicle, DHPG, or DHPG together with EPO. To measure Pf, cells were initially perfused with isoosmotic PBS followed by a switch to a hypoosmotic PBS (200 mOsm). The part of the obtained curves recorded immediately after the solution switch was used for Pf calculation as previously described (35, 37). The initial region of the fluorescence curve was fitted with a single exponential function and its time constant was used as a measure of the rate of cell swelling. For the astroglial cells in primary culture, the time constant of the initial swelling curve was used directly as a measure of water permeability and is here reported as arbitrary units (a.u.).

Hippocampal Slice Swelling.

Acute hippocampal slice swelling was measured using a method that we described in ref. 14. Sprague–Dawley rats (20 days old, B & K Universal) were killed by decapitation, the brain was quickly removed and dipped in cold artificial cerebrospinal fluid (aCSF) oxygenated with 95% O2/5% CO2. Transverse hippocampal slices of 300 μm thickness were cut using Vibratome 3000 sectioning system in cold oxygenated aCSF and incubated for 2 h at RT to allow recovery before experiments. The slices were then loaded with calcein by incubation with 10 μM calcein-AM in aCSF for 15 min at RT, placed in the closed chamber and mounted on the stage of Zeiss 410 inverted confocal laser scanning microscope with a 5x/0.25 objective. All further procedures were performed at 25 °C in oxygenated aCSF. The slices were first incubated with drugs or vehicle for 6 min and then perfused with aCSF containing the same agents. After 1 minute of the perfusion, the solution was switched to a hypotonic solution obtained by decreasing NaCl concentration to 110 mM in the aCSF. Images of the slices were recorded each 3 s, with excitation by a 488-nm argon laser and emission collected using a bandpass filter at 515–525 nm. Obtained images were analyzed using ImageJ software (Rasband, W.S. ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, 1997–2007) to calculate the slice surface area. The initial part of the slice surface area curve (≈10 s after application of hypoosmotic solution) was fitted by a single exponential function. The time constant of this function was used as a measure of the slice swelling rate.

Calcium Measurements.

Primary cultured astroglial cells were incubated with 3 μM Fura-2/AM (Molecular Probes) for 20 min in PBS at 37 °C for intracellular Ca2+ ([Ca2+]i)measurements. Single cell ratiometric imaging was performed by using an upright microscope (Axioskop 2 FS, Zeiss) with a 40 × 0.8 NA water dipping lens with a cooled CCD camera (ORCA-ERG, Hamamatsu). Cytosolic free Ca2+ changes was calculated from the emission ratio detected with a bandpass filter at 510/30 nm using excitation at 340/10 nm and 380/10 nm performed with a monochromator (Polychrome IV, TILL Photonics). All devices were controlled and data were analyzed with computer software (MetaFluor, Molecular Devices). To measure calcium dynamics, cells were excited with a frequency of 1.0 Hz, which corresponds to a Nyqvist frequency of 0.5 Hz. [Ca2+]i were measured after application of DHPG (100 μM) or after preincubation with EPO (10−7 g/ml) for 10 min before application of DHPG.

DHPG-induced intracellular calcium oscillations were analyzed in cells pretreated with EPO and compared with those from cells exposed to DHPG only. We recorded the pattern and synchronicity of the responses, the proportion of cells not responding to DHPG, the proportion of cells responding to DHPG with oscillatory changes in [Ca2+]i (definitions in the SI), the proportion of cells responding with more than 3 Ca2+-peaks in each group and the total number of Ca2+-peaks, respectively. We also performed a Power Spectrum Analysis, provided as SI.


Data are presented as means ± SEM. Statistical analysis was performed using Student's t test or, when appropriate, ANOVA followed by Student-Newman-Keuls test for pairwise comparisons (MedCalc Software). A difference was considered statistically significant when P < 0.05.

Supplementary Material

Supporting Information:


This work was supported by the Swedish Research Council (A.A., H.B., U.A.), The Nordic Centre of Excellence Program in Molecular Medicine, The Persson Family Foundation, HKH Kronprinsessan Lovisas förening för barnsjukvård/Stiftelsen Axel Tielmans Minnesfond.


Conflict of interest statement: M.B. and A.C. are employees of Warren Pharmaceuticals, which is developing erythropoietin analogues and tissue-protective compounds for potential clinical uses.

This article contains supporting information online at www.pnas.org/cgi/content/full/0812708106/DCSupplemental.


1. Brines M, Cerami A. Emerging biological roles for erythropoietin in the nervous system. Nat Rev Neurosci. 2005;6:484–494. [PubMed]
2. Brines M, Cerami A. Discovering erythropoietin's extrahematopoietic functions: Biology and clinical promise. Kidney Int. 2006;70:246–250. [PubMed]
3. Genc S, Koroglu TF, Genc K. Erythropoietin and the nervous system. Brain Res. 2004;1000:19–31. [PubMed]
4. Ehrenreich H, et al. Erythropoietin therapy for acute stroke is both safe and beneficial. Mol Med. 2002;8:495–505. [PMC free article] [PubMed]
5. Brines ML, et al. Erythropoietin crosses the blood-brain barrier to protect against experimental brain injury. Proc Natl Acad Sci USA. 2000;97:10526–10531. [PMC free article] [PubMed]
6. Liu R, Suzuki A, Guo Z, Mizuno Y, Urabe T. Intrinsic and extrinsic erythropoietin enhances neuroprotection against ischemia and reperfusion injury in vitro. J Neurochem. 2006;96:1101–1110. [PubMed]
7. Yamasaki M, et al. Neuroprotective effects of erythropoietin on glutamate and nitric oxide toxicity in primary cultured retinal ganglion cells. Brain Res. 2005;1050:15–26. [PubMed]
8. Di X, Alves OL, Bullock R. Cytotoxic edema is independent of NMDA ion channel activation following middle cerebral artery occlusion (MCAO). An in vivo autoradiographic and MRI study. Neurol Res. 2003;25:329–334. [PubMed]
9. Weinstein PR, Hong S, Sharp FR. Molecular identification of the ischemic penumbra. Stroke. 2004;35:2666–2670. [PubMed]
10. Amiry-Moghaddam M, et al. An alpha-syntrophin-dependent pool of AQP4 in astroglial end-feet confers bidirectional water flow between blood and brain. Proc Natl Acad Sci USA. 2003;100:2106–2111. [PMC free article] [PubMed]
11. Manley GT, et al. Aquaporin-4 deletion in mice reduces brain edema after acute water intoxication and ischemic stroke. Nat Med. 2000;6:159–163. [PubMed]
12. Nielsen S, et al. Specialized membrane domains for water transport in glial cells: high-resolution ImmunoGold cytochemistry of aquaporin-4 in rat brain. J Neurosci. 1997;17:171–180. [PubMed]
13. Eid T, et al. Increased expression of erythropoietin receptor on blood vessels in the human epileptogenic hippocampus with sclerosis. J Neuropathol Exp Neurol. 2004;63:73–83. [PubMed]
14. Gunnarson E, et al. Identification of a molecular target for glutamate regulation of astrocyte water permeability. Glia. 2008;56:587–596. [PubMed]
15. Gullans SR, Verbalis JG. Control of brain volume during hyperosmolar and hypoosmolar conditions. Annu Rev Med. 1993;44:289–301. [PubMed]
16. Xenocostas A, et al. The pharmacokinetics of erythropoietin in the cerebrospinal fluid after intravenous administration of recombinant human erythropoietin. Eur J Clin Pharmacol. 2005;61:189–195. [PubMed]
17. Nakanishi S, Masu M. Molecular diversity and functions of glutamate receptors. Annu Rev Biophys Biomol Struct. 1994;23:319–348. [PubMed]
18. Zur Nieden R, Deitmer JW. The role of metabotropic glutamate receptors for the generation of calcium oscillations in rat hippocampal astrocytes in situ. Cereb Cortex. 2006;16:676–687. [PubMed]
19. Wang X, et al. The nonerythropoietic asialoerythropoietin protects against neonatal hypoxia-ischemia as potently as erythropoietin. J Neurochem. 2004;91:900–910. [PubMed]
20. Ghezzi P, Brines M. Erythropoietin as an antiapoptotic, tissue-protective cytokine. Cell Death Differ. 2004;11(suppl 1):S37–S44. [PubMed]
21. Villa P, et al. Erythropoietin selectively attenuates cytokine production and inflammation in cerebral ischemia by targeting neuronal apoptosis. J Exp Med. 2003;198:971–975. [PMC free article] [PubMed]
22. Simard M, Nedergaard M. The neurobiology of glia in the context of water and ion homeostasis. Neuroscience. 2004;129:877–896. [PubMed]
23. Kimelberg HK. Astrocytic swelling in cerebral ischemia as a possible cause of injury and target for therapy. Glia. 2005;50:389–397. [PubMed]
24. Amiry-Moghaddam M, Ottersen OP. The molecular basis of water transport in the brain. Nat Rev Neurosci. 2003;4:991–1001. [PubMed]
25. Kawakami M, Sekiguchi M, Sato K, Kozaki S, Takahashi M. Erythropoietin receptor-mediated inhibition of exocytotic glutamate release confers neuroprotection during chemical ischemia. J Biol Chem. 2001;276:39469–39475. [PubMed]
26. Brines M, et al. Erythropoietin mediates tissue protection through an erythropoietin and common beta-subunit heteroreceptor. Proc Natl Acad Sci USA. 2004;101:14907–14912. [PMC free article] [PubMed]
27. Lee SM, et al. EPO receptor-mediated ERK kinase and NF-κB activation in erythropoietin-promoted differentiation of astrocytes. Biochem Biophys Res Commun. 2004;320:1087. [PubMed]
28. Wojchowski DM, Gregory RC, Miller CP, Pandit AK, Pircher TJ. Signal transduction in the erythropoietin receptor system. Exp Cell Res. 1999;253:143–156. [PubMed]
29. Hanlon PR, et al. Mechanisms of erythropoietin-mediated cardioprotection during ischemia-reperfusion injury: role of protein kinase C and phosphatidylinositol 3-kinase signaling. FASEB J. 2005 04–3545fje. [PubMed]
30. Villa P, et al. Reduced functional deficits, neuroinflammation, and secondary tissue damage after treatment of stroke by nonerythropoietic erythropoietin derivatives. J Cereb Blood Flow Metab. 2007;27:552–563. [PubMed]
31. Dirnagl U, Iadecola C, Moskowitz MA. Pathobiology of ischaemic stroke: an integrated view. Trends Neurosci. 1999;22:391–397. [PubMed]
32. Kimelberg HK. Current concepts of brain edema. Review of laboratory investigations. J Neurosurg. 1995;83:1051–1059. [PubMed]
33. Siren AL, et al. Erythropoietin prevents neuronal apoptosis after cerebral ischemia and metabolic stress. Proc Natl Acad Sci USA. 2001;98:4044–4049. [PMC free article] [PubMed]
34. Feustel PJ, Jin Y, Kimelberg HK. Volume-regulated anion channels are the predominant contributors to release of excitatory amino acids in the ischemic cortical penumbra. Stroke. 2004;35:1164–1168. [PubMed]
35. Zelenina M, Zelenin S, Bondar AA, Brismar H, Aperia A. Water permeability of aquaporin-4 is decreased by protein kinase C and dopamine. Am J Physiol Renal Physiol. 2002;283:F309–318. [PubMed]
36. Gunnarson E, et al. Lead induces increased water permeability in astrocytes expressing aquaporin 4. Neuroscience. 2005;136:105–114. [PubMed]
37. Zelenina M, Brismar H. Osmotic water permeability measurements using confocal laser scanning microscopy. Eur Biophys J. 2000;29:165–171. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...


  • Compound
    PubChem chemical compound records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records. Multiple substance records may contribute to the PubChem compound record.
  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem chemical substance records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records.

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...