Neurochemistry International 58 (2011) 414–422

Contents lists available at ScienceDirect

Neurochemistry International

journa l homepage: www.e lsev ier .com/ locate /neuint

Release of ATP from avian Muller glia cells in culture

Erick Correia Loiola, Ana Lucia Marques Ventura *

Department of Neurobiology, Neuroscience Program, Institute of Biology, Federal Fluminense University, Niteroi, RJ, Brazil

A R T I C L E I N F O

Article history:

Received 22 September 2010

Received in revised form 17 December 2010

Accepted 20 December 2010

Available online 28 December 2010

Keywords:

ATP release

Glutamate receptors

Muller glia

Avian retina

A B S T R A C T

ATP can be released from neurons and act as a neuromodulator in the nervous system. Besides neurons,

cortical astrocytes also are capable of releasing ATP from acidic vesicles in a Ca2+-dependent way. In the

present work, we investigated the release of ATP from Muller glia cells of the chick embryo retina by

examining quinacrine staining and by measuring the extracellular levels of ATP in purified Muller glia

cultures. Our data revealed that glial cells could be labeled with quinacrine, a reaction that was

prevented by incubation of the cells with 1 mM bafilomycin A1 or 2 mM Evans blue, potent inhibitors of

vacuolar ATPases and of the vesicular nucleotide transporter, respectively. Either 50 mM KCl or 1 mM

glutamate was able to decrease quinacrine staining of the cells, as well as to increase the levels of ATP in

the extracellular medium by 77% and 89.5%, respectively, after a 5 min incubation of the cells.

Glutamate-induced rise in extracellular ATP could be mimicked by 100 mM kainate (81.5%) but not by

100 mM NMDA in medium without MgCl2 but with 2 mM glycine. However, both glutamate- and

kainate-induced increase in extracellular ATP levels were blocked by 50 mM of the glutamatergic

antagonists DNQX and MK-801, suggesting the involvement of both NMDA and non-NMDA receptors.

Extracellular ATP accumulation induced by glutamate was also blocked by incubation of the cells with

30 mM BAPTA-AM or 1 mM bafilomycin A1. These results suggest that glutamate, through activation of

both NMDA and non-NMDA receptors, induces the release of ATP from retinal Muller cells through a

calcium-dependent exocytotic mechanism.

� 2010 Elsevier Ltd. Open access under the Elsevier OA license.

1. Introduction

In the CNS, ATP mediates a broad range of effects, varying fromtrophic to toxic effects, both in neurons and glial cells (for review,see Franke and Illes, 2006; Verkhratsky et al., 2009). In the retina, itis also emerging as an important signaling molecule that can bereleased, through a calcium-dependent mechanism, by applicationof several depolarizing stimuli such as light, KCl and glutamateagonists (Newman, 2005; Perez et al., 1986; Santos et al., 1999).ATP can also be released from pigmented epithelium (PE) by theopening of connexin 43 hemichannels (Pearson et al., 2005) orNMDA receptor stimulation (Reigada et al., 2006). Recently, therelease of ATP in the retina or in cultures of retinal cells wasobserved in pathological conditions such as high glucose (Costaet al., 2009) or elevated intraocular pressure (Resta et al., 2007).

The expression of several nucleotide receptor subtypes wasdescribed in the retina. Besides mRNAs for several P2X and P2Yreceptors (Fries et al., 2004a,b; Greenwood et al., 1997; Jabs et al.,2000; Wheeler-Schilling et al., 2000, 2001), several receptor

* Corresponding author at: Department of Neurobiology, Universidade Federal

Fluminense, Cx. Postal 100180, Niteroi, RJ 24001-970, Brazil. Tel.: +55 21 26292274;

fax: +55 21 26292268.

E-mail address: almv@vm.uff.br (A.L.M. Ventura).

0197-0186/� 2010 Elsevier Ltd. Open access under the Elsevier OA license.

doi:10.1016/j.neuint.2010.12.019

proteins, including both P2Y and P2X sub-types of receptors, werecharacterized in this tissue (for review, see Housley et al., 2009).

During development, nucleotide-mediated responses wereprimarily associated with the induction of cell proliferation inthe retina (Milenkovic et al., 2003; Moll et al., 2002; Pearson et al.,2002; Sanches et al., 2002; Sugioka et al., 1999). In the chick retina,while activation of P2Y2/4 receptors by ATP or UTP induces theproliferation of early developing progenitors that will generateganglion, amacrine, horizontal cells and photoreceptors (Pearsonet al., 2002, 2005), activation of P2Y1 receptors by ATP or ADPinduces the proliferation of late developing glial/bipolar progeni-tors (Franca et al., 2007; Sanches et al., 2002) by a mechanisminvolving PKC, MAPK and PI3K/AKT pathways (Nunes et al., 2007;Ornelas and Ventura, 2010; Sholl-Franco et al., 2010). In thedeveloping rat retina, ATP signaling was also associated with theinduction of cell death through the activation of P2X7 receptors(Resta et al., 2005).

The Muller cell is the predominant glial cell type that interactswith the majority of neurons in the retina (for review, Sarthy andRipps, 2001). Muller cells have a supportive function for retinalneurons, responding to and releasing a variety of signalingmolecules during development as well as in the adult tissue (Reiset al., 2008, for review). Muller cells, for example, are involved in thecontrol of the extracellular levels of K+, H+ and neurotransmitters, inthe release of vasoactive agents and D-serine, in light conduction to

E.C. Loiola, A.L.M. Ventura / Neurochemistry International 58 (2011) 414–422 415

photoreceptors, in inhibition of cell swelling under hypotonicconditions, among other functions (Bringmann et al., 2006).

Some of the above functions of the retinal glia involve activationof nucleotide receptors primarily associated with the mobilizationof intracellular calcium levels (Li et al., 2001). It was demonstrated,for example, that light or mechanical stimulation of the retinainduces Ca2+ waves that propagate from Muller cell to Muller cellby the release of ATP and activation of P2 receptors (Newman,2001, 2003). The release of ATP during calcium transients inhibitslight-induced ganglion cell spiking, a response that is blocked byadenosine antagonists or inhibitors of ectonucleotidases (New-man, 2004; Newman and Zahs, 1997). ATP can induce a P2Y1-mediated release of adenosine from Muller cells that inhibits theirswelling in tissues submitted to hypotonic conditions (Uckermannet al., 2006). Activation of P2Y1 receptors is also involved in Mullercell gliosis after ouabain-induced cell injury in the fish retina(Battista et al., 2009).

Although ATP is released from Muller cells when calciumtransients are induced in the retina (Newman, 2001), themechanism by which these cells release this nucleotide is poorlyunderstood. In the present study, we investigated the release ofATP from Muller glia cells of the chick embryo retina by examiningquinacrine staining and by measuring the extracellular levels ofATP in purified Muller glia cultures. Our data revealed that glialcells could be labeled with quinacrine, a reaction that wasprevented by incubation of the cells with bafilomycin A1, a potentinhibitor of vacuolar ATPases. Moreover, 50 mM KCl, glutamateand kainate were able to decrease quinacrine staining in the cellsas well as to increase the extracellular levels of ATP in the medium.Glutamate-induced rise in extracellular ATP was completely

[()TD$FIG]

Fig. 1. Quinacrine staining of living purified cultures of Muller glia. Glia enriched culture

fluorescence microscopy. (A) Photomicrograph showing quinacrine fluorescent granules

under phase contrast illumination. (C) Photomicrograph showing glial cultures pre-treate

Phase contrast image of cells shown in (C). Bar = 10 mm.

blocked by the glutamatergic antagonists DNQX and MK-801, aswell as by BAPTA-AM and bafilomycin A1, suggesting thatglutamate, through activation of NMDA and non-NMDA receptors,induces the release of ATP from retinal Muller cells through acalcium-dependent exocytotic mechanism.

2. Materials and methods

2.1. Reagents

Glutamine, penicillin-G, streptomycin sulfate, glutamate, kainate, 6,7-dinitro-

quinoxaline-2,3-dione (DNQX), dizocilpine maleate (MK-801), BAPTA-AM, EGTA,

quinacrine, Evans blue were from Sigma (St. Louis, MO, USA); ATP determination

kit, MEM, Fetal Bovine Serum, Life Technologies Inc.; trypsin, Worthington

Biochemical (Freehold, NJ, USA); all other reagents were of analytical grade.

2.2. Glial cell cultures

The use of animals was in accordance with the ‘‘NIH guide for the Care and Use of

Laboratory Animals’’ and approved by the department’s commission for animal

care. Glial cultures were obtained according to a previous published procedure

(Cossenza and Paes De Carvalho, 2000). Retinas from White-Leghorn chick embryos

were used and monolayer retinal cultures enriched in glial cells prepared.

Neuroretinas from 8-day-old embryos were dissected from other structures of the

eye and immediately transferred to 1 mL of Ca2+ and Mg2+-free balanced salt

solution (CMF). Trypsin, at a final concentration of 0.1%, was added and the

suspension incubated at 37 8C for 20–25 min. Trypsin solution was removed and

the retinas resuspended in MEM containing 5% fetal calf serum, 2 mM glutamine,

100 U/mL penicillin and 100 mg/mL streptomycin. The tissues were mechanically

dissociated by successive aspirations of the medium. For quinacrine staining

experiments, the cells were seeded at a density of 2.5 � 103 cells/mm2 on 40 mm

plastic Petri dishes. For experiments measuring the extracellular levels of ATP, cells

were seeded on 4-well dishes, at the same density. Cultures were maintained in a

humidified atmosphere of 95% air/5% CO2, at 37 8C. Medium changes were

performed 3 times a week. Cultures were used after 14–20 days, when almost all

neurons died and the culture contained only glial cells.

s were incubated with 5 mM quinacrine for 5 min, washed and then observed under

located over the cytoplasm of Muller glial cells. (B) Photomicrograph of cells in (A)

d with 1 mM bafilomycin A1 for 1 h and labeled with 5 mM quinacrine for 5 min. (D)

E.C. Loiola, A.L.M. Ventura / Neurochemistry International 58 (2011) 414–422416

2.3. Quinacrine staining

Quinacrine staining of ATP-containing vesicles was performed as

described previously (Bodin and Burnstock, 2001a). Briefly, Muller glial cell

cultures were incubated with 5 mM quinacrine for 5 min, at 37 8C. The cultures

were washed 5� with Hank’s balanced salt solution (128 mM NaCl, 4 mM KCl,

1 mM Na2HPO4, 0.5 mM KH2PO4, 1 mM MgCl2, 3 mM CaCl2, 20 mM HEPES,

12 mM glucose, pH 7.4). The cells were immediately observed on a Nikon TE

2000-U fluorescence microscope using a B-2E/C filter block for FICT. Fluores-

cence of quinacrine was acquired by a digital camera immediately before

treatment (time = 0) or after cells were incubated with 50 mM KCl, 1 mM

glutamate or 100 mM kainate for 10 min, at room temperature. The glutamate

antagonists MK-801 and DNQX (50 mM) were always added 10 min prior to

glutamate or glutamatergic agonists addition. To examine the effect of 1 mM

bafilomycin A1 or 2 mM Evans blue, cells were treated for 1 h with the drug

prior to incubation with quinacrine. To examine the reversibility of Evans blue

blockade of quinacrine staining, stained cells treated with Evans blue were

washed once and incubated with 2 mL of complete MEM medium for 2 h, at

37 8C. After this incubation, cultures were stained again with quinacrine for

5 min, washed and observed under fluorescence illumination.

[()TD$FIG]

Fig. 2. Effect of Evans blue on quinacrine staining of living purified cultures of Muller glia.

incubated with 5 mM quinacrine for 5 min, washed and observed under fluorescence mic

as shown in (C) that was washed, incubated in complete culture medium for 2 h in the ab

under fluorescence microscopy. (B, D and F) Photomicrographs under phase contrast il

2.4. Measurement of extracellular ATP levels

Prior to measurement of the extracellular ATP levels, culture medium was

removed, cells washed twice with 0.5 mL of Hank’s balanced salt solution and

incubated for 5 min, at 37 8C, in 0.2 mL of Hank’s. This bathing solution was

discarded and cells incubated in fresh solution for another 5 min (basal level).

Medium was collected and cells incubated for an additional period of 5 min in

the presence of 50 mM KCl, 1 mM glutamate or 100 mM kainate (stimulated

level). The glutamate antagonists MK-801 and DNQX (50 mM) were added 5 min

before stimulation. BAPTA-AM (30 mM) and bafilomycin A (1 mM) were added

15 and 60 min prior stimulation, respectively. ATP release was measured by the

luciferin-luciferase assay using an ATP determination kit, following the

manufacturer’s instructions (Invitrogen). Briefly, ATP standards (25 nM–

400 nM) and test samples were added to eppendorf tubes containing the

luciferin–luciferase mixture. Tubes were immediately placed in a luminometer

(Turner BioSystems, Sunnyvale, CA) and luminescence measured for 10 s. A

calibration curve was constructed using ATP standards and used to calculate

ATP levels in test samples. Data in figures were expressed as normalized [ATP]

that represents the stimulated levels of extracellular ATP divided by the basal

levels of extracellular nucleotide.

Glia enriched cultures were pre-treated or not with 2 mM Evans blue for 60 min and

roscopy. (A) Control cultures. (C) Cultures treated with Evans blue. (E) Same culture

sence of Evans blue, labeled with 5 mM quinacrine for 5 min, washed and observed

lumination of cells shown in (A), (C) and (E), respectively. Bar = 10 mm.

E.C. Loiola, A.L.M. Ventura / Neurochemistry International 58 (2011) 414–422 417

2.5. Statistical analysis

Statistical comparisons were made by Student’s t test or one-way analysis of

variance (ANOVA) followed by the Bonferroni post-test. Data are presented as the

mean � S.E.M. for at least 3 or 4 experiments performed in duplicate or triplicate. A

P < 0.05 was taken as significant.

3. Results

Although strong previous evidences suggest that the pigmentedepithelium and retinal neurons are a main source of ATP in thedeveloping chick retina (Pearson et al., 2005; Santos et al., 1999),Muller glial cells were shown to release ATP during thepropagation of calcium waves induced by mechanical stimulationin the adult rat retina (Newman, 2001). In order to verify if Mullerglial cells from the developing chick retina could release ATP, we

[()TD$FIG]

Fig. 3. Effect of KCl depolarization on quinacrine staining and extracellular ATP levels in M

for 5 min, washed and photographed under fluorescence illumination (A and C). Then

photographed after 10 min (B and D). (A and B) control; (C and D), KCl. (E) Glial cultur

medium collected for determination of ATP content as described in Section 2. Data were

the mean value � S.E.M. of 3 independent experiments performed in duplicate. Control le

first investigated whether these cells presented ATP-filled vesiclesthat could be labeled by quinacrine as described in rat astrocytes(Coco et al., 2003). This acridine derivative is a weak-base thatbinds ATP with high affinity and is widely used to visualize ATP-containing sub-cellular compartments in living cells (Bodin andBurnstock, 2001b; Irvin and Irvin, 1954). Enriched Muller glia cellcultures were incubated with 5 mM quinacrine for 5 min, washedand immediately visualized under fluorescence illumination(Fig. 1A). An abundant punctate fluorescent staining, distributedover cell cytoplasm, was observed.

Neurotransmitter uptake into secretory vesicles requires anelectrochemical proton gradient that is maintained by a v-ATPase(Montana et al., 2006). In order to verify if fluorescent puncta weresecretory vesicles or other acidic organelles, enriched glial cultureswere incubated with the v-ATPase inhibitor bafilomycin A1 (1 mM)

uller glia cell cultures. Glia enriched cultures were incubated with 5 mM quinacrine

, solution was changed to a 50 mM KCl containing Hank’s solution and cultures

es were treated or not with 50 mM KCl for 5 min and samples of the extracellular

normalized to the mean control value of each experiment performed and represent

vels of ATP = 1.68 � 0.3 pmol/culture. **P < 0.01. Bar = 10 mm.

[()TD$FIG]

Fig. 4. Glutamate-induced decrease in quinacrine staining of Muller glial cells in culture. Quinacrine-loaded Muller glial cell cultures, pre-treated or not with 50 mM DNQX or

50 mM MK-801 for 10 min, were incubated in the absence or presence of 1 mM glutamate and photographed under fluorescence illumination immediately (A, C, E and G) or

after 10 min of incubation with drugs (B, D, F and H). (A) Control, 0 min. (B) Control, 10 min. (C) Glutamate, 0 min. (D) Glutamate, 10 min. (E) Glutamate + DNQX, 0 min. (F)

Glutamate + DNQX, 10 min. (G) Glutamate + MK-801, 0 min. (H) Glutamate + MK-801, 10 min. Experiments were performed 3 times with similar results.

E.C. Loiola, A.L.M. Ventura / Neurochemistry International 58 (2011) 414–422418

[()TD$FIG]

0.0

0.5

1.0

1.5

2.0

2.5

***

#

##

#

Glutamate - + - + - + - + DNQX - - + + - - + + MK-801 - - - - + + + +

No

rmal

ized

[A

TP

]

Fig. 5. Glutamate-induced increase in extracellular ATP levels in purified retinal

glial cultures. Purified glial cultures were incubated with or without 50 mM DNQX,

50 mM MK-801 or both antagonists for 5 min, at 37 8C, after which medium was

collected to determine the basal levels of extracellular ATP. Cells were further

incubated for 5 min at 37 8C with 1 mM glutamate, in the presence or not of

antagonists. Medium was then collected to measure glutamate-stimulated levels of

extracellular ATP. Data were normalized to the mean control value of each

experiment performed and represent the mean � S.E.M. of at least 5 independent

experiments performed in duplicate. Control levels of ATP = 1.81 � 0.15 pmol/culture.

***P < 0.001, versus control. #P < 0.05 and ##P < 0.01, versus glutamate.

E.C. Loiola, A.L.M. Ventura / Neurochemistry International 58 (2011) 414–422 419

for 1 h, prior to quinacrine staining. As shown in Fig. 1C, thisprocedure completely blocked the appearance of fluorescentgranules within cultured cells.

Recently, Sawada et al. (2008) identified a novel member of theSLC17 family of anion transporters (VNUT) that could activelyaccumulate nucleotides into liposomes. The uptake of ATP byVNUT was dependent on membrane potential and could be greatlyinhibited by DIDS and Evans blue, two potent blockers of theglutamate transporter VGLUT. Since quinacrine staining of Mullerglia in culture was blocked by the v-ATPase inhibitor bafilomycinA1, the effect of Evans blue on quinacrine staining of culturedMuller cells was investigated (Fig. 2). Enriched glial cultures wereincubated with 2 mM Evans blue for 1 h prior to quinacrinestaining. In contrast to control cultures where fluorescent granulescould be easily noticed (Fig. 2A), no quinacrine fluorescence wasdetected in cultures pre-treated with Evans blue (Fig. 2C).Moreover, quinacrine labeling over glial cells was restored whenquinacrine negative, Evans blue-treated cultures were washedbriefly and re-incubated in complete culture medium for 2 h, at37 8C. When these cultures were stained again with quinacrine, anabundant punctuate fluorescent labeling over the cytoplasm ofcells was observed (Fig. 2E).

Elevation of extracellular potassium depolarizes glial cellswhich can lead to an increase in the internal levels of Ca2+ in Mullercells (Keirstead and Miller, 1995; Wakakura and Yamamoto, 1994).Treatment of enriched Muller glia cultures with 50 mM KClresulted in reduced intracellular levels of quinacrine staining aftera 10 min period of stimulation (Fig. 3). No difference in quinacrinestaining could be detected in control cultures after the same periodof incubation, discarding the possibility that the decrease inquinacrine staining observed in KCl-treated cultures was due tofluorescence fading.

The effect of 50 mM KCl on the accumulation of extracellularATP was also investigated in these cultures (Fig. 3E). The amountof ATP in the bathing solution of the cultures was measured afteran incubation of 5 min in Hank’s balanced salt solution and afteran additional incubation of 5 min in Hank’s solution containing50 mM KCl. An increase of �77% over the basal levels of ATP wasobserved when cultures were incubated with 50 mM KCl. Whilenon-stimulated levels of ATP were 1.68 � 0.3 pmol/culture, levelsof ATP observed in KCl-treated cultures were 2.97 � 0.45 pmol/culture.

Both NMDA and AMPA/KA ionotropic glutamate receptors wereshown to be expressed in chick Muller glial cells (Lamas et al.,2005; Lopez et al., 1994, 1997). Activation of these sites was shownto elicit the release of ATP from astrocytes, although activation ofNMDA receptors elicits ATP release to a lesser extent (Pangrsicet al., 2007; Queiroz et al., 1997, 1999). To investigate if glutamatecould also induce the release of ATP from retinal Muller glia cells inculture, the effect of 1 mM glutamate on quinacrine staining ofglial cultures was evaluated (Fig. 4). Treatment of the cultures withglutamate for 10 min induced a decrease in both the intensity ofquinacrine fluorescence as in the number of quinacrine stainedgranules (Fig. 4D). No reduction in quinacrine staining wasobserved in control, non-treated cultures (Fig. 4B). Also, nodecrease in quinacrine staining was observed when glutamate-treated cultures were co-incubated with 50 mM of the antagonistsDNQX (Fig. 4F) or MK-801 (Fig. 4H).

The amount of extracellular ATP in glutamate-stimulated glialcultures was also evaluated using the luciferin–luciferase assay(Fig. 5). Consistent with the reduction in quinacrine staining,incubation with 1 mM glutamate, for 5 min, at 37 8C, induced asignificant increase in the extracellular levels of ATP in thecultures. While the medium of control cultures showed ATP levelsof 1.81 � 0.15 pmol/culture, glutamate-treated cultures showed ATPlevels of 3.43 � 0.33 pmol/culture.

The effect of glutamate selective agonists for NMDA and non-NMDA receptors on quinacrine staining of retina glial cells orextracellular levels of ATP is shown in Fig. 6. While no decrease inquinacrine staining was detected in control, non-treated cultures,incubation of the cultures with 100 mM concentration of kainatefor 10 min induced a consistent decrease in quinacrine fluorescentstaining in the cultured cells. Moreover, incubation of the cells with100 mM kainate for 5 min, at 37 8C, also induced a significantchange in extracellular ATP levels that increased from1.73 � 0.17 pmol/culture in control cultures to 3.14 � 0.55 pmol/culture in kainate-treated cultures. This increase in extracellular ATPlevels induced by kainate was completely prevented by theincubation of the cultures with the agonist in the presence of50 mM DNQX or 50 mM MK-801 or in the presence of bothantagonists.

Since MK-801, an NMDA receptor antagonist, blocked theincrease in extracellular ATP levels in both glutamate- and kainate-treated cultures, the effect of NMDA on ATP levels was alsoevaluated (Fig. 6F). Muller glia cultures were incubated for 5 min,at 37 8C, with 100 mM NMDA in Hank’s medium without MgCl2,but with 2 mM glycine. However, no increase in extracellular ATPlevels was observed in NMDA-treated cultures. No significantchange was also noticed in cultures treated with NMDA in thepresence of 50 mM of the antagonist MK-801.

Exocytosis is a regulated pathway of transmitter release thatdepends on intracellular calcium elevation. To investigate ifglutamate-induced increase in extracellular ATP level was depen-dent on intracellular calcium rise, glia-enriched cultures were pre-incubated with 30 mM of the Ca2+ chelator BAPTA-AM for 15 min,at 30 8C and incubated with 1 mM glutamate for an additional5 min period. As can be observed in Fig. 7, glutamate induced a�2� increase in extracellular nucleotide levels, an increase thatwas completely blocked by the addition of BAPTA-AM to theincubation medium. No significant difference in ATP levels wasobserved in BAPTA-AM-treated cultures, either in the presence orabsence of glutamate, as compared to the control cultures.

According to the evidences showing that bafilomycin A1impairs ATP storage in secretory organelles, a decrease inglutamate-induced rise in extracellular ATP levels was expectedto occur in bafilomycin A1-treated cultures. Muller glial cultureswere pre-incubated with 1 mM bafilomycin A1 for 1 h and thenincubated with 1 mM glutamate for 5 min. A significant reduction

[()TD$FIG]

Fig. 6. Effect of glutamate agonists on quinacrine staining and extracellular ATP levels in purified retinal glia cultures. Quinacrine-loaded cultured Muller glial cells were

incubated in the absence (A and B) or presence of 100 mM kainate (C and D) and photographed immediately (A and C) or after 10 min of incubation (B and D). Purified glial

cultures were incubated with or without 50 mM DNQX, 50 mM MK-801 or both antagonists for 5 min, at 37 8C. Then, medium was collected to determine the basal levels of

extracellular ATP. Cells were further incubated for 5 min at 37 8C with 100 mM kainate (E) or 100 mM NMDA (F), in the presence or not of the antagonists. Medium was then

collected to measure agonist-stimulated levels of extracellular ATP. Data were normalized to the mean control value of each experiment performed and represent the

mean � S.E.M. of 7 independent experiments performed in duplicate (E) or 3 experiments performed in quadruplicate (F). Control levels of ATP = 1.73 � 0.17 and 1.97 � 0.22 pmol/

culture, respectively. ***P < 0.001, versus control. #P < 0.05 and ##P < 0.01, versus kainate.

E.C. Loiola, A.L.M. Ventura / Neurochemistry International 58 (2011) 414–422420

in the glutamate-evoked increase in extracellular nucleotide levelswas observed in cultures treated with the v-ATPase inhibitor.Nucleotide levels decreased to only 60% and 92% of the controllevels in bafilomycin A1-treated and glutamate plus bafilomycin-treated cultures, respectively.

4. Discussion

Quinacrine is an acridine derivative that binds ATP with highaffinity and is widely used to visualize ATP-containing sub-cellularcompartments in living cells (Bodin and Burnstock, 2001b; Irvinand Irvin, 1954). In glial cells, quinacrine labeling of ATP-filledvesicles was first demonstrated in rat astrocytes (Coco et al., 2003).In the present study, we show that cultured chick Muller glia cellscould also be stained with quinacrine, with a pattern of stainingthat was granular and located in the cytoplasm of cells. Moreover,

quinacrine labeling of Muller cells could be prevented bybafilomycin A1, a potent inhibitor of vacuolar-ATPases and byEvans blue, a compound that blocks the uptake of ATP intoliposomes containing the newly identified vesicular nucleotidetransporter VNUT (Sawada et al., 2008). Together, these observa-tions strongly suggest that ATP is localized in secretory acidicvesicles in cultured Muller cells. Moreover, together with theobservation that Evans blue blockade of quinacrine staining wasreversible, our results also suggest that cultured avian Muller cellsstore ATP in acidic vesicles through the functioning of VNUT or arelated vesicular anion transporter sensitive to Evans blue. Oneinteresting point to be further explored is whether cultured Mullercells express this or some other similar transporter.

One major role of Muller glial cells is to regulate thecomposition of the retinal extracellular fluid. Neuronal activityresults in increases in extracellular K+ in the inner and outer

[()TD$FIG]

0.0

0.5

1.0

1.5

2.0

2.5

Glutamate 1 mM - + - + - + BAPTA-AM 30 µM - - + + - -Bafilomycin A1 1 µM - - - - + +

***

####

No

rmal

ized

[A

TP

]

Fig. 7. Effect of BAPTA-AM and bafilomycin A1 on glutamate-induced increase in

extracellular ATP levels. Cultures of Muller cells were incubated with 30 mM

BAPTA-AM or 1 mM bafilomycin A1 for 5 min, at 37 8C. Then, medium was collected

to determine the basal levels of extracellular ATP. Cells were further incubated for

5 min at 37 8C with 1 mM glutamate in the presence of inhibitors after which

medium was collected to estimate glutamate-induced increase in ATP levels.

BAPTA-AM and bafilomycin A1 were added to the cultures 15 min and 1 h before

the beginning of the experiment, respectively. Data were normalized to the mean

control value of each experiment and represents the mean � S.E.M. of at least 3

experiments performed in duplicate or quadruplicate. Control levels of

ATP = 1.76 � 0.15 pmol/culture. ***P < 0.001, versus control. ##P < 0.01, versus

glutamate.

E.C. Loiola, A.L.M. Ventura / Neurochemistry International 58 (2011) 414–422 421

plexiform layers and these variations lead to an influx of K+ intoMuller cells by a spatial-buffering mechanism, also known as ‘‘K+

siphoning’’, that depolarizes glial cells (Newman and Reichenbach,1996). Moreover, Muller cells express voltage-dependent calciumchannels (Newman, 1985) that were characterized as L-type ofcalcium channels in the human retina (Puro et al., 1996).Accordingly, high concentrations of extracellular K+ can inducean increase in intracellular calcium levels (Keirstead and Miller,1995; Wakakura and Yamamoto, 1994). In the present work, weshow that incubation of chick Muller glial cells with a 50 mMsolution of KCl induced both a decrease in quinacrine staining ofcell vesicles and a significant accumulation of ATP in the culturemedium, suggesting that under depolarization, cultured Mullerglia cells release ATP through the exocytosis of nucleotide-filledvesicles. Although ATP release from glial cells can occur by manydifferent pathways, such as connexin hemichannels (Stout et al.,2002), purinergic P2X7 receptor (Anderson et al., 2004) and ATPtransporter proteins (Abraham et al., 1993), the release of ATP byexocytosis was demonstrated in astrocytes (Bal-Price et al., 2002;Coco et al., 2003; Pangrsic et al., 2007) and Schwann cells (Liu et al.,2005).

Muller glial cells express several glutamate receptors, includingNMDA, AMPA/KA and metabotropic glutamate receptors (Keirst-ead and Miller, 1997; Lamas et al., 2005; Lopez et al., 1994, 1997;Lopez-Colome and Romo-de-Vivar, 1991; Uchihori and Puro, 1993;Wakakura and Yamamoto, 1994). As for KCl-mediated depolariza-tion, incubations with glutamate induced a decrease in quinacrinestaining as well as an increase in extracellular ATP content inretinal Muller cells in culture (Figs. 4 and 5). Both responses wereprevented by co-incubation of the cells with DNQX and/or MK-801,two selective antagonists for AMPA/KA and NMDA receptors,respectively, suggesting the participation of both ionotropicreceptors in glutamate-induced release of ATP in Muller glial cells.

Although NMDA and non-NMDA receptor antagonists blockedglutamate-induced increase in extracellular ATP, only kainate wascapable of inducing nucleotide accumulation in medium. Noincrease was observed by incubating cells with NMDA. Bothantagonists also blocked the increase in extracellular ATP levelsinduced by kainate. At least two possibilities could account for thisobservation. The first would be that NMDA receptor antagonistMK-801 blocked non-NMDA receptor stimulation. This possibilityhowever, does not seem plausible since no evidences for such non-

specific effect of MK-801 were found so far. Another possibilitywould be that kainate induced the release of endogenousglutamate as already suggested by Uckermann et al. (2006) inthe rat retina. In this scenario, released glutamate would stimulateNMDA receptors that together with the activation of non-NMDAreceptors by glutamate or kainate would induce the release of ATPfrom cultured Muller cells. This possibility is particularlyinteresting since a kainate-induced, calcium-dependent releaseof [3H]-D-aspartate was previously demonstrated in mixed chickretinal cultures (Duarte et al., 1996) as well as in the retina of otherspecies (Ohia et al., 2000). Since Muller cells seems to take up andrelease glutamate (Gadea et al., 2004; Newman and Zahs, 1998;Reis et al., 2008), one interesting point that deserves furtherinvestigation is whether glutamate itself can induce the release ofD-aspartate or glutamate from cultured chick Muller cells.

Previous evidences have shown that glutamate does not inducecalcium mobilization in Muller cells from adult rodent retinas(Newman, 2005; Newman and Zahs, 1997; Rillich et al., 2009;Uckermann et al., 2004). Moreover, in same preparations, the releaseof ATP from Muller cells was shown to be a calcium-independent,non-exocytotic process (Uckermann et al., 2006; Wurm et al., 2008).In the present study, glutamate-induced accumulation of extracel-lular ATP was blocked by BAPTA-AM, a chelator of intracellularcalcium and by bafilomycin A1, a v-ATPase inhibitor. Thediscrepancies between our findings and those mentioned abovemay have several explanations, including species or age differences.An interesting hypothesis is that the glutamate-induced calcium-dependent exocytotic release of ATP observed in the present studyoccurred only in cultured Muller cells, but not in freshly dissociatedor non-dissociated Muller cells as those used in the mentionedstudies. It is known that Muller cells in purified cultures candedifferentiate to progenitors and express different sets of signalingcomponents (Reis et al., 2008; Bringmann et al., 2009). In goodagreement with this hypothesis are the previous observations thatglutamate agonists do evoke intracellular calcium rise when Mullercells obtained from adult human, rabbit or salamander retinas arecultured for several days (Keirstead and Miller, 1997; Uchihori andPuro, 1993; Wakakura and Yamamoto, 1994).

Acknowledgements

We would like to thank Maria Leite Eduardo for technicalassistance. This work was supported by grants from FAPERJ, CAPES,MCT-PRONEX, CNPQ and PROPPI-UFF.

References

Abraham, E.H., et al., 1993. The multidrug resistance (mdr1) gene product functionsas an ATP channel. Proc. Natl. Acad. Sci. U.S.A. 90, 312–316.

Anderson, C.M., et al., 2004. ATP-induced ATP release from astrocytes. J. Neuro-chem. 88, 246–256.

Bal-Price, A., et al., 2002. Nitric oxide induces rapid, calcium-dependent release ofvesicular glutamate and ATP from cultured rat astrocytes. Glia 40, 312–323.

Battista, A.G., et al., 2009. Extracellular ADP regulates lesion-induced in vivo cellproliferation and death in the zebrafish retina. J. Neurochem. 111, 600–613.

Bodin, P., Burnstock, G., 2001a. Evidence that release of adenosine triphosphatefrom endotelial cells during increased shear stress is vesicular. J. Cardiovasc.Pharmacol. 38, 900–908.

Bodin, P., Burnstock, G., 2001b. Purinergic signalling: ATP release. Neurochem. Res.26, 959–969.

Bringmann, A., et al., 2006. Muller cells in the healthy and diseased retina. Prog.Retin. Eye Res. 25, 397–424.

Bringmann, A., et al., 2009. Cellular signaling and factors involved in Muller cellgliosis: neuroprotective and detrimental effects. Prog. Retin. Eye Res. 28, 423–451.

Coco, S., et al., 2003. Storage and release of ATP from astrocytes in culture. J. Biol.Chem. 278, 1354–1362.

Cossenza, M., Paes De Carvalho, R., 2000. L-Arginine uptake and release by culturedavian retinal cells: differential cellular localization in relation to nitric oxidesynthase. J. Neurochem. 74, 1885–1894.

Costa, G., et al., 2009. High glucose changes extracellular adenosine triphosphatelevels in rat retinal cultures. J. Neurosci. Res. 87, 1375–1380.

E.C. Loiola, A.L.M. Ventura / Neurochemistry International 58 (2011) 414–422422

Duarte, C.B., et al., 1996. Glutamate release evoked by glutamate receptor agonistsin cultured chick retina cells: modulation by arachidonic acid. J. Neurosci. Res.44, 363–373.

Franca, G.R., et al., 2007. ATP-induced proliferation of developing retinal cells:regulation by factors released from postmitotic cells in culture. Int. J. Dev.Neurosci. 25, 283–291.

Franke, H., Illes, P., 2006. Involvement of P2 receptors in the growth and survival ofneurons in the CNS. Pharmacol. Ther. 109, 297–324.

Fries, J.E., et al., 2004b. Expression of P2Y1, P2Y2, P2Y4, and P2Y6 receptor subtypesin the rat retina. Invest. Ophthalmol. Vis. Sci. 45, 3410–3417.

Fries, J.E., et al., 2004a. Distribution of metabotropic P2Y receptors in the rat retina:a single-cell RT-PCR study. Mol. Brain Res. 130, 1–6.

Gadea, A., et al., 2004. Glutamate-induced inhibition of D-aspartate uptake inMuller glia from the retina. Neurochem. Res. 29, 295–304.

Greenwood, D., et al., 1997. Expression of the P2X2 receptor subunit of the ATP-gated ion channel in the retina. Neuroreport 8, 1083–1088.

Housley, G.D., et al., 2009. Purinergic signaling in special senses. Trends Neurosci.32, 128–141.

Irvin, J.L., Irvin, E.M., 1954. The interaction of quinacrine with adenine nucleotides. J.Biol. Chem. 210, 45–56.

Jabs, R., et al., 2000. Evidence for P2X3, P2X4, P2X5 but not for P2X7 containingpurinergic receptors in Muller cells of the rat retina. Mol. Brain Res. 76, 205–210.

Keirstead, S.A., Miller, R.F., 1995. Calcium waves in dissociated retinal glial (Muller)cells are evoked by release of calcium from intracellular stores. Glia 14, 14–22.

Keirstead, S.A., Miller, R.F., 1997. Metabotropic glutamate receptor agonists evokecalcium waves in isolated Muller cells. Glia 21, 194–203.

Lamas, M., et al., 2005. Cell-specific expression of N-methyl-D-aspartate receptorsubunits in Muller glia and neurons from the chick retina. Invest. Ophthalmol.Vis. Sci. 46, 3570–3577.

Li, Y., et al., 2001. Muller cell Ca2+ waves evoked by purinergic receptor agonists inslices of rat retina. J. Neurophysiol. 85, 986–994.

Liu, G.J., et al., 2005. Secretion of ATP from Schwann cells in response to uridinetriphosphate. Eur. J. Neurosci. 21, 151–160.

Lopez, T., et al., 1994. AMPA/KA receptor expression in radial glia. Neuroreport 5,504–506.

Lopez, T., et al., 1997. NMDA receptors in cultured radial glia. FEBS Lett. 405, 245–248.

Lopez-Colome, A.M., Romo-de-Vivar, M., 1991. Excitatory amino acid receptors inprimary cultures of glial cells from the retina. Glia 4, 431–439.

Milenkovic, I., et al., 2003. P2Y receptor-mediated stimulation of Muller glial cellDNA synthesis: dependence on EGF and PDGF receptor transactivation. Invest.Ophthalmol. Vis. Sci. 44, 1211–1220.

Moll, V., et al., 2002. P2Y receptor-mediated stimulation of Muller glial DNAsynthesis. Invest. Ophthalmol. Vis. Sci. 43, 766–773.

Montana, V., et al., 2006. Vesicular transmitter release from astrocytes. Glia 54,700–715.

Newman, E.A., 1985. Voltage-dependent calcium and potassium channels in retinalglial cells. Nature 17, 809–811.

Newman, E.A., 2001. Propagation of intracellular calcium waves in retinal astro-cytes and Muller cells. J. Neurosci. 2, 2215–2223.

Newman, E.A., 2003. Glial cell inhibition of neurons by release of ATP. Neuroscience23, 1659–1666.

Newman, E.A., 2004. Glial modulation of synaptic transmission in the retina. Glia47, 268–274.

Newman, E.A., 2005. Calcium increases in retinal glial cells evoked by light-inducedneuronal activity. J. Neurosci. 25, 5502–5510.

Newman, E.A., Zahs, K.R., 1997. Calcium waves in retinal glial cells. Science 275,844–847.

Newman, E.A., Zahs, K.R., 1998. Modulation of neuronal activity by glial cells in theretina. J. Neurosci. 18, 4022–4028.

Newman, E.A., Reichenbach, A., 1996. The Muller cell: a functional element of theretina. Trends Neurosci. 19, 307–312.

Nunes, P.H., et al., 2007. Signal transduction pathways associated with ATP-inducedproliferation of cell progenitors in the intact embryonic retina. Int. J. Dev.Neurosci. 25, 499–508.

Ohia, S.E., et al., 2000. Human, bovine, and rabbit retinal glutamate-induced [3H]D-aspartate release: role in excitotoxicity. Neurochem. Res. 25, 853–860.

Ornelas, I.M., Ventura, A.L.M., 2010. Involvement of the PI3K/AKT pathway in ATP-induced proliferation of developing retinal cells in culture. Int. J. Dev. Neurosci.28, 503–511.

Pangrsic, T., et al., 2007. Exocytotic release of ATP from cultured astrocytes. J. Biol.Chem. 282, 28749–28758.

Pearson, R.A., et al., 2002. Purinergic and muscarinic modulation of the cell cycleand calcium signaling in the chick retinal ventricular zone. J. Neurosci. 22,7569–7579.

Pearson, R.A., et al., 2005. ATP released via gap junction hemichannels from thepigment epithelium regulates neural retinal progenitor proliferation. Neuron46, 731–744.

Perez, M.T., et al., 1986. Release of endogenous and radioactive purines from therabbit retina. Brain Res. 398, 106–112.

Puro, D.G., et al., 1996. Characterization of an L-type calcium channel expressed byhuman retinal Muller (glial) cells. Brain Res. Mol. Brain Res. 37, 41–48.

Queiroz, G., et al., 1997. Release of ATP from cultured rat astrocytes elicited byglutamate receptor activation. Neuroscience 78, 1203–1208.

Queiroz, G., et al., 1999. A study of the mechanism of the release of ATP from ratcortical astroglial cells evoked by activation of glutamate receptors. Neurosci-ence 91, 1171–1181.

Reigada, D., et al., 2006. Glutamate acts at NMDA receptors on fresh bovine and oncultured human retinal pigment epithelial cells to trigger release of ATP. J.Physiol. 575, 707–720.

Reis, R.A.M., et al., 2008. Muller glia as an active compartment modulating nervousactivity in the vertebrate retina: neurotransmitters and trophic factors. Neu-rochem. Res. 33, 1466–1474.

Resta, V., et al., 2005. Neuronal death induced by endogenous extracellular ATP inretinal cholinergic neuron density control. Development 132, 2873–2882.

Resta, V., et al., 2007. Acute retinal ganglion cell injury caused by intraocularpressure spikes is mediated by endogenous extracellular ATP. Eur. J. Neurosci.25, 2741–2754.

Rillich, K., et al., 2009. Light stimulation evokes two different calcium responses inMuller glial cells of the guinea pig retina. Eur. J. Neurosci. 29, 1165–1176.

Sanches, G., et al., 2002. ATP induces proliferation of retinal cells in culture viaactivation of PKC and extracellular signal-regulated kinase cascade. Int. J. Dev.Neurosci. 20, 21–27.

Santos, P.F., et al., 1999. Characterization of ATP release from cultures enriched incholinergic amacrine-like neurons. J. Neurobiol. 41, 340–348.

Sarthy, V., Ripps, H., 2001. The Retinal Muller Cell. Kluwer Academic/PlenumPublishers, NewYork.

Sawada, K., et al., 2008. Identification of a vesicular nucleotide transporter. Proc.Natl. Acad. Sci. U.S.A. 105, 5683–5686.

Sholl-Franco, A., et al., 2010. ATP controls cell cycle and induces proliferation in themouse developing retina. Int. J. Dev. Neurosci. 28, 63–73.

Stout, C.E., et al., 2002. Intercellular Calcium signaling in astrocytes via ATP releasethrough connexin hemichannels. J. Biol. Chem. 277, 10482–10488.

Sugioka, M., et al., 1999. Involvement of P2 purinoceptors in the regulation of DNAsynthesis in the neural retina of chick embryo. Int. J. Dev. Neurosci. 17, 135–144.

Uchihori, Y., Puro, D.G., 1993. Glutamate as a neuron-to-glial signal for mitogenesis:role of glial N-methyl-D-aspartate receptors. Brain Res. 613, 212–220.

Uckermann, O., et al., 2004. Glutamate-evoked alterations of glial and neuronal cellmorphology in the guinea pig retina. J. Neurosci. 24, 10149–10158.

Uckermann, O., et al., 2006. Glutamate release by neurons evokes a purinergicinhibitory mechanism of osmotic glial cell swelling in the rat retina: activationby neuropeptide Y. J. Neurosci. Res. 83, 538–550.

Verkhratsky, A., et al., 2009. Purinoceptors on neuroglia. Mol. Neurobiol. 39, 190–208.

Wakakura, M., Yamamoto, N., 1994. Cytosolic calcium transient increase throughthe AMPA/kainate receptor in cultured Muller cells. Vision Res. 34, 1105–1109.

Wheeler-Schilling, T.H., et al., 2000. Expression of purinergic receptors in bipolarcells of the rat retina. Mol. Brain Res. 76, 415–418.

Wheeler-Schilling, T.H., et al., 2001. Identification of purinergic receptors in retinalganglion cells. Mol. Brain Res. 92, 177–180.

Wurm, A., et al., 2008. Purinergic receptor activation inhibits osmotic glial cellswelling in the diabetic rat retina. Exp. Eye Res. 87, 385–393.